Intelligent Reflecting Surfaces: Sum-Rate Optimization Based on Statistical Position Information

In this paper, we consider a multi-user multiple-input multiple-output (MIMO) system aided by multiple intelligent reflecting surfaces (IRSs) that are deployed to increase the coverage and, possibly, the rank of the channel. We propose an optimization algorithm to configure the IRSs, which is aimed at maximizing the network sum-rate by exploiting only the statistical characterization of the locations of the mobile users. As a consequence, the proposed approach does not require the estimation of either instantaneous channel state information (CSI) or second-order channel statistics for IRS optimization, thus significantly relaxing (or even avoiding) the need of frequently reconfiguring the IRSs, which constitutes one of the most critical issues in IRS-assisted systems. Numerical results confirm the validity of the proposed approach. It is shown, in particular, that IRS-assisted wireless systems that are optimized based on statistical position information still provide large performance gains as compared to the baseline scenarios in which no IRSs are deployed.

Automated summary

This paper explores a multi-user MIMO system aided by multiple intelligent reflecting surfaces (IRSs) and proposes an optimization algorithm aimed at maximizing network sum-rate based on statistical position information, without the need for frequent reconfiguration. Numerical results confirm the effectiveness of the approach, showing that IRS-assisted wireless systems optimized using statistical position information still provide significant performance gains.