Coverage Extension through MA-DQN-based Relaying for V2V Communications

In the realm of vehicular communication, ensuring robust and extensive coverage in high-speed highway environments is paramount. This paper introduces an innovative approach that employs Multi-Agent Reinforcement Learning (MARL) to optimize the selection of relay entities, aiming to enhance V2V communication coverage. Also, we propose a novel resource reservation scheme based on 5G New Radio (NR) Mode 2, tailored for relaying entities to ensure adherence to the latest 3rd Generation Partnership Project (3GPP) protocols. Simulation results reveal that the implementation of our resource reservation strategy markedly boosts the Packet Reception Ratio (PRR), confirming the effectiveness of the proposed method. By juxtaposing scenarios with and without our resource allocation technique, we demonstrate an enhancement in PRR performance, thereby validating the benefits of our approach.