Data-Enabled Learning based Intelligent Resource Allocation for Multi-RIS Assisted Dynamic Wireless Network

Abstract

This paper investigated the optimal dynamic re-source allocation problem for multi mobile reconfigurable intelligent surface (RIS) aided wireless network with uncertain time-varying wireless channels. Recently, RIS has been considered as one of the most promising techniques for enhancing dynamic wireless network quality, e.g. maximizing spectrum efficiency, etc., without increasing power consumption. Comparing with the traditional RIS techniques, the mobility of RIS through the unmanned aerial vehicles(UAV) is stimulated in this paper. Before harvesting the benefits from mobile RIS, a novel resource allocation technique needs to be developed that cannot only optimize the overall network quality, e.g. maximizing energy efficiency, coverage, minimizing power consumption, etc., but also adapt to the uncertainty of the environment, such as time-varying wireless channel, in real time. Hence, a novel online reinforcement learning based optimal resource allocation algorithm has been designed. Firstly, a Q-learning Adaptive Dynamic Programming algorithm is utilized to optimize the deployment of the RIS. Then, an online actor-critic reinforcement learning algorithm is developed along with neural networks (NNs)to learn the optimal transmit power control as well as mobile RIS phase shift control policy. Eventually, numerical simulations have been provided to demonstrate the effectiveness of the developed scheme.