In the light of the attacks on Controller Area Networks (CAN) recorded over the past decade, detecting intrusions has become a critical demand. While cryptographic mechanisms are largely absent on CAN buses and clever adversaries may evade intrusion detection mechanisms that rely solely on traffic analysis, using physical signal characteristics to detect the source of incoming frames started to attract a lot of interest in the recent years. This technique is based on physical imperfections of transceivers and microcontrollers as well as network characteristics that are hard if not impossible to clone. In this chapter we discuss the use of voltage fingerprints for source identification as well as the recently emerged topic of localizing controllers by means of signal propagation time. These techniques can have a number of applications ranging from forensics, the detection of unauthorized components and as a complementary mechanism to traditional cryptographic protection and intrusion detection mechanisms.
Physical Layer Intrusion Detection and Localization on CAN Bus
Machine Learning and Optimization Techniques for Automotive Cyber-Physical Systems ; Kapitel : 13 ; 399-423
2023-03-27
25 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Physical structure for vehicle intrusion detection device
Europäisches Patentamt | 2022
|PHYSICAL STRUCTURE FOR VEHICLE INTRUSION DETECTION DEVICE
Europäisches Patentamt | 2021
|