Computational Intelligence Inspired Data Delivery for Vehicle-to-Roadside Communications

We propose a vehicle-to-roadside communication protocol based on distributed clustering where a coalitional game approach is used to stimulate the vehicles to join a cluster, and a fuzzy logic algorithm is employed to generate stable clusters by considering multiple metrics of vehicle velocity, moving pattern, and signal qualities between vehicles. A reinforcement learning algorithm with game theory based reward allocation is employed to guide each vehicle to select the route that can maximize the whole network performance. The protocol is integrated with a multi-hop data delivery virtualization scheme that works on the top of the transport layer and provides high performance for multi-hop end-to-end data transmissions. We conduct realistic computer simulations to show the performance advantage of the protocol over other approaches.