Energy-Efficient Beamforming Design for RIS-Aided MIMO Downlink Communication With SWIPT

Abstract

This paper investigates an energy/power-efficient simultaneous wireless information and power transfer (SWIPT)-enabled reconfigurable intelligent surface (RIS)-assisted multi-input multi-output (MIMO) communication network, where a base station (BS) communicates with multiple information receivers (IRs) and energy receivers (ERs) simultaneously with the aid of a RIS. We formulate a power minimization problem and jointly optimize the active beamforming matrix at the BS and passive beamforming matrix at the RIS subject to the constraint of the minimum required rate at each IR while ensuring the minimum energy harvesting requirement of ERs. Due to the non-convex nature of the formulated problem, we transform the minimum rate constraint into a simpler form by adopting a mean square error (MSE) based approach. Next, an alternating optimization (AO) method is adopted to transform the main non-convex problem into two simpler sub-problems, for obtaining the optimal active and passive beamforming matrices, respectively. Thereafter, computationally efficient algorithms are proposed to determine the optimal solutions of each sub-problem utilizing optimization tools such as Lagrangian dual decomposition, sub-gradient method, majorization-minimization (MM) and pricing-based approaches. Then, we also propose an AO-based iterative algorithm that obtains the optimal values of both matrices iteratively by repeatedly using the aforementioned sub-optimal solutions. We also discuss the impact of imperfect channel state information (CSI) on the performance of the considered network and demonstrate the effectiveness of the proposed algorithm in providing an efficient beamforming design for the same. In addition to this, we highlight the impact of key system parameters such as the number of reflecting elements, and the minimum rate constraint. Finally, we establish that the use of an RIS in such a network along with the proposed algorithm provides a power saving of approximately 10% −26% compared to the other traditional schemes to achieve the same quality of service.