Development of an Attack Detection Module for Vehicular Ad-hoc Networks

Vehicular ad-hoc networks (VANETs) are rapidly expanding and providing benefits such as smarter traffic control, improved road safety, and increased passenger convenience. However, security remains a significant concern due to the rise in malicious activities aiming to compromise data security and endanger vehicle and passenger safety. The article reports the development of a classification module for VANET Intrusion Detection Systems (IDSs), which continuously monitors and analyzes data traffic to identify patterns indicative of malicious behavior within the network. In course of this research seven different classifiers were trained using the Basic Safety Message subset of the VerReMi dataset. The performance of these classifiers was comprehensively analyzed based on four metrics. Eventually, the kNN, and Random Forest models ensured the best average performance in the case of both investigated message types.