Static IRS Meets Distributed MIMO: A New Architecture for Dynamic Beamforming

TL;DR

This work tackles the overhead of dynamic IRS beamforming by proposing a D-MIMO empowered static IRS architecture that uses a single fixed pattern ${\Theta}$ and AP-subarea associations to maximize the worst-case received power over a target area. The approach decouples AP-subarea assignment from IRS beamforming, enabling offline optimization based on statistical CSI, and leverages MRT for transmit beams in conjunction with a beam-flattening based IRS design; theoretical results show the passive beamforming gain scales as ${|\chi^w|^2} \approx 4N^2/\pi^2$ when enough APs are deployed ($J \ge J_s$), with a worst-case performance gap to DIBF bounded by ${4/\pi^2}$ (≈3.9 dB) as $N\to\infty$. The analysis is complemented by simulations demonstrating substantial gains over static IRS setups, a clear reduction in angular deviation due to distributed APs, and practical viability for off-line deployment without real-time CSI or IRS reconfiguration. Overall, the paper provides a rigorous framework for achieving DIBF-like coverage using static IRS via D-MIMO, reducing overhead while preserving near-optimal performance.

Abstract

Intelligent reflecting surface (IRS) has been considered as a revolutionary technology to enhance the wireless communication performance. To cater for multiple mobile users, adjusting IRS beamforming patterns over time, i.e., dynamic IRS beamforming (DIBF), is generally needed for achieving satisfactory performance, which results in high controlling power consumption and overhead. To avoid such cost, we propose a new architecture based on the static regulated IRS for wireless coverage enhancement, where the principle of distributed multiple-input multiple-output (D-MIMO) is integrated into the system to exploite the diversity of spatial directions provided by multiple access points (APs). For this new D-MIMO empowered static IRS architecture, the total target area is partitioned into several subareas and each subarea is served by an assigned AP. We consider to maximize the worst-case received power over all locations in the target area by jointly optimizing a single set of IRS beamforming pattern and AP-subarea association. Then, a two-step algorithm is proposed to obtain its high-quality solution. Theoretical analysis unveils that the fundamental squared power gain can still be achieved over all locations in the target area. The performance gap relative to the DIBF scheme is also analytically quantified. Numerical results validate our theoretical findings and demonstrate the effectiveness of our proposed design over benchmark schemes.

Figures (3)

• Figure 1: D-MIMO empowered static IRS for coverage enhancement.
• Figure 2: Worst-case IRS passive beamforming gain versus $J$.
• Figure 3: Worst-case SNR versus Rician factor with noise power $- 90$ dBm .

Theorems & Definitions (5)

• Proposition 1
• Proposition 2
• Proposition 3
• Theorem 1
• Corollary 1