作者: Mirco Marchetti , Dario Stabili , Alessandro Guido , Michele Colajanni
DOI: 10.1109/RTSI.2016.7740627
关键词:
摘要: This paper evaluates the effectiveness of information-theoretic anomaly detection algorithms applied to networks included in modern vehicles. In particular, we focus on providing an experimental evaluation detectors based entropy. Attacks in-vehicle were simulated by injecting different classes forged CAN messages traces captured from a licensed vehicle. Experimental results show that if entropy-based is all it only possible detect attacks comprise high volume messages. On other hand, characterized injection few can be detected applying several independent instances entropy detector, one for each class