LAUSR

—In order to realize massive connectivity, high spec- tral, and energy efficiency in future internet of things (IoT) networks, reconfigurable intelligent surfaces (RISs) are considered as a promising solution with their low cost and channel-adjusting ability. However, the ever-growing massive number of devices in RIS-empowered IoT networks requires achieving energy self-sustainability for the entire network. Motivated by this, a simultaneous wireless information and power transfer system empowered by a wireless-powered RIS has been investigated under the condition of imperfect channel state information, in which both the RIS and zero-energy devices (ZEDs) harvest energy with the power-splitting (PS) scheme for their operation. Specifically, a non-convex transmit power minimizing problem is formulated to handle the power-limited issue of practical IoT networks, where the PS ratios of the ZEDs, the active beamforming of the access point and RIS reflective coefficients are jointly optimized and solved by an efficient alternating opti- mization framework. The improved performance of the proposed algorithm against the benchmark schemes and its robustness are verified by simulation results, which highlight the effectiveness of the proposed algorithm in practical IoT networks.