Interference mitigation with block diagonalization for IRS-aided MU-MIMO communications

TL;DR

This paper proposes two block diagonalization-based precoding methods for IRS-aided MU-MIMO networks, demonstrating that accounting for IRS interference can enhance system capacity and spectral efficiency.

Contribution

It introduces novel precoder design techniques that consider IRS-induced interference, improving interference mitigation in MU-MIMO systems.

Findings

Treating IRS interference can improve spectral efficiency.

Ignoring IRS interference is sometimes sufficient.

Proposed methods outperform traditional approaches in certain scenarios.

Abstract

This work investigates interference mitigation techniques in multi-user multiple input multiple output (MU-MIMO) Intelligent Reflecting Surface (IRS)-aided networks, focusing on the base station end. Two methods of precoder design based on block diagonalization are proposed. The first method does not consider the interference caused by the IRS, seeking to mitigate only the multi-user interference. The second method mitigates both the IRS-caused interference and the multi-user interference. A comparison between both methods within an no-IRS MU-MIMO network with strong direct links is provided. The results show that, although in some circumstances IRS interference can be neglected, treating it can improve system capacity and provide higher spectral efficiency