Joint optimization of hybrid beamforming and reflection coefficients for secure XL-RIS aided SWIPT system

The extremely large-scale array is a key feature of future wireless communication systems, and the reconfigurable intelligent surface (RIS) gradually develops toward extremely large-scale RIS (XL-RIS) to enhance the communication system performance and compensate for severe path loss. This paper focuses on a secure XL-RIS aided simultaneous wireless information and power transfer (SWIPT) system, where the hybrid analog-digital beamforming architecture of the base station (BS) and the near-field cascaded channels of the XL-RIS are considered. To ensure the security of information transmission for information decoding receivers, artificial noise (AN) is introduced into the transmit signals to prevent interception by potential eavesdroppers, i.e., energy harvesting receivers (EHRs). We aim to maximize the sum received power among EHRs via jointly optimizing the hybrid analog-digital precoders at the BS, the reflection-coefficient matrix, and the AN covariance matrix, subject to the transmit power budget constraint, the minimal rate requirement of information decoding receivers (IDRs), the secure transmission constraints of EHRs, and the unit-modulus phase shifts constraints. Since this optimization problem is non-convex and the variables are tightly coupled, a penalty-based block coordination descent (BCD) algorithm is designed to decouple the original problem into multiple sub-problems and then iteratively solve them by capitalizing on the successive convex approximation (SCA) technique. In particular, the baseband digital beamformers and the analog beamformer are obtained with closed-form expressions using the augmented Lagrangian and BCD methods. Simulation results are presented to demonstrate the effectiveness of the proposed algorithm and it achieves substantial performance gain compared to the other benchmarks.