Deep Reinforcement Learning Based Intelligent Reflecting Surface Optimization for TDD MultiUser MIMO Systems

TL;DR

This paper proposes a deep reinforcement learning approach using PPO-GRU to optimize IRS phase shifts in TDD multi-user MIMO systems, enhancing data rates without requiring detailed CSI.

Contribution

It introduces a novel DRL-based method with PPO-GRU for IRS optimization, improving performance and stability over existing benchmarks.

Findings

PPO-GRU outperforms benchmarks in data rate and convergence speed.

The method reduces reliance on detailed channel state information.

Simulation results validate the effectiveness of the proposed approach.

Abstract

In this letter, we investigate the discrete phase shift design of the intelligent reflecting surface (IRS) in a time division duplexing (TDD) multi-user multiple input multiple output (MIMO) system.We modify the design of deep reinforcement learning (DRL) scheme so that we can maximizing the average downlink data transmission rate free from the sub-channel channel state information (CSI). Based on the characteristics of the model, we modify the proximal policy optimization (PPO) algorithm and integrate gated recurrent unit (GRU) to tackle the non-convex optimization problem. Simulation results show that the performance of the proposed PPO-GRU surpasses the benchmarks in terms of performance, convergence speed, and training stability.