Trustworthiness Evaluation of Internet of Vehicles Nodes Based on Real-Time Perception and Long-Term Reputation

In Internet of Vehicles, the way of collecting road environment data and sharing data through vehicle equipment has been widely used. However, due to the impact of vehicle equipment, environment and malicious attacks, different vehicles have different source qualities, and the perceived data from multiple vehicle sources may be different from the real road conditions. Truth discovery is to infer the truth closest to the truth from multiple different or even conflicting information sources by comprehensively considering the quality and reliability assessment of these information sources. Aiming at the quality and reliability assessment of information sources of vehicle nodes, this paper proposes a node credibility assessment method based on real-time perception data and long-term reputation of vehicle sources. Compared with the existing scheme, the reliability of vehicle source nodes is used as the initial weight of truth discovery in intelligent networked vehicles, and the prior knowledge is used to reduce the number of iterations required for truth discovery and improve its accuracy. The percentage of relevant data is calculated according to the factual perception data of vehicle sources. Finally, the reliability of vehicle source nodes is obtained by combining the long-term reputation accumulated by long-term tasks of vehicle sources. Experiments on GPS data sets show that the method of assigning initial weights based on vehicle source node confidence can reduce the number of iterations and have higher accuracy.