Joint Resource Optimization for Secure Cooperative Perception in Vehicular Networks

In the realm of autonomous driving, cooperative perception serves as a crucial technology for mitigating the inherent constraints of individual vehicle’s perception. To enable cooperative perception, vehicle-to-vehicle (V2V) communication plays an indispensable role. Unfortunately, owing to weak virus protection in V2V networks, the emergence and widespread adoption of V2V communications have also created fertile soil for the breeding and rapid spreading of worms. To stimulate vehicles to participate in cooperative perception while blocking the spreading of worms through V2V communications, we design an incentive mechanism, in which the utility of each sensory data requester and that of each sensory data provider are defined, respectively, to maximize the total utility of all the vehicles. To deal with the highly non-convex problem, we propose a pairing and resource allocation (PRA) scheme based on the Stackelberg game theory. Specifically, we decompose the problem into two subproblems. The subproblem of maximizing the utility of the requester is solved via a two-stage iterative algorithm, while the subproblem of maximizing the utility of the provider is addressed using the linear search method. The results demonstrate that our proposed PRA approach addresses the challenges of cooperative perception and worm spreading while efficiently converging to the Stackelberg equilibrium point, jointly maximizing the utilities for both the requester and the provider.