Projects - User contributions [en]

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-10-11T05:41:20Z

A1621205: /* Booth List (for internal use only) */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== ''RadBeacon Tag'' ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter, it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting<ref name=ref1/>.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods.<ref name=ref1/> Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime.<ref name=ref1/> In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter takes a large number of consecutive measurements to create the optional estimation of the state.<ref name=ref24/> It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node.<ref name=ref1/> The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance.<ref name=ref21/> It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications.<ref name=ref25/> In particular PiBeacon is able to work with iBeacon technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>

<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>

<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>

<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>

<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (377,150)
*<project_04> (485,150)
*<project_05> (595,150)
*<project_06> (700,150)
*<project_07> (377,180)
*<project_08> (485,180)
*<project_10> (595,180)
*<project_11> (700,180)
*<project_12> (128,315)
*<project_13> (225,315)
*<project_15> (328,315)
*<project_16> (438,315)
*<project_17> (128,344)
*<project_18> (225,344)
*<project_21> (328,344)
*<project_25> (438,344)
*<project_26> (642,315)
*<project_28> (745,315)
*<project_31> (845,315)
*<project_32> (937,315)
*<project_36> (642,344)
*<project_40> (745,344)
*<project_42> (845,344)
*<project_46> (937,344)
*<project_45> (377,482)
*<project_50> (485,482)
*<project_56> (595,482)
*<project_61> (700,482)
*<project_70> (377,517)
*<project_72> (485,517)
*<project_73> (595,517)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-10-11T05:06:14Z

A1621205: /* Booth List (for internal use only) */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== ''RadBeacon Tag'' ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter, it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting<ref name=ref1/>.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods.<ref name=ref1/> Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime.<ref name=ref1/> In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter takes a large number of consecutive measurements to create the optional estimation of the state.<ref name=ref24/> It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node.<ref name=ref1/> The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance.<ref name=ref21/> It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications.<ref name=ref25/> In particular PiBeacon is able to work with iBeacon technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>

<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>

<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>

<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>

<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (520,208)
*<project_04> (670,208)
*<project_05> (820,208)
*<project_06> (974,208)
*<project_07> (520,254)
*<project_08> (670,254)
*<project_10> (820,254)
*<project_11> (974,254)
*<project_12> (178,430)
*<project_13> (320,430)
*<project_15> (450,430)
*<project_16> (600,430)
*<project_17> (178,480)
*<project_18> (320,480)
*<project_21> (450,480)
*<project_25> (600,480)
*<project_26> (896,430)
*<project_28> (1028,430)
*<project_31> (1176,430)
*<project_32> (1304,430)
*<project_36> (896,480)
*<project_40> (1028,480)
*<project_42> (1176,480)
*<project_46> (1304,480)
*<project_45> (520,670)
*<project_50> (670,670)
*<project_56> (820,670)
*<project_61> (974,670)
*<project_70> (520,720)
*<project_72> (670,720)
*<project_73> (820,720)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-09-25T04:00:26Z

A1621205: /* Booth List (for internal use only) */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== ''RadBeacon Tag'' ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter, it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting<ref name=ref1/>.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods.<ref name=ref1/> Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime.<ref name=ref1/> In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter takes a large number of consecutive measurements to create the optional estimation of the state.<ref name=ref24/> It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node.<ref name=ref1/> The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance.<ref name=ref21/> It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications.<ref name=ref25/> In particular PiBeacon is able to work with iBeacon technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>

<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>

<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>

<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>

<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (40,15)
*<project_04> (50,15)
*<project_05> (60,15)
*<project_06> (70,15)
*<project_07> (40,25)
*<project_08> (50,25)
*<project_10> (60,25)
*<project_11> (70,25)
*<project_12> (15,35)
*<project_13> (25,35)
*<project_15> (35,35)
*<project_16> (45,35)
*<project_17> (15,45)
*<project_18> (25,45)
*<project_21> (35,45)
*<project_25> (45,45)
*<project_26> (65,35)
*<project_28> (75,35)
*<project_31> (85,35)
*<project_32> (95,35)
*<project_36> (65,45)
*<project_40> (75,45)
*<project_42> (85,45)
*<project_46> (95,45)
*<project_45> (40,55)
*<project_50> (50,55)
*<project_56> (60,55)
*<project_61> (70,55)
*<project_70> (40,65)
*<project_72> (50,65)
*<project_73> (60,65)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-09-25T01:13:29Z

A1621205: /* Booth List (for internal use only) */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== ''RadBeacon Tag'' ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter, it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting<ref name=ref1/>.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods.<ref name=ref1/> Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime.<ref name=ref1/> In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter takes a large number of consecutive measurements to create the optional estimation of the state.<ref name=ref24/> It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node.<ref name=ref1/> The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance.<ref name=ref21/> It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications.<ref name=ref25/> In particular PiBeacon is able to work with iBeacon technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>

<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>

<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>

<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>

<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (40,15,0)
*<project_04> (50,15,0)
*<project_05> (60,15,0)
*<project_06> (70,15,0)
*<project_07> (40,25,0)
*<project_08> (50,25,0)
*<project_10> (60,25,0)
*<project_11> (70,25,0)
*<project_12> (15,35,0)
*<project_13> (25,35,0)
*<project_15> (35,35,0)
*<project_16> (45,35,0)
*<project_17> (15,45,0)
*<project_18> (25,45,0)
*<project_21> (35,45,0)
*<project_25> (45,45,0)
*<project_26> (65,35,0)
*<project_28> (75,35,0)
*<project_31> (85,35,0)
*<project_32> (95,35,0)
*<project_36> (65,45,0)
*<project_40> (75,45,0)
*<project_42> (85,45,0)
*<project_46> (95,45,0)
*<project_45> (40,55,0)
*<project_50> (50,55,0)
*<project_56> (60,55,0)
*<project_61> (70,55,0)
*<project_70> (40,65,0)
*<project_72> (50,65,0)
*<project_73> (60,65,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-09-10T14:53:06Z

A1621205: /* Booth List (for internal use only) */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== ''RadBeacon Tag'' ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter, it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting<ref name=ref1/>.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods.<ref name=ref1/> Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime.<ref name=ref1/> In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter takes a large number of consecutive measurements to create the optional estimation of the state.<ref name=ref24/> It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node.<ref name=ref1/> The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance.<ref name=ref21/> It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications.<ref name=ref25/> In particular PiBeacon is able to work with iBeacon technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>

<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>

<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>

<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>

<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_72> (31,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-27T10:42:13Z

A1621205: /* References */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== ''RadBeacon Tag'' ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter, it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting<ref name=ref1/>.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods.<ref name=ref1/> Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime.<ref name=ref1/> In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter takes a large number of consecutive measurements to create the optional estimation of the state.<ref name=ref24/> It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node.<ref name=ref1/> The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance.<ref name=ref21/> It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications.<ref name=ref25/> In particular PiBeacon is able to work with iBeacon technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>

