Extremely Large-Scale Movable Antenna-Enabled Multiuser Communications: Modeling and Optimization

TL;DR

This paper introduces an extremely large-scale movable antenna system for multiuser wireless communication, modeling near-field effects and optimizing antenna placement to significantly improve system performance over fixed antennas.

Contribution

It proposes a novel XL-MA architecture with a spatially non-stationary channel model and develops an efficient placement optimization algorithm for enhanced multiuser communication.

Findings

Optimized XL-MA placement increases user signal power.

Placement reduces multiuser interference.

Achieves near-optimal performance compared to benchmarks.

Abstract

Movable antenna (MA) has been recognized as a promising technology to improve communication performance in future wireless networks such as 6G. To unleash its potential, this paper proposes a novel architecture, namely extremely large-scale MA (XL-MA), which allows flexible antenna/subarray positioning over an extremely large spatial region for effectively enhancing near-field effects and spatial multiplexing performance. In particular, this paper studies an uplink XL-MA-enabled multiuser system, where single-antenna users distributed in a coverage area are served by a base station (BS) equipped with multiple movable subarrays. We begin by presenting a spatially non-stationary channel model to capture the near-field effects, including positiondependent large-scale channel gains and line-of-sight visibility. To evaluate system performance, we further derive a closedform approximation of the expected weighted sum rate under maximum ratio combining (MRC), revealing that optimizing XLMA placement enhances user channel power gain to increase desired signal power and reduces channel correlation to decreases multiuser interference. Building upon this, we formulate an antenna placement optimization problem to maximize the expected weighted sum rate, leveraging statistical channel conditions and user distribution. To efficiently solve this challenging non-linear binary optimization problem, we propose a polynomial-time successive replacement algorithm. Simulation results demonstrate that the proposed XL-MA placement strategy achieves nearoptimal performance, significantly outperforming benchmark schemes based on conventional fixed-position antennas.