Trust Management Mechanism in IoV based on Bayesian classification

The Internet of Vehicles (IoV) significantly enhances the perception of the traffic environment and facilitates information exchange among vehicles, thereby bolstering transportation safety. Nevertheless, the intricate nature of traffic scenes, the swift movement of vehicles and the dynamic nature of the vehicle network's topology present challenges in guaranteeing the reliability of shared messages among vehicles. The intricacy of the situation additionally poses challenges for vehicles in evaluating the reliability of received messages, creating substantial potential for traffic accidents. Therefore, this article proposes an IoV trust management mechanism based on Bayesian classification, which designs a complete vehicle trust value calculation scheme. The verification of messages received from neighboring vehicles is carried out using the Bayesian inference model, which can effectively detect vehicles sending malicious information. After the verification results, rating parameters are produced for every source vehicle of the messages. Ultimately, the trust value of the vehicle is calibrated and penalized based on these rating parameters. The vehicle uploads rating parameters to the RSU, which calculates the trust value of the relevant vehicles. By excluding vehicles with low trust values, the cooperation success rate among vehicles within the network has risen from 65% to approximately 70%.