A Reservoir Computing-Based Distributed Trust Sharing Provisioning for Internet of Vehicle

The ultimate goal of Internet of Vehicles (IoV) is to enhance the safety, intelligence, and efficiency of traffic systems. However, dynamic network topology, high speed, and exposed communication links inevitably pose looming threats to vehicle safety. Thus, establishing trust sharing mechanism between traffic entities is pivotal to safeguarding the safety of IoV. This paper proposes a distributed trust sharing provisioning utilizing Reservoir Computing-based path predicting algorithm. In this paper, Echo State Network and Deep Q-Network are novelly combined to realize trust information sharing in Partially Observable Markov Decision Process (POMDP) scenarios. In addition, in DRL state space design, we consider the current trust assessment results of different clusters in front of the vehicle to maximize the protection of vehicle network security. Finally, extensive simulations are conducted to prove that the proposed scheme has excellent convergence speed, verifying the feasibility and effectiveness.