<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>

<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>

<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>

<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-27T10:41:32Z

A1621205: /* References */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== ''RadBeacon Tag'' ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter, it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting<ref name=ref1/>.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods.<ref name=ref1/> Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime.<ref name=ref1/> In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter takes a large number of consecutive measurements to create the optional estimation of the state.<ref name=ref24/> It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node.<ref name=ref1/> The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance.<ref name=ref21/> It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications.<ref name=ref25/> In particular PiBeacon is able to work with iBeacon technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>

<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref14>IDC: Smartphone OS Market Share. (2015). [Online]. Available: http://www.idc.com/prodserv/smartphone-os-market-share.jsp. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>

<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref23>D. Stojanović, N. Stojanović, "INDOOR LOCALIZATION AND TRACKING: METHODS, TECHNOLOGIES AND RESEARCH CHALLENGES", Faculty of Electronic Engineering, Dept. Computer Science, Republic of Serbia, Vol. 13, No 1, 2014, pp. 57 - 72, 2014</ref>
<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-27T10:33:41Z

A1621205: /* References */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== ''RadBeacon Tag'' ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter, it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting<ref name=ref1/>.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods.<ref name=ref1/> Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime.<ref name=ref1/> In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter takes a large number of consecutive measurements to create the optional estimation of the state.<ref name=ref24/> It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node.<ref name=ref1/> The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance.<ref name=ref21/> It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications.<ref name=ref25/> In particular PiBeacon is able to work with iBeacon technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>

<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref14>IDC: Smartphone OS Market Share. (2015). [Online]. Available: http://www.idc.com/prodserv/smartphone-os-market-share.jsp. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>
<ref name=ref16>iBeacon for Developers - Apple Developer. (2015). [Online]. Available:https://developer.apple.com/ibeacon/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref17>Radius Networks. (2015). [Online]. Available: http://developer.radiusnetworks.com/pibeacon/pibeacon-instructions.html. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref18>RadBeacon Config App. (2015). [Online]. Available: http://store.radiusnetworks.com/products/radbeacon-config. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref19>Estimote Community Portal. What are Broadcasting Power, RSSI and Measured Power? (2015). [Online]. Available: https://community.estimote.com/hc/en-us/articles/201636913-What-are-Broadcasting-Power-RSSI-and-Measured-Power-. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref20>E. Lau, W. Chung, "Enhanced RSSI-based Real-time User Location Tracking System for Indoor and Outdoor Environments," IEE, 2007</ref>
<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref23>D. Stojanović, N. Stojanović, "INDOOR LOCALIZATION AND TRACKING: METHODS, TECHNOLOGIES AND RESEARCH CHALLENGES", Faculty of Electronic Engineering, Dept. Computer Science, Republic of Serbia, Vol. 13, No 1, 2014, pp. 57 - 72, 2014</ref>
<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-27T10:31:48Z

A1621205: /* Project Information */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== ''RadBeacon Tag'' ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter, it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting<ref name=ref1/>.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods.<ref name=ref1/> Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime.<ref name=ref1/> In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter takes a large number of consecutive measurements to create the optional estimation of the state.<ref name=ref24/> It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node.<ref name=ref1/> The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance.<ref name=ref21/> It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications.<ref name=ref25/> In particular PiBeacon is able to work with iBeacon technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref3>The University of Adelaide. (2015). About Ingenuity 2015. [Online]. Available:http://www.ecms.adelaide.edu.au/ingenuity/about/. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref4>The University of Adelaide. (2015). Projects. [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Main_page#Final_Year_Projects. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref5>The University of Adelaide. (2015). [Online]. Available: https://myuni.adelaide.edu.au/webapps/blackboard/content/listContent.jsp?course_id=_318829_1&content_id=_6466732_1. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref6>Projects:2014S1-51 Heart Signal Processing Software for Evaluating Pacemaker Effectiveness - Projects. (2015). [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Projects:2014S151_Heart_Signal_Processing_Software_for_Evaluating_Pacemaker_Effectiveness. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref14>IDC: Smartphone OS Market Share. (2015). [Online]. Available: http://www.idc.com/prodserv/smartphone-os-market-share.jsp. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>
<ref name=ref16>iBeacon for Developers - Apple Developer. (2015). [Online]. Available:https://developer.apple.com/ibeacon/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref17>Radius Networks. (2015). [Online]. Available: http://developer.radiusnetworks.com/pibeacon/pibeacon-instructions.html. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref18>RadBeacon Config App. (2015). [Online]. Available: http://store.radiusnetworks.com/products/radbeacon-config. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref19>Estimote Community Portal. What are Broadcasting Power, RSSI and Measured Power? (2015). [Online]. Available: https://community.estimote.com/hc/en-us/articles/201636913-What-are-Broadcasting-Power-RSSI-and-Measured-Power-. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref20>E. Lau, W. Chung, "Enhanced RSSI-based Real-time User Location Tracking System for Indoor and Outdoor Environments," IEE, 2007</ref>
<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref23>D. Stojanović, N. Stojanović, "INDOOR LOCALIZATION AND TRACKING: METHODS, TECHNOLOGIES AND RESEARCH CHALLENGES", Faculty of Electronic Engineering, Dept. Computer Science, Republic of Serbia, Vol. 13, No 1, 2014, pp. 57 - 72, 2014</ref>
<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T07:02:22Z

A1621205: /* RadBeacon Tag */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) (see Appendix C) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) (see Appendix C) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon (see section 2.1.3.1.1) provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== ''RadBeacon Tag'' ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned in section 2.1.3.1, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter [1], it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods [1]. Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime [1]. In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter [24] takes a large number of consecutive measurements to create the optional estimation of the state. It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node [1]. The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance [21]. It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon [25] is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications. In particular PiBeacon is able to work with iBeacon (section 2.1.3.1.1) technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag (see section 2.1.3.1.1.1) devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 (see Appendix A) a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon (see section 2.3) was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed in section 2.2. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided in section 3.1.3 into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined in Section 3.1 and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref3>The University of Adelaide. (2015). About Ingenuity 2015. [Online]. Available:http://www.ecms.adelaide.edu.au/ingenuity/about/. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref4>The University of Adelaide. (2015). Projects. [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Main_page#Final_Year_Projects. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref5>The University of Adelaide. (2015). [Online]. Available: https://myuni.adelaide.edu.au/webapps/blackboard/content/listContent.jsp?course_id=_318829_1&content_id=_6466732_1. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref6>Projects:2014S1-51 Heart Signal Processing Software for Evaluating Pacemaker Effectiveness - Projects. (2015). [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Projects:2014S151_Heart_Signal_Processing_Software_for_Evaluating_Pacemaker_Effectiveness. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref14>IDC: Smartphone OS Market Share. (2015). [Online]. Available: http://www.idc.com/prodserv/smartphone-os-market-share.jsp. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>
<ref name=ref16>iBeacon for Developers - Apple Developer. (2015). [Online]. Available:https://developer.apple.com/ibeacon/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref17>Radius Networks. (2015). [Online]. Available: http://developer.radiusnetworks.com/pibeacon/pibeacon-instructions.html. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref18>RadBeacon Config App. (2015). [Online]. Available: http://store.radiusnetworks.com/products/radbeacon-config. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref19>Estimote Community Portal. What are Broadcasting Power, RSSI and Measured Power? (2015). [Online]. Available: https://community.estimote.com/hc/en-us/articles/201636913-What-are-Broadcasting-Power-RSSI-and-Measured-Power-. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref20>E. Lau, W. Chung, "Enhanced RSSI-based Real-time User Location Tracking System for Indoor and Outdoor Environments," IEE, 2007</ref>
<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref23>D. Stojanović, N. Stojanović, "INDOOR LOCALIZATION AND TRACKING: METHODS, TECHNOLOGIES AND RESEARCH CHALLENGES", Faculty of Electronic Engineering, Dept. Computer Science, Republic of Serbia, Vol. 13, No 1, 2014, pp. 57 - 72, 2014</ref>
<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T07:02:06Z

