Joint Optimization of Channel Allocation and Power Control Algorithm for AoI-Aware Industrial IoT with Deep Reinforcement Learning

Abstract

With the rapid improvement of information freshness and energy efficiency requirements of industrial Internet of Things, this paper proposes a channel resource allocation algorithm and a power control algorithm based on deep reinforcement learning to solve the resource scheduling problem of multi-device and multi-channel age sensitive industrial Internet of things environment. In addition, by analyzing the system environment, a reward function based on the age of data information and the backlog state of the queue is designed, so that the trained reinforcement learning agent can reach the convergence state faster. The simulation results show that the proposed algorithm has better performance in time-varying dynamic industrial Internet of things environment.