Hybrid Beamforming with Discrete Phase Shifts for RIS-assisted Multiuser SWIPT System

Reconfigurable Intelligent Surfaces (RIS) provides an emerging affordable solution by leveraging massive low-cost reconfigurable reflective array elements. In this letter, by utilizing low-cost passive reflecting elements, a joint hybrid beamforming with discrete phase shifts design is studied in RIS-assisted multiuser simultaneous wireless information and power transfer (SWIPT) systems. However, the formulated sum rate maximization problem is nonconvex and very challenging to solve. To overcome this difficulty, we decouple the original maximization problem into the digital-, phase shifts and analog subproblem. Specifically, for the nonconvexity of the digital subproblem, we perform the first order Taylor series expansion on the objective function, and then an iterative strategy is developed to tackle the difficulty of the maximum transmit power constraint imposed by discrete optimization variables. Subsequently, an effective convex optimization technique was employed to optimize reflecting elements of RIS. Finally, the greedy selection method based on minimum mean squared error (MMSE) scheme is proposed to solve the analog subproblem. Numerical simulations are conducted to demonstrate the performance advantages of the proposed scheme.