A1621205: /* iBeacon */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) (see Appendix C) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) (see Appendix C) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon (see section 2.1.3.1.1) provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== ''iBeacon'' ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.

====== RadBeacon Tag ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned in section 2.1.3.1, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter [1], it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods [1]. Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime [1]. In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter [24] takes a large number of consecutive measurements to create the optional estimation of the state. It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node [1]. The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance [21]. It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon [25] is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications. In particular PiBeacon is able to work with iBeacon (section 2.1.3.1.1) technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag (see section 2.1.3.1.1.1) devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 (see Appendix A) a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon (see section 2.3) was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed in section 2.2. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided in section 3.1.3 into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined in Section 3.1 and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref3>The University of Adelaide. (2015). About Ingenuity 2015. [Online]. Available:http://www.ecms.adelaide.edu.au/ingenuity/about/. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref4>The University of Adelaide. (2015). Projects. [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Main_page#Final_Year_Projects. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref5>The University of Adelaide. (2015). [Online]. Available: https://myuni.adelaide.edu.au/webapps/blackboard/content/listContent.jsp?course_id=_318829_1&content_id=_6466732_1. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref6>Projects:2014S1-51 Heart Signal Processing Software for Evaluating Pacemaker Effectiveness - Projects. (2015). [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Projects:2014S151_Heart_Signal_Processing_Software_for_Evaluating_Pacemaker_Effectiveness. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref14>IDC: Smartphone OS Market Share. (2015). [Online]. Available: http://www.idc.com/prodserv/smartphone-os-market-share.jsp. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>
<ref name=ref16>iBeacon for Developers - Apple Developer. (2015). [Online]. Available:https://developer.apple.com/ibeacon/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref17>Radius Networks. (2015). [Online]. Available: http://developer.radiusnetworks.com/pibeacon/pibeacon-instructions.html. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref18>RadBeacon Config App. (2015). [Online]. Available: http://store.radiusnetworks.com/products/radbeacon-config. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref19>Estimote Community Portal. What are Broadcasting Power, RSSI and Measured Power? (2015). [Online]. Available: https://community.estimote.com/hc/en-us/articles/201636913-What-are-Broadcasting-Power-RSSI-and-Measured-Power-. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref20>E. Lau, W. Chung, "Enhanced RSSI-based Real-time User Location Tracking System for Indoor and Outdoor Environments," IEE, 2007</ref>
<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref23>D. Stojanović, N. Stojanović, "INDOOR LOCALIZATION AND TRACKING: METHODS, TECHNOLOGIES AND RESEARCH CHALLENGES", Faculty of Electronic Engineering, Dept. Computer Science, Republic of Serbia, Vol. 13, No 1, 2014, pp. 57 - 72, 2014</ref>
<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T07:01:42Z

A1621205: /* Bluetooth Low Energy */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) (see Appendix C) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) (see Appendix C) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== ''Bluetooth Low Energy'' ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon (see section 2.1.3.1.1) provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.

====== iBeacon ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.
====== RadBeacon Tag ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned in section 2.1.3.1, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter [1], it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods [1]. Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime [1]. In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter [24] takes a large number of consecutive measurements to create the optional estimation of the state. It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node [1]. The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance [21]. It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon [25] is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications. In particular PiBeacon is able to work with iBeacon (section 2.1.3.1.1) technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag (see section 2.1.3.1.1.1) devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 (see Appendix A) a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon (see section 2.3) was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed in section 2.2. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided in section 3.1.3 into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined in Section 3.1 and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref3>The University of Adelaide. (2015). About Ingenuity 2015. [Online]. Available:http://www.ecms.adelaide.edu.au/ingenuity/about/. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref4>The University of Adelaide. (2015). Projects. [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Main_page#Final_Year_Projects. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref5>The University of Adelaide. (2015). [Online]. Available: https://myuni.adelaide.edu.au/webapps/blackboard/content/listContent.jsp?course_id=_318829_1&content_id=_6466732_1. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref6>Projects:2014S1-51 Heart Signal Processing Software for Evaluating Pacemaker Effectiveness - Projects. (2015). [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Projects:2014S151_Heart_Signal_Processing_Software_for_Evaluating_Pacemaker_Effectiveness. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref14>IDC: Smartphone OS Market Share. (2015). [Online]. Available: http://www.idc.com/prodserv/smartphone-os-market-share.jsp. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>
<ref name=ref16>iBeacon for Developers - Apple Developer. (2015). [Online]. Available:https://developer.apple.com/ibeacon/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref17>Radius Networks. (2015). [Online]. Available: http://developer.radiusnetworks.com/pibeacon/pibeacon-instructions.html. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref18>RadBeacon Config App. (2015). [Online]. Available: http://store.radiusnetworks.com/products/radbeacon-config. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref19>Estimote Community Portal. What are Broadcasting Power, RSSI and Measured Power? (2015). [Online]. Available: https://community.estimote.com/hc/en-us/articles/201636913-What-are-Broadcasting-Power-RSSI-and-Measured-Power-. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref20>E. Lau, W. Chung, "Enhanced RSSI-based Real-time User Location Tracking System for Indoor and Outdoor Environments," IEE, 2007</ref>
<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref23>D. Stojanović, N. Stojanović, "INDOOR LOCALIZATION AND TRACKING: METHODS, TECHNOLOGIES AND RESEARCH CHALLENGES", Faculty of Electronic Engineering, Dept. Computer Science, Republic of Serbia, Vol. 13, No 1, 2014, pp. 57 - 72, 2014</ref>
<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T06:39:08Z

