Age-of-Information Based Power Control with Reinforcement Learning in Vehicular Networks

Vehicular networks are more and more popular all over the world owing to their provision of safer and smarter traffic. In intelligent traffic, the information received by vehicles needs to be fresh because of the importance for traffic safety and passenger experience. In this paper, we construct the system where the vehicles acquire information from the roadside units (RSUs) connected with storage servers. The age of information (AoI) is adopted to quantify the freshness of the information. In order to minimize the total AoI of the system, we propose a power management scheme to allocate the power of RSUs. The multi-agent twin delayed deep deterministic policy gradient (MATD3) is deployed to solve the problem. The simulation results show that the proposed approach has superior performance over the deep deterministic policy gradient (DDPG) and greedy algorithms.