Throughput Maximization Design for RIS-Assisted mmWave WPCN-NOMA Systems

This paper considers an reconfigurable intelligent surface (RIS)-assisted millimeter-wave (mmWave) wireless-powered communication network (WPCN). In the network, multiple energy-constrained devices harvest radio frequency (RF) energy signals transmitted by the power station (PS) with the assistance of the RIS in the downlink (DL), and then transmit the signals to the data sink (DS) in the uplink (UL) employing non-orthogonal multiple access (NOMA) strategy. Under the considered framework, we aim to maximize the sum achievable throughput of all devices by jointly optimizing the device grouping, transmit power allocation of PS, energy beamforming vectors, and phase shift matrix of RIS in the DL transmission, as well as the phase shift matrix of RIS and transmission timeslot allocation for the UL transmission. To solve this non-convex problem with multiple high coupling parameters, the optimization problem is divided into three sub-problems. In the first sub-problem, an algorithm combining K-means++ clustering algorithm and exclusion theory is proposed to obtain optimal device grouping. For the second sub-problem, the transmit power allocation of PS, energy beamforming vectors, and phase shift matrix of RIS in the DL are obtained by alternating optimization (AO) algorithm. Then, we optimize the transmission timeslot allocation and phase shift matrix of RIS in the UL according to a successive convex approximation (SCA)-based 1-D search method in the third sub-problem. Numerical simulations verify the superiority of the proposed optimization scheme in improving the achievable throughput compared to other comparison schemes.