A1621205: /* Proposal */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) (see Appendix C) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) (see Appendix C) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== Bluetooth Low Energy ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon (see section 2.1.3.1.1) provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.
====== iBeacon ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.
====== RadBeacon Tag ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned in section 2.1.3.1, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter [1], it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods [1]. Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime [1]. In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter [24] takes a large number of consecutive measurements to create the optional estimation of the state. It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node [1]. The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance [21]. It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon [25] is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications. In particular PiBeacon is able to work with iBeacon (section 2.1.3.1.1) technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
The proposed approach to this project has been defined into two key components that will primarily be undertaken separately (each team member is assigned a component) but simultaneously (each team member will work on their component at the same time) with room for overlap and integration of components. These sections are titled Algorithmic Development and Application Development. The Algorithmic Development component of the project is primarily concerned with the setup and deployment of beacons and the determination of the algorithmic approaches and the application of these positioning algorithms that will be used in the final indoor positioning system. Whereas, the Application Development component is concerned with development of an application for iOS that will utilise these algorithmic approaches determined in the Algorithmic Development component, Graphical User Interface (GUI) design, and the testing procedures involved. The project has been split in this way such, to minimise risk, in that there is a clear cut distinction between components and in the event of failure to meet the demands of one aspect of the project the other aspect will still be semi-functional. However, the overall system will rely on the success of both aspects of the project. The Algorithmic Development component will be further discussed in this section with a brief outline of the Application Development approach.
==== Algorithmic Development ====
The Algorithmic Development component consists of four main stages. The first stage is concerned primarily with the calibration of the beacon devices. The second stage is concerned with the collection of Received Signal Strength Indicator (RSSI) values from different scenarios to be used to form the basis of testing for the positioning algorithms. The third stage involves the discovery, testing and analysis of different positioning algorithms based on the RSSI signal parameter. The fourth and final stage is concerned with the coding of the chosen algorithm/algorithms into MATLAB to be passed on to be used in the final application.
===== Beacon Calibration =====
The initial stage of the Algorithmic Development component of this project is concerned primarily with the calibration of the Radius Networks’ RadBeacon Tag (see section 2.1.3.1.1.1) devices. Further research and experimentation into the trade-offs between Transmission (Tx) power (and hence, battery life) and the signal range will need to be explored in order to make a feasible decision as to an appropriate Tx power level for the purposes of this project. The Advertising Rate (AR) of the RadBeacon Tags will also need to be considered in this initial calibration stage. Due to its nature, this stage may need to be re-factored or undertaken in more details near the end of this project.
===== RSSI Data Collection =====
The next stage of the Algorithmic Development stage will be concerned with further researching and investigating how to obtain the RSSI parameter values from the RadBeacon Tags. At Ingenuity 2014 (see Appendix A) a test scenario was setup such that RSSI values could be obtained in the Adelaide Convention Centre environment (which is where this system is intended to first be deployed). The use of Radius Networks’ PiBeacon (see section 2.3) was utilised in order to obtain and store this data. This step involves the extraction of this data and the assessment of its possible use in assisting in testing the positioning algorithms in step three of this component. Unfortunately the Adelaide Convention Centre will not be accessible up until the date of Ingenuity 2015. Due to this restriction a test site will be simulated at one of the laboratories at the University of Adelaide. At least three beacons will be deployed over a suitable (to be determined) distance using the workbenches provided within the laboratory and multiple tests will be undertaken such that data to be analysed can be obtained. These tests will be left over night and during the day such to simulate the environment of a convention centre which will suffer from signal interference, noise and possible disruption of beacons. These factors present themselves as a technical challenge that the project group must overcome.
Some of the tests that will be included in this stage of the Algorithmic Development component involve:
• Collecting RSSI data for three stationary beacons with known positions with a stationary receiver (PiBeacon)
• Collecting RSSI data for three stationary beacons with known positions with a moving receiver
• Collecting RSSI data for more than three stationary beacons with known positions with a stationary and/or moving receiver
The scope of these tests is not limited to those described above.
===== Positioning Algorithm identification and development =====
The last stage in the Algorithmic Development component of the project is concerned with the research and analysis of different positioning algorithms that can utilise the obtained RSSI values and infer position. The algorithmic approaches that will be explored are discussed in section 2.2. The main objective in this step is the need to determine an algorithm/set of algorithms that are/is both easy to implement (considering it needs to be written as part of the application) and provides the highest feasible position accuracy and precision with the minimum amount of beacons.
===== Positioning Algorithm Coding =====
The last stage of this component of the project is to write up the algorithmic approach/approaches decided in section 3.1.3 into MATLAB such that they can be handed over to be used in a similar fashion for the application.
Throughout the entire Algorithmic Development process the need to obtain the highest feasible position accuracy with the minimal amount of beacons will be looked into and addressed throughout each stage listed above.
==== Application Development ====
The goal of the application development component of the project is to develop an app which applies the algorithms determined in Section 3.1 and the necessary background calculations needed to determine distances/location. The app will allow a user to view a map of the location and what booths/objects are in the location. It will provide a wiki page for the booth that a user is closest to. This stage involves the design of a GUI on the top of the app that enables easy interaction with the system to users. The application development component will involve the design and implementation of the functionalities outlined in Figure 1 below. It will then involve the design, storyboard and development of a GUI which will display the map and wiki page of exhibit points at Ingenuity 2015. An idea of what this will looks like are shown in Figures 2 and 3 below. The final stage of the Application development will be the integration of the algorithmic techniques into the app and testing of the final system.

=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref3>The University of Adelaide. (2015). About Ingenuity 2015. [Online]. Available:http://www.ecms.adelaide.edu.au/ingenuity/about/. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref4>The University of Adelaide. (2015). Projects. [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Main_page#Final_Year_Projects. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref5>The University of Adelaide. (2015). [Online]. Available: https://myuni.adelaide.edu.au/webapps/blackboard/content/listContent.jsp?course_id=_318829_1&content_id=_6466732_1. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref6>Projects:2014S1-51 Heart Signal Processing Software for Evaluating Pacemaker Effectiveness - Projects. (2015). [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Projects:2014S151_Heart_Signal_Processing_Software_for_Evaluating_Pacemaker_Effectiveness. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref14>IDC: Smartphone OS Market Share. (2015). [Online]. Available: http://www.idc.com/prodserv/smartphone-os-market-share.jsp. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>
<ref name=ref16>iBeacon for Developers - Apple Developer. (2015). [Online]. Available:https://developer.apple.com/ibeacon/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref17>Radius Networks. (2015). [Online]. Available: http://developer.radiusnetworks.com/pibeacon/pibeacon-instructions.html. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref18>RadBeacon Config App. (2015). [Online]. Available: http://store.radiusnetworks.com/products/radbeacon-config. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref19>Estimote Community Portal. What are Broadcasting Power, RSSI and Measured Power? (2015). [Online]. Available: https://community.estimote.com/hc/en-us/articles/201636913-What-are-Broadcasting-Power-RSSI-and-Measured-Power-. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref20>E. Lau, W. Chung, "Enhanced RSSI-based Real-time User Location Tracking System for Indoor and Outdoor Environments," IEE, 2007</ref>
<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref23>D. Stojanović, N. Stojanović, "INDOOR LOCALIZATION AND TRACKING: METHODS, TECHNOLOGIES AND RESEARCH CHALLENGES", Faculty of Electronic Engineering, Dept. Computer Science, Republic of Serbia, Vol. 13, No 1, 2014, pp. 57 - 72, 2014</ref>
<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T06:33:10Z

