Collaboration-based Misbehavior Detection System in Hybrid Vehicular Ad Hoc Networks

Abstract

Hybrid vehicular ad hoc network (VANET) will become one of the most important dynamic communication infrastructures for human’s daily life in the next decade. A hybrid VANET is composed of mobile vehicles and static road-side units (RSU) in which information is distributed among them by transmitting messages through shared wireless communication channels. Useful applications for entertainment, banking, traffic safety, etc. can be deployed onto the hybrid VANET infrastructure. As wireless technology is vulnerable to various malicious attacks, hybrid VANET also suffers from similar security threats, especially the threats from malicious vehicles. In this paper, a misbehavior detection system called MEDIA is developed to mitigate the effect of message randomdropping attack, which is a stereotype of security attack and very easy to launch by a malicious vehicle. Based on collaboration concept, the proposed system can be installed in vehicles and RSUs to monitor vehicles, detect malicious ones and isolate these vicious vehicles. The simulation results show that the MEDIA system can isolate malicious vehicles effectively, increase the data packet delivery ratio and reduce the number of control packets sent even up to 40% of vehicles are malicious in a hybrid VANET.