Research on the Intrusion Detection Methods of Automotive Networks Based on Generative Adversarial Networks
Wei Xu, Guojun Huang · 2024
With the popularization of intelligent connected cars, the number of interactions between the vehicle's interior and the external network has increased, and the difficulty of malicious attacks on the car's internal network has also been greatly reduced. In order to effectively detect these malicious attacks, an intrusion detection system (IDS) is required. Unlike IDS, there are few known attack signatures for vehicle networks. In addition, IDS for vehicles requires high accuracy because any false positive error can seriously affect driver safety. In order to solve this problem, an IDS based on the deep learning Model-Generative Adversarial Nets is proposed. Experimental results show that this model has high accuracy in detecting malicious attacks and can effectively detect intrusions.