A1621205: /* Indoor Positioning System Technologies */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10"/> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2/> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9/> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices.<ref name=ref1/> Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it.<ref name=ref1/> The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) (see Appendix C) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) (see Appendix C) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost.<ref name=ref1/><ref name=ref12/>
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology.<ref name=ref1/>
====== Bluetooth Low Energy ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery.<ref name=ref1/>
Apple’s standardised protocol iBeacon (see section 2.1.3.1.1) provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves.<ref name=ref11/> BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11.<ref name=ref7/> The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device.<ref name=ref1/> The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal.<ref name=ref1/> Based on this, RSSI will be a primary starting point for investigation throughout this project.
====== iBeacon ======
Introduced in iOS 7, iBeacon technology establishes a region around an object.<ref name=ref15/> It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters:<ref name=ref22/>
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.
====== RadBeacon Tag ======
Radius Networks RadBeacon Tag, as stated in the specifications, is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology.<ref name=ref8/> iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops.<ref name=ref11/><ref name=ref13/> RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned in section 2.1.3.1, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter [1], it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods [1]. Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime [1]. In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter [24] takes a large number of consecutive measurements to create the optional estimation of the state. It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node [1]. The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance [21]. It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon [25] is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications. In particular PiBeacon is able to work with iBeacon (section 2.1.3.1.1) technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref3>The University of Adelaide. (2015). About Ingenuity 2015. [Online]. Available:http://www.ecms.adelaide.edu.au/ingenuity/about/. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref4>The University of Adelaide. (2015). Projects. [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Main_page#Final_Year_Projects. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref5>The University of Adelaide. (2015). [Online]. Available: https://myuni.adelaide.edu.au/webapps/blackboard/content/listContent.jsp?course_id=_318829_1&content_id=_6466732_1. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref6>Projects:2014S1-51 Heart Signal Processing Software for Evaluating Pacemaker Effectiveness - Projects. (2015). [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Projects:2014S151_Heart_Signal_Processing_Software_for_Evaluating_Pacemaker_Effectiveness. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref14>IDC: Smartphone OS Market Share. (2015). [Online]. Available: http://www.idc.com/prodserv/smartphone-os-market-share.jsp. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>
<ref name=ref16>iBeacon for Developers - Apple Developer. (2015). [Online]. Available:https://developer.apple.com/ibeacon/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref17>Radius Networks. (2015). [Online]. Available: http://developer.radiusnetworks.com/pibeacon/pibeacon-instructions.html. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref18>RadBeacon Config App. (2015). [Online]. Available: http://store.radiusnetworks.com/products/radbeacon-config. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref19>Estimote Community Portal. What are Broadcasting Power, RSSI and Measured Power? (2015). [Online]. Available: https://community.estimote.com/hc/en-us/articles/201636913-What-are-Broadcasting-Power-RSSI-and-Measured-Power-. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref20>E. Lau, W. Chung, "Enhanced RSSI-based Real-time User Location Tracking System for Indoor and Outdoor Environments," IEE, 2007</ref>
<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref23>D. Stojanović, N. Stojanović, "INDOOR LOCALIZATION AND TRACKING: METHODS, TECHNOLOGIES AND RESEARCH CHALLENGES", Faculty of Electronic Engineering, Dept. Computer Science, Republic of Serbia, Vol. 13, No 1, 2014, pp. 57 - 72, 2014</ref>
<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T06:27:43Z

A1621205: /* References */

The desire and need for indoor localisation is becoming more widespread in society. In particular, exhibition event-goers are often faced with the problem of wasting time looking for items of interest to them and may find themselves missing out on the exhibits that are more suited to their taste. This Honours Project proposes a solution to this problem. It aims to provide useful advertisement about nearby exhibits to users based on their location. It concerns the development of an indoor positioning system based on Bluetooth Low Energy (BLE) technology and concerns its integration into an application designed for Apple Inc’s Mobile iPhone Operating System (iOS). This project aims to further prove the viability of the use of BLE as an indoor positioning technology that has the potential to become widespread in society and to enhance the experience of event-goers.

== Project Information ==
=== Aims and Objectives ===
The main objective of this project is the development of a simple, adaptable, user friendly application for iOS devices that will be capable of displaying/providing useful information about a user’s surrounding based on their location within a convention sized setting. It aims to utilise Apple Inc’s iBeacon Technology which is a class of BLE device that can broadcast its location to nearby portable electronic devices to serve as transmitters within the system. More specifically Radius Networks’ RadBeacon Tag will be the chosen proximity beacon device based on iBeacon technology although alternative options are in existence. The application/system is intended to be initially deployed at the Adelaide Convention Centre for the Ingenuity 2015 event. It will be targeted towards final year students and teaching staff at the University of Adelaide in the School of Electrical and Electronic Engineering and the general public who will be attending Ingenuity 2015. The application will be capable of locating the nearest exhibit booth to a user and will provide the name and description by popping up with its related wiki page. The application/system is expected to be adaptable such that it can be expanded into other locations with similar application.

