Projects:2019s1-105 Hacking CAN Bus

Utilising a testbed environment to develop and demonstrate man-in-the-middle deception attacks on the internal communication networks of modern vehicles

Modern cars have multiple dedicated computers that control all of the car’s operations, such as the engine, braking, steering and entertainment. These computers are called electronic control units (ECUs) and communicate primarily by a highly vulnerable serial network, the CAN bus protocol.

Cyber security concerns surrounding these vehicles are increasing, particularly with the introduction of Wi-Fi and other wireless vehicle services. Using these wireless services, adversaries can find their way on the CAN bus and gain control of the vehicle. In a different scenario, infiltrating a vehicle and connecting a device to subtly alter a vehicle’s display information, without any overt consequences, may result in an unaware user who, at best, might have a slight suspicion.

The research presented in this paper discusses this type of security threat and the mistrust it can cause. The intent of the research is to assert the usefulness of a testbed environment in exploiting the vulnerabilities of the CAN bus protocol by developing and weaponizing a deceptive man-in-the-middle type attack.

Project Team

Student Researchers

Stefan Smiljanic

Charlie Tran

Project Supervisors

Dr. Matthew Sorell

Aaron Frishling (DSTG)

Bradley Cooney (DSTG)

Daniel Coscia (DSTG)

CAN Controller Area Network ECU Electronic Control Unit MITM Man-in-the-Middle MOTS Man-on-the-Side OBD On Board Diagnostics

Research Questions

• How can a testbed environment be utilised for safe and effective research in automotive security?
• Using a testbed environment, how can the vulnerabilities of the CAN bus be exploited to perform a deceptive MITM type attack?

Related Work and Motivation

To be filled...

Objectives

To be filled...

Method

To be filled...

Results

To be filled...

References

To be filled...