Secure Beamforming Optimization for STAR-RIS-Assisted SWIPT Networks Using Modified Differentiated Creative Search Algorithm

This paper introduces a secure simultaneous wireless information and power transfer (SWIPT) system using simultaneous transmitting and reflecting reconfigurable intelligent surfaces (STAR-RISs). To enhance secrecy and energy harvesting, we propose a Modified Differentiated Creative Search (MDCS) technique to optimize the transmit beamforming, artificial noise, and STAR-RIS transmission/reflection coefficients simultaneously. The proposed paradigm considers imperfect cascaded channel state information (CSI) and integrates a viable non-linear energy harvesting model with limited CSI error models. The simulation results show that the proposed MDCS-based strategy exceeds existing benchmark techniques in terms of secrecy performance, robustness, and energy harvesting efficiency. The findings demonstrate that the proposed MDCS algorithm achieves a mean objective value of 0.1167, outperforming Differentiated Creative Search (DCS), Draco Lizard Optimizer (DLO), mean Differential Evolution algorithm (meanDE), Octopus, and Nutcracker Optimization Algorithm (NOA) by approximately 75.1%, 86.4%, 45.0%, 85.0%, and 74.7%, respectively. Furthermore, MDCS achieves faster convergence and improved solution stability, demonstrating its efficiency for robust beamforming optimization in STAR-RIS-assisted secure SWIPT systems with suboptimal CSI conditions.