Real Time Anomaly Detection and Intrusion Detection for Safeguarding Intra-Vehicle Communication Powered by AI

Kumar, Chitoor Venkat Rao Ajay and Venkatagirish, Parnam and Patibandla, Sai Srinivas and Rathod, Kapil (2025) Real Time Anomaly Detection and Intrusion Detection for Safeguarding Intra-Vehicle Communication Powered by AI. World Journal of Advanced Research and Reviews, 25 (1). pp. 1992-2000. ISSN 2581-9615

This study addresses cyber-attacks in Electric Vehicles (EVs) and proposes an intelligent, secure framework to protect both in-vehicle and vehicle-to-vehicle communication systems. The proposed model uses an improved support vector machine (SVM) for anomaly and intrusion detection based on the Controller Area Network (CAN) protocol a critical component in vehicle communication. To further enhance detection speed and accuracy a new optimization algorithm the Social Spider Optimization (SSO) is introduced for reinforcing the offline training process. Simulation results on real-world datasets demonstrate the model's high performance, reliability and ability to defend against denial-of-service (DoS) attacks in EVs.