XAI Integrated Misbehavior Detection in VANETs

This document is a research article that discusses a machine learning-based approach for detecting position falsification attacks in Vehicular Ad-Hoc Networks (VANETs). It outlines the critical security concerns associated with VANETs, particularly the integrity of data exchanged among vehicles. The authors propose a novel detection system that integrates Explainable Artificial Intelligence (XAI) principles to enhance the interpretability of the model's outcomes. The paper details the architecture of VANETs, the types of applications they support, and the vulnerabilities they face, including insider attacks. It emphasizes the limitations of traditional cryptographic techniques in addressing these vulnerabilities and highlights the necessity for advanced detection methods. The proposed approach aims to provide transparency in the decision-making processes of machine learning models, thereby fostering trust and facilitating validation by human experts. The document also reviews existing literature on machine learning techniques for misbehavior detection and the integration of XAI for improved understanding of model predictions.