Visible to the public "IntelliCAN: Attack-resilient Controller Area Network (CAN) for secure automobiles"

Title"IntelliCAN: Attack-resilient Controller Area Network (CAN) for secure automobiles"
Publication TypeConference Paper
Year of Publication2015
AuthorsAnsari, M. R., Yu, S., Yu, Q.
Conference Name2015 IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems (DFTS)
ISBN Number978-1-4799-8606-4
Keywordsattack-resilient controller area network, automobile, automobiles, CAN protocols, computer network security, controller area network(CAN), controller area networks, Discrete Fourier transforms, error detection, Fault tolerance, Fault tolerant systems, hardware security, IntelliCAN, masquerade attack, mechanical engineering computing, Nanotechnology, pubcrawl170106, security, Very large scale integration

Controller Area Network (CAN) is the main bus network that connects electronic control units in automobiles. Although CAN protocols have been revised to improve the vehicle safety, the security weaknesses of CAN have not been fully addressed. Security threats on automobiles might be from external wireless communication or from internal malicious CAN nodes mounted on the CAN bus. Despite of various threat sources, the security weakness of CAN is the root of security problems. Due to the limited computation power and storage capacity on each CAN node, there is a lack of hardware-efficient protection methods for the CAN system without losing the compatibility to CAN protocols. To save the cost and maintain the compatibility, we propose to exploit the built-in CAN fault confinement mechanism to detect the masquerade attacks originated from the malicious CAN devices on the CAN bus. Simulation results show that our method achieves the attack misdetection rate at the order of 10-5 and reduces the encryption latency by up to 68% over the complete frame encryption method.

Citation Keyansari_intellican:_2015