Performance Analysis under IRS-User Association for Distributed IRSs Assisted MISO Systems

TL;DR

This paper develops a theoretical framework for analyzing and optimizing the average SINR in distributed IRS-assisted multi-user MISO systems, demonstrating improved performance through optimized IRS-user association.

Contribution

It introduces a practical SINR analysis method and a successive refinement optimization for IRS-user association in distributed IRS-assisted systems.

Findings

Distributed IRSs outperform centralized IRSs in SINR performance.

Optimized IRS-user association significantly improves system fairness and SINR.

The proposed analysis matches simulation results accurately.

Abstract

Distributed intelligent reflecting surfaces (IRSs) deployed in multi-user wireless communication systems promise improved system performance. However, the signal-to-interference-plus-noise ratio (SINR) analysis and IRSs optimization in such a system become challenging, due to the large number of involved parameters. The system optimization can be simplified if users are associated with IRSs, which in turn focus on serving the associated users. We provide a practical theoretical framework for the average SINR analysis of a distributed IRSs-assisted multi-user MISO system, where IRSs are optimized to serve their associated users. In particular, we derive the average SINR expression under maximum ratio transmission (MRT) precoding at the BS and optimized reflect beamforming configurations at the IRSs. A successive refinement (SR) method is then outlined to optimize the IRS-user association parameters for the formulated max-min SINR problem which motivates user-fairness. Simulations validate the average SINR analysis while confirming the superiority of a distributed IRSs system over a centralized IRS system as well as the gains with optimized IRS-user association as compared to random association.