An Integrated Security VANET Algorithm for Threat Mitigation and Performance Improvement Using Machine Learning

Vehicular Ad-Hoc Networks (VANETs) have emerged as a captivating field of research due to the escalating number of vehicles on the road in recent years. Ensuring a secure Intelligent Transportation System (ITS) is imperative for passenger and driver safety. However, the dynamic nature of VANETs presents challenges for real-time implementation. This study proposes an enhanced security algorithm tailored for VANETs, adept at mitigating threats such as Denial of Service Attacks (DoS), Sybil, and Replay attacks. The proposed method employs an Enhanced K-Means approach to form clusters for diverse attacks, coupled with a hybrid technique utilizing Support Vector Machine (SVM) and Feed-forward backpropagation for classifier accuracy assessment which incorporates firefly algorithm for optimization. The findings demonstrate notable enhancements in terms of Throughput, Jitter, True Detection Rate, and Packet Delivery Ratio (PDR). Lastly, we delineate future directions and identify open issues for further investigation.