Optimization of RIS-aided MIMO Systems via the Cutoff Rate

TL;DR

This paper proposes using the cutoff rate as a tractable proxy for mutual information in RIS-aided MIMO systems and develops two optimization methods to maximize it, leading to improved system performance.

Contribution

It introduces two novel optimization algorithms based on PGM and SCA to maximize the cutoff rate in RIS-aided MIMO systems, addressing the challenge of MI optimization.

Findings

Significant enhancement of cutoff rate achieved.

Improved mutual information performance demonstrated.

Optimization methods outperform baseline approaches.

Abstract

The main difficulty concerning optimizing the mutual information (MI) in reconfigurable intelligent surface (RIS)-aided communication systems with discrete signaling is the inability to formulate this optimization problem in an analytically tractable manner. Therefore, we propose to use the cutoff rate (CR) as a more tractable metric for optimizing the MI and introduce two optimization methods to maximize the CR, assuming perfect knowledge of the channel state information (CSI). The first method is based on the projected gradient method (PGM), while the second method is derived from the principles of successive convex approximation (SCA). Simulation results show that the proposed optimization methods significantly enhance the CR and the corresponding MI.