Misbehavior Detection based on Multi-Head Deep Learning for V2X Network Security

Misbehavior detection in vehicular networks is an important field of research that focuses on designing and developing mechanisms to detect anomalous behaviors pertaining to vehicle movement, data transmission. Misbehavior detection systems help to protect vehicular networks from insider attacks that are generated within the vehicular network system and avoid dangerous situations on time. In this paper, we propose a deep learning based misbehavior detection architecture in vehicular networks. The proposed architecture consists of a multi-head convolutional neural network (CNN) and long short-term memory (LSTM) network. The simulation results demonstrate that our proposed method is more effective to detect misbehaviors as compared to previous methods in terms of accuracy.