Detection and Prevention From Misbehaving Intruders in Vehicular Networks

In this paper, we design and implement a new intrusion detection and prevention schema for vehicular networks. It has the ability to detect and predict with a high accuracy a future malicious behavior of an attacker. This is unlike the current detection schemas, where there is no prevention technique since they aim to detect only current attackers that occur in the network. We used game theory concept to predict the future behavior of the monitored vehicle and categorize it into the appropriate list (White, White & Gray, Gray, and Revocation_Black) according to its predicted attack severity. In this paper, our aim is to prevent from the most dangerous attack that targets a vehicular network, which is false alert's generation attack. Simulation results show that our intrusion detection and prevention schema exhibits a high detection rate and generates a low false positive rate. In addition, it requires a low overhead to achieve a high-level security.