Difference between revisions of "Projects:2019s1-101 Behavioural Analytics of Mobile Applications"

From Projects
Jump to: navigation, search
(Created page with "The last two decades have marked a dramatic shift in our reliance on the internet. The most significant catalyst for our newfound reliance has been the universal adoption of m...")
 
(Introduction)
 
Line 6: Line 6:
 
__TOC__
 
__TOC__
 
== Introduction ==
 
== Introduction ==
Intro here
+
 
 +
Network visibility is of paramount importance from a variety of standpoints, from security to general systems maintenance, and being able to determine the devices and applications on a network is essential. This however is becoming increasingly difficult in modern day networking as increasingly secure encryption protocols make traditional (packet inspection based) network analytics vastly more complex. A potential solution to this issue is to perform network analytics based on network flows, rather than packet contents or other more specific information. Network flows are sequences of packets sent from one source to another over a network, and these flows may be analysed and used to determine selected characteristics.
 +
 
 +
This project intends to analyse network flows to characterise traffic based on the application which generated it, and characteristics of the device from which it originated. It also intends to expand on a testbed made up of physical and emulated mobile devices, which are able to be automated to perform a variety of user actions (such as navigating applications) without a human user being present. This work expands on the that of a number of previous years honours projects at the University of Adelaide, and is to be completed under sponsorship and guidance of Defense Science and Technology, formally known as DSTG.
  
 
==Project Team ==
 
==Project Team ==

Latest revision as of 22:17, 14 April 2019

The last two decades have marked a dramatic shift in our reliance on the internet. The most significant catalyst for our newfound reliance has been the universal adoption of mobile devices (i.e., smartphones and tablets), which now account for over half of all time spent online. From games to online banking, mobile applications have made their way into almost every market. This ubiquity however has also made apps a prime target for user analytics and malicious attack.

This project aims to characterise the behaviour of apps through analysis of their network traffic, and from this, infer specific characteristics of the device from which that traffic originated.

Introduction

Network visibility is of paramount importance from a variety of standpoints, from security to general systems maintenance, and being able to determine the devices and applications on a network is essential. This however is becoming increasingly difficult in modern day networking as increasingly secure encryption protocols make traditional (packet inspection based) network analytics vastly more complex. A potential solution to this issue is to perform network analytics based on network flows, rather than packet contents or other more specific information. Network flows are sequences of packets sent from one source to another over a network, and these flows may be analysed and used to determine selected characteristics.

This project intends to analyse network flows to characterise traffic based on the application which generated it, and characteristics of the device from which it originated. It also intends to expand on a testbed made up of physical and emulated mobile devices, which are able to be automated to perform a variety of user actions (such as navigating applications) without a human user being present. This work expands on the that of a number of previous years honours projects at the University of Adelaide, and is to be completed under sponsorship and guidance of Defense Science and Technology, formally known as DSTG.

Project Team

Students

  • Leon Dimitrakopoulos
  • Dave Rattan
  • Leo Hague

Supervisors

  • Dr. Hong Gunn Chew
  • Dr. Cheng-Chew Lim
  • Dr. Adriel Cheng (DST)
  • Mr. Kyle Millar