=== Background ===
==== Indoor Positioning System Technologies ====
The use of GPS in the context of indoor localisation has many limitations. Consequently, alternative technologies have been proposed and discussed in literature.<ref name="ref10">Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref> Some of the proposed alternatives include: Wi-Fi, ZigBee, Ultra wideband radio, and Bluetooth. Although an interesting study of the use of GPS for indoor localisation was found, the technologies that have caught the most attention in the field of indoor positioning are Bluetooth and Wi-Fi and as such literature pertaining to these will be primarily discussed. For the purposes of this project, Bluetooth and more specifically BLE propose many advantages over other technologies and these will be discussed.
===== Global Positioning System =====
An indoor positioning system, based on GPS repeaters and modified positioning algorithms, was proposed in an article named "Indoor positioning based on global positioning system signals".<ref name=ref2> K. Ozsoy, A. Bozkurt and I. Tekin, 'Indoor positioning based on global positioning system signals', Microwave and Optical Technology Letters, vol. 55, no. 5, pp. 1091-1097, 2013.</ref> The results indicated that an Indoor Positioning System (IPS) based on GPS technologies can produce positioning accuracy comparable to that of outdoors GPS positioning with the restraint of a need for additional infrastructure in addition to buildings. Although this additional infrastructure is cost effective and simple it is not a feasible solution for this project in which light and easily deployable beacons is a necessity.
===== WI-FI =====
A study on the effects of indoor localisation techniques in a Wi-Fi access point (AP) intense environment is presented in an article named “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs”.<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref> Although the technology and algorithms applied in this study were based on Wi-Fi, the results can be useful in this project. In particular, the algorithms explored throughout this paper may prove useful when applied in a Bluetooth context. This paper alludes to some factors that may need to be considered throughout this project that have the potential to affect the accuracy of localisation. These include: the effect of a large amount of APs (or beacons), time tolerance i.e. signal strength decreasing over time, APs disappearing and appearing, different mobile device specifications and orientation of devices.
===== Bluetooth =====
Bluetooth is a technology that has received much attention in the field of indoor positioning. It is designed to enable short range wireless communication between devices [1]. Bluetooth uses radio signals in the 2.4 GHz range to transmit data between devices. It has been designed to support low power wireless communication and hence, power control is one of the features associated with it [1]. The power control feature allows a transmitter to adjust its strength based on the Received Signal Strength Indicator (RSSI) (see Appendix C) from another device. RSSI, which is the signal parameter of focus within this project, is among other Bluetooth measures that can be used as input data into positioning algorithms. Other measures that have been looked into are Link Quality (LQ) which is another measure that can be picked up by a Bluetooth receiver and the transmitted power level (TPL) (see Appendix C) which is transmitted. One major advantage attributed to the use of Bluetooth technology is that it has high penetration in society meaning that Bluetooth capable devices are easily obtainable and the required hardware is mass produced such that Bluetooth technology has a very low unit cost [1][12].
In 2010, Bluetooth Low Energy (BLE) was introduced along with the specification for Bluetooth 4.0 technology [1].
====== Bluetooth Low Energy ======
Bluetooth Low Energy (BLE), as the name implies, was primarily designed to produce lower power consumption than classic Bluetooth technology whilst maintaining a similar communication range. This attribute is ideal for this project as amongst other trade-offs, power consumption and battery life of a BLE enabled device have a direct correlation. It is claimed that the low power consumption of BLE capable devices means they can potentially last for years powered by a single coin cell battery [1].
Apple’s standardised protocol iBeacon (see section 2.1.3.1.1) provides a means for BLE enabled devices to communicate through “Advertisements.” These “Advertisements” are small packets of data, that are broadcasted at regular intervals by beacons or other BLE enabled devices through the use of radio waves [11]. BLE contains three dedicated advertisement channels: 37, 38, and 39, which have been specifically allocated in the frequency spectrum such to minimise interference with the most common Wi-Fi channels 1, 6 and 11 [7]. The use of these three dedicated advertisement channels provides a speed advantage over tradition Bluetooth as the time it takes to locate all beacons within a vicinity is dependent on the interval in which the beacons are programmed to perform advertisement.
There are several trade-offs that need to be considered with the use of BLE technology. The first trade-off is between advertisement time and device discovery. BLE devices may advertise as little as once every 10 seconds or as fast as once every 20 milliseconds. This advertisement interval has a direct correlation to the time it takes to discover a device [1]. The transmit range of BLE based beacons has a direct correlation with the Transmission (Tx) power level. This is another trade-off that must be taken into account when using BLE based beacons for indoor localisation.
A limitation to the use of BLE for indoor localisation is that the LQ parameter is no longer able to be obtained and beacon devices based on BLE typically have a fixed transmission (Tx) power level implying that the use of the TPL parameter is also no longer a viable option. Fortunately, the RSSI parameter is still easily accessible in BLE via simply receiving a broadcasted message. It is claimed to be the parameter within Bluetooth that has received the most attention and consensus to be the best suited for positioning applications despite its flaw in not being optimal [1]. Based on this, RSSI will be a primary starting point for investigation throughout this project.
====== iBeacon ======
Introduced in iOS 7, iBeacon technology establishes a region around an object [15]. It allows an iOS device to detect when it has entered or left this established region, along with an estimation of proximity to a beacon. A key feature of iBeacon technology is that iOS devices can be configured to generate iBeacon advertisements, which is crucial to this project.
A typical advertisement packet will contain the following parameters [22]:
• Proximity UUID (16 byte): which is a unique identifier used to distinguish a set of beacons from another. For example if a set of beacons were used to present special offers to customers in a chain of stores, beacons with the same proximity UUID would be associated with the same chain and such an application for an iOS device for a specific chain will be configured to search for beacons with that associated UUID.
• Major (2 byte): which is a decimal number used to group a set of related beacons. For example all beacons inside the same location will have the same major number which will help distinguish one location from another.
• Minor (2 bytes): which is a decimal number used to distinguish individual beacons within a group of beacons. For example each beacon within a particular location will have a different minor number in order to help distinguish which part of that location the user is in.
• Transmission Power (Tx) (2 bytes): is a representation of the strength of the signal measured at 1 metre from the device.
For this project Radius Networks’ RadBeacon Tag was chosen as the device that can implement this iBeacon technology. RadBeacon Tags are small, easily deployable and able to provide the RSSI parameter which makes them an ideal choice for this project.
====== RadBeacon Tag ======
Radius Networks RadBeacon Tag, as stated in the specifications [8], is a fully standalone Bluetooth Low Energy (BLE) proximity beacon that uses iBeacon technology. iBeacon technology has already been successfully applied in society for both indoor seat location/directing in large arenas and for advertisement purposes within shops [11][13]. RadBeacon Tags are powered by a replaceable coin-cell battery and can be adjusted to provide signal ranges from 5m to 50m. RadBeacon Tags have a configurable transmit power range of +4dBm to -20dBm and a configurable Advertisement Rate (AR) of 1Hz to 10Hz. The battery life of the RadBeacon Tags has a direct relation to the AR. A RadBeacon Tag battery will typically last 30 days on a High AR (100ms), 145 days on a medium AR (500ms) and 285 days on a Low AR.

