Machine Learning for Misbehavior Detection in Intelligent Transportation Systems Using BSM Data

Vehicular networks focus on improving traffic and safety concerns in an Intelligent Transportation System (ITS). However, the presence of misbehaving nodes can drastically impact the safety, traffic and other applications offered by ITS. Misbehavior detection forms an important mechanism in an ITS to detect and mitigate the effect of misbehaving nodes. The proposed work focuses on analyzing the performance of popular machine learning algorithms for misbehavior detection in VANETs. The effect on performance pre and post-regulation is also highlighted. VEREMI dataset is used as the dataset of choice for the analysis