Computational trust model for Internet of Vehicles

In this paper, we propose trust based estimation of the future behavior of the vehicles in IoV (Internet of Vehicles). In an IoV, vehicles can communicate among themselves directly and interact through internet. This allows a vehicle to get past information of other vehicles and obtain trust values of its neighbors from the Internet. An iterative trust computation mechanism is developed when the past trust values of a vehicle available in the network, direct interaction with the vehicles and feedback from the neighbors are used for trust computation. This mechanism employs Dempster Shafer Theorem (DST) to imbibe uncertainty and lack of sufficient data about a vehicle for quick trust update. The model considers four different trust levels to characterize vehicle's behavior. Results indicate that the mechanism is scalable and can be used in IoV environment even when vehicle exhibits changing behavior in presence of small number of vehicles in the neighborhood.