Spatial Multiplexing Oriented Channel Reconfiguration in Multi-IRS Aided MIMO Systems

TL;DR

This paper proposes a novel channel reconfiguration strategy using multiple IRSs to enhance spatial multiplexing in MIMO systems, addressing IRS placement and element allocation for improved spectral efficiency.

Contribution

It introduces an orthogonal placement strategy and a joint optimization approach for IRS elements and power allocation, with theoretical analysis of asymptotic optimality.

Findings

Multi-IRS systems outperform single IRS when IRS elements or power are sufficiently large.

Orthogonal placement reduces channel correlation, enabling interference-free multi-stream transmission.

Equal IRS elements and power allocation become optimal as IRS size and power grow large.

Abstract

Spatial multiplexing plays a significant role in improving the capacity of multiple-input multiple-output (MIMO) communication systems. To improve the spectral efficiency (SE) of a point-to-point MIMO system, we exploit the channel reconfiguration capabilities provided by multiple intelligent reflecting surfaces (IRSs) to enhance the spatial multiplexing. Unlike most existing works, we address both the issues of the IRSs placement and elements allocation. To this end, we first introduce an orthogonal placement strategy to mitigate channel correlation, thereby enabling interference-free multi-stream transmission. Subsequently, we propose a successive convex approximation (SCA)-based approach to jointly optimize the IRS elements and power allocation. Our theoretical analysis unveils that equal IRS elements/power allocation scheme becomes asymptotically optimal as the number of IRS elements and transmit power tend to be infinite. Numerical results demonstrate that when the total number of IRS elements or the power exceeds a certain threshold, a multi-IRS assisted system outperforms a single IRS configuration.