Secure Transmission Optimization for IRS-Aided WPCNs With Linear EH Model

Abstract

In this paper, we present a secure transmission optimization scheme for intelligent reflecting surface (IRS) aided wireless powered communication networks (WPCNs) in the presence of eavesdropping in the uplink, where the linear energy harvesting (EH) model and imperfect channel state information (CSI) are considered. We aim to maximize the secrecy rate at the hybrid access point (HAP) by jointly optimizing the transmission time allocation factor, HAP transmit beamforming vector, IRS energy reflection coefficients, and IRS information reflection coefficients, which is formulated as a joint multivariate non-convex optimization problem. By using the alternatively iterative optimization method and after several manipulations on the subproblems, the optimal solutions of time allocation factor and HAP transmit beamforming vector are obtained with standard convex optimization tools, while the sub-optimal solutions of IRS energy and information reflection coefficients are obtained with semidefinite relaxation (SDR) and Gaussian randomization procedure. Numerical results show that the proposed algorithm achieves substantial performance enhancement while maintaining high robustness, and significantly outperforms the benchmark schemes under various CSI error variance. It is also shown that optimizing IRS phase shifts or increasing the number of IRS reflection elements is better than optimizing HAP transmit beamforming or increasing the number of HAP antennas, respectively.