A LightGBM-Based Approach to improve Sybil Attack Detection in VANET networks

Intelligent Transport Systems (ITS) usually rely on Vehicular Ad Hoc Networks (VANETs) to efficiently enable communication and information exchange between vehicles and the infrastructure. However, the openness and dynamic nature of these networks make them vulnerable to security issues. In fact, VANET networks can suffer from several types of attack due to the large number of participants and the homogeneity of the infrastructure. However, we only focus on the Sybil attack that involve malicious entities creating multiple fake identities to disrupt network operations, compromise data integrity, and deceive legitimate vehicles. On the other hand, Machine Learning caught the attention of several researchers in order to solve these security issues, due to it reliability to detect several types of attacks. In this respect, our paper proposes a LightGBM algorithm to detect sybil nodes in VANETs using the position verification approach. When Using the VeReMi dataset, simulations show that the suggested model has better performance when compared to existing methods.