The field of vehicular network security is moving towards the development of more robust and efficient intrusion detection systems, with a focus on addressing the unique challenges posed by the Controller Area Network (CAN) and Electronic Control Units (ECUs). Researchers are exploring the use of advanced techniques such as graph neural networks, attention mechanisms, and autoencoders to improve detection accuracy and reduce false positives. Furthermore, there is a growing interest in developing secure communication protocols that can prevent Denial of Service (DoS) attacks and ensure the integrity of cooperative awareness messages. Innovative approaches, such as distributed federated learning and message verification facilitators, are being proposed to enhance the security and accountability of vehicular networks. Noteworthy papers include: Are GNNs Worth the Effort for IoT Botnet Detection, which evaluates the effectiveness of graph neural networks for IoT botnet detection, and Accountable, Scalable and DoS-resilient Secure Vehicular Communication, which proposes a novel approach to prevent DoS attacks in vehicular networks.
Advances in Vehicular Network Security
Sources
Vehicular Intrusion Detection System for Controller Area Network: A Comprehensive Survey and Evaluation
Graph Attention Neural Network for Botnet Detection: Evaluating Autoencoder, VAE and PCA-Based Dimension Reduction
Are GNNs Worth the Effort for IoT Botnet Detection? A Comparative Study of VAE-GNN vs. ViT-MLP and VAE-MLP Approaches
A Joint Reconstruction-Triplet Loss Autoencoder Approach Towards Unseen Attack Detection in IoV Networks
Accountable, Scalable and DoS-resilient Secure Vehicular Communication
Distributed Federated Learning for Vehicular Network Security: Anomaly Detection Benefits and Multi-Domain Attack Threats