EAMI: Efficient Privacy Preserving Authentication and Machine Learning Based Intrusion Detection in VANETs

Vehicular ad hoc networks (VANETs) offer an effective communication platform between the vehicles and Road Side Units (RSUs), providing the backbone of intelligent transportation systems. However, ensuring secure communication in VANET is a major challenge due to the Open nature of wireless communication. Authentication plays a crucial role in VANET security, as it checks the authenticity of the users communicating in the network. This paper explores anonymous authentication with the support of Elliptic Curve Cryptography (ECC) and bilinear pairing to allow only the authenticated entities to communicate in VANETs. Moreover, Support Vector Machine (SVM) based intrusion detection system is introduced in this paper to avoid unauthorized intrusion of malicious users. Further the performance of the proposed EAMI scheme is analyzed in terms of computational complexity with other existing authentication mechanisms to guarantee the efficiency of the proposed work. Through experimental analysis, the effectiveness of SVM based intrusion detection system is demonstrated in this paper to identify various types of intrusion, providing a robust solution to secure VANETs.