Deep Reinforcement Learning based Distributed Resource Allocation for V2V Broadcasting

Abstract

In this article, we exploit deep reinforcement learning for joint resource allocation and scheduling in vehicle-to-vehicle (V2V) broadcast communications. Each vehicle, considered as an autonomous agent, makes its decisions to find the messages and spectrum for transmission based on its local observations without requiring or having to wait for global information. From the simulation results, each vehicle can effectively learn how to ensure the stringent latency constraints on V2V links while minimizing the interference to vehicle-to-infrastructure (V2I) links.