Reinforcement Learning Base Energy Efficient Route Optimization Model to Enhance Network Lifetime on the Internet of Vehicles

Abstract

The Internet of Vehicles (IoV) has emerged as a promising paradigm to revolutionize transportation systems by enabling seamless communication and data exchange among vehicles and infrastructure. A crucial challenge in IoV is optimizing the routes of information, which is used to enhance network efficiency, reduce congestion, and enhance network lifetime. This research presents a centralized reinforcement base learning framework that leverages the power of reinforcement learning algorithms to optimize the routes by minimizing communication overhead. To evaluate the proposed approach, extensive simulations are conducted in a realistic IoV environment, incorporating various scenarios and traffic conditions. Comparative analyses are performed against LEACH, PEGASIS and EER_RL traditional widely used route optimization algorithms and heuristic-based methods to assess the effectiveness and efficiency of the reinforcement learning-based approach. Our findings demonstrate that the reinforcement learning-based route optimization approach exhibits superior performance in terms of reducing energy consumption, minimizing communication overhead, and enhancing network lifetime compared to conventional methods. This work opens new avenues for future research in leveraging RL algorithms for optimizing various aspects of IoV systems and addressing the challenges posed by dynamic and complex vehicular networks.