==== Positioning Algorithms ====
The core of any localisation method relies on the real-time measurement of one to several parameters, such as angles, distances, or distance differences [23]. These measurement parameters reflect the relative location of a target object (an iOS device) to a single or several fixed points (beacons) with known locations. As mentioned in section 2.1.3.1, the measurement parameter of interest within this project is the RSSI. In order to infer position several types of algorithms have been proposed throughout literature. In a study on the best types of algorithms suited for indoor positioning based on BLE technology and the RSSI parameter [1], it was concluded that the most suited types of algorithms are: Trilateration, Particle Filters and Fingerprinting.
===== Trilateration =====
In the context of location estimation, Trilateration is one of the oldest and most well renown methods [1]. Trilateration algorithms require three or more reference points in order to accurately determine distance. These three or more reference points are used to form the centres of three or more circles where the distances are considered to be the radii of these circles. In order to determine the relative or absolute position, simple geometric equations are used to calculate the intersection point between these circles. In the context of indoor localisation based on BLE, the three or more reference points are known reference positions determined by fixed beacon nodes and the distances can be obtained from received signal strength parameters. In the context of this project, the RSSI parameter values can be used.
===== Particle Filters =====
Particle filter algorithms involve the random and continuous generation of thousands of particles during runtime [1]. In the context of indoor positioning estimation these particles are the distance/difference to each of the reference beacons or data/values that can be obtained. A filtering method is applied to this data where the particles that are unlikely to represent the current position are filtered out of the data set while particles that are more likely to be true representations of the position are retained for calculating the estimation.
The Kalman Filter is a type of Particle Filter that was considered within this project.
===== Kalman Filter =====
The Kalmin Filter [24] takes a large number of consecutive measurements to create the optional estimation of the state. It relies on two phases that occur repeatedly. These are a prediction and correction phase. The first phase is a calculation of a predicted state in the next moment and the second phase is where this prediction is corrected using a measured value. It is suspected that this process of prediction and correction can be used to filter out RSSI values that may have been affected by a large signal fluctuation.
===== Fingerprinting =====
In fingerprinting, attributes are assigned to a segment or grid cell. In the context of Indoor positioning estimation, RSSI or other received signal strength parameters are used as unique attributes. Fingerprinting involves the division of a map into segments or grids. In terms of RSSI, these segments consist of an average of RSSI values per beacon node [1]. The fingerprint is thus made up of several RSSI averages and the position on the map segment or the grid cell that the fingerprinting belongs to. Ultimately, a fingerprint database is constructed.
The k-Nearest Neighbour (k-NN) is a type of algorithm based on Fingerprinting that was considered within this project.
===== K-Nearest Neighbour (k-NN) =====
The output of a k-Nearest Neighbour (k-NN) algorithm is based on a comparison between an input vector and the training set (or fingerprint database) using a distance measure such as Euclidean distance [21]. It selects k-instances of the data set that are most similar to the input (those that have the shortest distance to the input) and those instances determine the class (or location) of the output. The class that has the majority of votes among the k-nearest neighbours of the input is selected as the output. For the purposes of this project k-NN can be applied by setting up a database of advertisement packet data within a location. The advertisement packet data received by a user at any point within this location can then be compared with the k-nearest neighbouring database reference points (which will be associated with an object (or booth) ) and the object/booth with the most matching points will be determined as the closest location to the user.
==== PiBeacon ====
Radius Networks’ PiBeacon [25] is a Raspberry Pi based proximity beacon tool that is designed to work with multiple Bluetooth Low Energy proximity beacon specifications. In particular PiBeacon is able to work with iBeacon (section 2.1.3.1.1) technology. PiBeacons have the ability to transmit as a beacon or to scan for nearby beacons. As such PiBeacon has been identified as a potential substitute for an iOS device in the testing of the algorithmic development stage of the project.

=== Proposal ===
=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==
{{reflist|refs=
<ref name=ref1>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref2>E. Dahlgren and H. Mahmood, “Evaluation of indoor positioning based on Bluetooth Smart technology,” M.S. thesis, Dept. Computer Science and Engineering, Chalmers University of Technology, Göteborg, Sweden, 2014.</ref>
<ref name=ref3>The University of Adelaide. (2015). About Ingenuity 2015. [Online]. Available:http://www.ecms.adelaide.edu.au/ingenuity/about/. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref4>The University of Adelaide. (2015). Projects. [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Main_page#Final_Year_Projects. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref5>The University of Adelaide. (2015). [Online]. Available: https://myuni.adelaide.edu.au/webapps/blackboard/content/listContent.jsp?course_id=_318829_1&content_id=_6466732_1. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref6>Projects:2014S1-51 Heart Signal Processing Software for Evaluating Pacemaker Effectiveness - Projects. (2015). [Online]. Available: https://www.eleceng.adelaide.edu.au/students/wiki/projects/index.php/Projects:2014S151_Heart_Signal_Processing_Software_for_Evaluating_Pacemaker_Effectiveness. [Accessed: 16- Apr- 2015].</ref>
<ref name=ref7>E. Mackensen, M. Lai, T. Wendt, Bluetooth low energy (ble) based wireless sensors,in: Sensors, 2012 IEEE, 2012, pp. 1-4.</ref>
<ref name=ref8>Radius Networks. (2015). RadBeacon Tag Specs. [Online]. Available:http://www.radiusnetworks.com/ibeacon/radbeacon/tag/specs.html. [Accessed:17- Apr- 2015].</ref>
<ref name=ref9>X. Chen, J. Kong, Y. Guo and X. Chen, “An Empirical Study of IndoorLocalization Algorithms with Densely Deployed APs,” IEEE, pp. 517 – 522, 2014</ref>
<ref name=ref10>Hossain, A.K.M.M.; Hien Nguyen Van; Yunye Jin; Wee-Seng Soh, "Indoor Localization Using Multiple Wireless Technologies," Mobile Adhoc and Sensor Systems, 2007. MASS 2007. IEEE International Conference on , vol., no., pp.1,8, 8-11 Oct. 2007</ref>
<ref name=ref11>iBeacon Insider. (2014). 'What is iBeacon? A Guide to iBeacons'. [Online]. Available:http://www.ibeacon.com/what-is-ibeacon-a-guide-to-beacons/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref12>Oksar, I., "A Bluetooth signal strength based indoor localization method," Systems, Signals and Image Processing (IWSSIP), 2014 International Conference on , vol., no., pp.251,254, 12-15 May 2014</ref>
<ref name=ref13>How Stadiums can use Beacons to Enhance Fans’ Experiences. (2015). [Online]. Available: http://blog.beaconstac.com/how-stadiums-can-use-beacons-to- enhance-fans/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref14>IDC: Smartphone OS Market Share. (2015). [Online]. Available: http://www.idc.com/prodserv/smartphone-os-market-share.jsp. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref15>Getting Started with iBeacon. (2014). [Online]. Available: https://developer.apple.com/ibeacon/Getting-Started-with-iBeacon.pdf. [Accessed: 26- Mar- 2015].</ref>
<ref name=ref16>iBeacon for Developers - Apple Developer. (2015). [Online]. Available:https://developer.apple.com/ibeacon/. [Accessed: 17- Apr- 2015].</ref>
<ref name=ref17>Radius Networks. (2015). [Online]. Available: http://developer.radiusnetworks.com/pibeacon/pibeacon-instructions.html. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref18>RadBeacon Config App. (2015). [Online]. Available: http://store.radiusnetworks.com/products/radbeacon-config. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref19>Estimote Community Portal. What are Broadcasting Power, RSSI and Measured Power? (2015). [Online]. Available: https://community.estimote.com/hc/en-us/articles/201636913-What-are-Broadcasting-Power-RSSI-and-Measured-Power-. [Accessed: 03- Jun- 2015].</ref>
<ref name=ref20>E. Lau, W. Chung, "Enhanced RSSI-based Real-time User Location Tracking System for Indoor and Outdoor Environments," IEE, 2007</ref>
<ref name=ref21>A. Bekkelien, "Bluetooth Indoor Positioning", M.S these, Dept. Computer Science, University of Geneva, March 2012.</ref>
<ref name=ref22>How do iBeacons Work? (2015). [Online]. Available: http://java.dzone.com/articles/how-do-ibeacons-work?page=0,1. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref23>D. Stojanović, N. Stojanović, "INDOOR LOCALIZATION AND TRACKING: METHODS, TECHNOLOGIES AND RESEARCH CHALLENGES", Faculty of Electronic Engineering, Dept. Computer Science, Republic of Serbia, Vol. 13, No 1, 2014, pp. 57 - 72, 2014</ref>
<ref name=ref24>How To Determine Location If You have A Roof Over Your Head. (2015). [Online]. Available: http://blog.lemberg.co.uk/how-determine-location-if-you-have-roof-over-your-head. [Accessed: 05- Jun- 2015].</ref>
<ref name=ref25>PiBeacon. (2015). [Online]. Available: http://store.radiusnetworks.com/products/pibeacon. [Accessed: 05- Jun- 2015].</ref>
}}

== Booth List (for internal use only) ==
This section is intended for the use of project teams. It contains the locations of each project booth in the Convention Centre.

