Modern vehicles are equipped with electronic control units (ECUs) which are connected through an in-vehicle controller area network (CAN) to control the electrical systems or subsystems in a vehicle, such as engine and powertrain systems. With the proliferation of V2X applications, more and more vehicles are to be connected to the Internet. Attackers then may find new attacking vectors to penetrate into the vehicle to affect or steal the vehicle control, which raises serious safety concerns. CAN bus intrusion detection is an important function to identify malicious attacks on the bus in realtime. Mitigation or recovery mechanisms can be triggered if a malicious attack on the CAN bus is detected at its early stage before it could cause any damage. This project aims to develop realtime CAN bus intrusion detection systems which include a local monitor and a cloud-based service. A small-scale CAN bus testbed will be built based on PiCAN and Raspberry Pi.
-
Amanuel Kidanu
- Project Manager
- Cloud Infrastructure (AWS)
- Hardware Implementation (Raspberry Pi)
-
Lia Chiflemariam
- Machine Learning Research and Implementation
-
Luke Francis
- Cloud Infrastructure (AWS)
- Hardware Implementation (Raspberry Pi)
-
Walid Jami
- Machine Learning Research and Implementation
-
Dr. Kai Zeng
- Faculty Supervisor