Anomaly Intrusion Detection for CAN-FD Bus by Support Vector Machine

Abstract:

Aiming at the potential network attack for vehicles， an abnormal intrusion detection algorithm based on support vector machine is proposed for the controller area network with flexible data-rate （CAN-FD） bus. With the framework of common intrusion detection framework （CIDF）， the method uses message identifier （ID）， time period and data field as intrusion detection features. Using the binary classification and small sample features of the support vector machine algorithm， the identification of intrusion message data in CAN-FD network environment is realized. The simulation data show that the proposed method has a high accuracy of intrusion detection， and this method can be used for periodic packets and aperiodic messages.

Key words:cyber security;the controller area network with flexible data-rate （CAN-FD);intrusion detection;support vector machine