MAC Contention Protocol Based on Reinforcement Learning for IoV Communication Environments

Abstract

The Medium Access Control (MAC) layer contention protocol is closely related to the performance of network throughput, end-to-end delay, and access fairness on the Internet of Vehicles (IoV) communication. Based on the MAC layer protocol of the Wireless Access in Vehicular Environments (WAVE) standard system, this paper proposes a MAC layer contention window adaptive adjustment policy using Reinforcement Learning. Through the detection of the number of neighbors and the application of the Q-Learning algorithm, the vehicle can adjust the contention window according to the number of nodes competing for the same channel to adapt to the changing environments of the IoV. Three different MAC protocols are simulated and analyzed under the Vehicle in Network Simulation (Veins) platform. The results show that the proposed MAC protocol based on neighbor detection and Q-Learning performs better than WAVE MAC protocol and general MAC protocol based on Q-Learning.