*<project_01> (1,0,0)
*<project_04> (2,0,0)
*<project_05> (3,0,0)
*<project_06> (4,0,0)
*<project_07> (5,0,0)
*<project_08> (6,0,0)
*<project_10> (7,0,0)
*<project_11> (8,0,0)
*<project_12> (9,0,0)
*<project_13> (10,0,0)
*<project_15> (11,0,0)
*<project_16> (12,0,0)
*<project_17> (13,0,0)
*<project_18> (14,0,0)
*<project_21> (15,0,0)
*<project_25> (16,0,0)
*<project_26> (17,1,2)
*<project_28> (18,0,0)
*<project_31> (19,0,0)
*<project_32> (20,0,0)
*<project_36> (21,0,0)
*<project_40> (22,0,0)
*<project_42> (23,0,0)
*<project_46> (24,0,0)
*<project_45> (25,0,0)
*<project_50> (26,3,4)
*<project_56> (27,0,0)
*<project_61> (28,0,0)
*<project_70> (29,0,0)
*<project_73> (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T06:19:12Z

A1621205: /* Indoor Positioning System Technologies */

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T06:10:33Z

A1621205: /* Indoor Positioning System Technologies */

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T06:08:39Z

A1621205: /* References */

Template:Reflist

2015-08-26T06:00:48Z

A1621205: Replaced content with "{{#tag:references|{{{refs|}}}|group={{{group|}}}}}"

{{#tag:references|{{{refs|}}}|group={{{group|}}}}}

Template:Reflist

2015-08-26T05:59:45Z

A1621205:

<div class="reflist {{#if: {{{1|}}}
| columns {{#iferror: {{#ifexpr: {{{1|1}}} > 1 }}
| references-column-width
| references-column-count references-column-count-{{#if:1|{{{1}}}}} }}
| {{#if: {{{colwidth|}}}
| references-column-width }} }}" style="{{#if: {{{1|}}}
| {{#iferror: {{#ifexpr: {{{1|1}}} > 1 }}
| {{column-width|{{#if:1|{{{1}}}}}}}
| {{column-count|{{#if:1|{{{1}}}}}}} }}
| {{#if: {{{colwidth|}}}
| {{column-width|{{{colwidth}}}}} }} }} list-style-type: {{{liststyle|{{#switch: {{{group|}}}
| upper-alpha
| upper-roman
| lower-alpha
| lower-greek
| lower-roman = {{{group}}}
| #default = decimal}}}}};">
{{#tag:references|{{{refs|}}}|group={{{group|}}}}}</div><noinclude>
</noinclude>

Template:Reflist

2015-08-26T05:39:48Z

A1621205: Blanked the page

Template:Reflist

2015-08-26T05:37:54Z

A1621205: Created page with "<div class="reflist {{#if: {{{1|}}} | columns {{#iferror: {{#ifexpr: {{{1|1}}} > 1 }} | references-column-width | references-column-count references-..."

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T05:06:45Z

A1621205: /* Background */

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-26T04:56:52Z

A1621205: /* Aims and Objectives */

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-20T16:32:14Z

A1621205: /* Booth List (for internal use only) */

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-20T16:31:07Z

A1621205: /* Booth List (for internal use only) */

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-20T13:41:54Z

A1621205:

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-20T13:37:09Z

A1621205:

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-20T13:22:04Z

A1621205:

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-13T12:54:34Z

A1621205: /* Booth List (for internal use only) */

introduction here
== Project Information ==
=== Aims and Objectives ===
=== Background ===
=== Proposal ===
=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==

== Booth List (for internal use only) ==
*01 (1,0,0)
*04 (2,0,0)
*05 (3,0,0)
*06 (4,0,0)
*07 (5,0,0)
*08 (6,0,0)
*10 (7,0,0)
*11 (8,0,0)
*12 (9,0,0)
*13 (10,0,0)
*15 (11,0,0)
*16 (12,0,0)
*17 (13,0,0)
*18 (14,0,0)
*21 (15,0,0)
*25 (16,0,0)
*26 (17,0,0)
*28 (18,0,0)
*31 (19,0,0)
*32 (20,0,0)
*36 (21,0,0)
*40 (22,0,0)
*42 (23,0,0)
*46 (24,0,0)
*45 (25,0,0)
*50 (26,0,0)
*56 (27,0,0)
*61 (28,0,0)
*70 (29,0,0)
*73 (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-13T12:52:14Z

A1621205: /* Booth List (for internal use only) */

introduction here
== Project Information ==
=== Aims and Objectives ===
=== Background ===
=== Proposal ===
=== Outcome ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== References ==

== Booth List (for internal use only) ==
01 (1,0,0)
04 (2,0,0)
05 (3,0,0)
06 (4,0,0)
07 (5,0,0)
08 (6,0,0)
10 (7,0,0)
11 (8,0,0)
12 (9,0,0)
13 (10,0,0)
15 (11,0,0)
16 (12,0,0)
17 (13,0,0)
18 (14,0,0)
21 (15,0,0)
25 (16,0,0)
26 (17,0,0)
28 (18,0,0)
31 (19,0,0)
32 (20,0,0)
36 (21,0,0)
40 (22,0,0)
42 (23,0,0)
46 (24,0,0)
45 (25,0,0)
50 (26,0,0)
56 (27,0,0)
61 (28,0,0)
70 (29,0,0)
73 (30,0,0)

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-12T05:55:45Z

A1621205:

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-12T05:54:14Z

A1621205:

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-12T05:42:54Z

A1621205:

== Project Information ==
=== Aims and Objectives ===
=== Background ===

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
=== Supervisors ===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim A/Prof Cheng-Chew Lim]

== Booth List (for internal use only) ==

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-12T05:21:49Z

A1621205: /* Supervisors */

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
===Supervisors===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim Associate Professor Cheng-Chew Lim]

==Booth List (for internal use only)==

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-12T05:07:44Z

A1621205: /* Supervisors */

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
===Supervisors===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*[http://www.adelaide.edu.au/directory/cheng.lim Prof Cheng Chew Lim]

==Booth List (for internal use only)==

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-12T05:05:20Z

A1621205: /* Supervisors */

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
===Supervisors===
*[http://www.adelaide.edu.au/directory/honggunn.chew Dr Hong Gunn Chew]
*Prof Cheng Chew Lim

==Booth List (for internal use only)==

Projects:2015s1-25 Indoor localisation using Bluetooth LE for event advertising

2015-08-12T05:02:29Z

A1621205:

== Project Team ==
=== Group members ===
*Mr Huajian Huang
*Ms Kyriaki Georgia Morias
===Supervisors===
*Dr Hong Gunn Chew
*Prof Cheng Chew Lim
==Booth List (for internal use only)==