Abstract
Reconfigurable intelligent surface (RIS) can extend signal coverage and improve spectral efficiency, but conventional RIS still cannot unlock the potential in improving the wireless communication performance due to the fixed positions of their reflecting elements. In this letter, we propose a new movable-element RIS (ME-RIS) architecture and investigate its performance optimization in the downlink transmission from a multi-antenna base station (BS) to a single-antenna user. In particular, we aim to jointly optimize the ME-RIS's beamforming and element positions to maximize the achievable rate at the user under the maximum-ratio transmission at the BS. Unlike most existing works relying on gradient-based algorithms for element/antenna position optimization, a lower-complexity hierarchical quantum-behaved particle swarm optimization (QPSO) algorithm is proposed to jointly determine the ME-RIS's beamforming and element positions. Simulation results show that the ME-RIS yields considerably better achievable rate performance than the conventional fixed-element RIS even with reflection amplitude loss, thanks to the available degrees of freedom provided by element movement.