Abstract
This paper investigates an intelligent reflecting surface (IRS)-aided movable antenna (MA) system, where multiple IRSs cooperate with a multi-MA base station to extend wireless coverage to multiple designated target areas. The objective is to maximize the worst-case signal-to-noise ratio (SNR) across all locations within these areas through joint optimization of MA positions, IRS reflection coefficients, and transmit beamforming. To achieve this while balancing the performance-cost trade-off, we propose three coverage-enhancement schemes: the area-adaptive MA-IRS scheme, the area-adaptive MA-staIRS scheme, and the shared MA-staIRS scheme, where staIRS denotes static IRSs with reflection coefficients configured only once during installation. These schemes lead to challenging non-convex optimization problems with implicit objective functions, which are difficult to solve optimally. To address these problems, we propose a general algorithmic framework that can be applied to solve each problem efficiently albeit suboptimally. Simulation results demonstrate that: 1) the proposed MA-based schemes consistently outperform their fixed-position antenna (FPA)-based counterparts under both area-adaptive and static IRS configurations, with the area-adaptive MA-IRS scheme achieving the best worst-case SNR performance; 2) as transmit antennas are typically far fewer than IRS elements, the area-adaptive MA-staIRS scheme may underperform the baseline FPA scheme with area-adaptive IRSs in terms of the worst-case SNR, but a modest increase in antenna number can reverse this trend; 3) under a fixed total cost, the optimal MA-to-IRS-element ratio for the worst-case SNR maximization is empirically found to be proportional to the reciprocal of their unit cost ratio.