Performance Analysis and Resource Allocation of STAR-RIS-Aided Wireless-Powered NOMA System

This paper proposes a simultaneous transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) aided wireless-powered non-orthogonal multiple access (NOMA) system, which includes an access point (AP), a STAR-RIS, and two non-orthogonal users located at both sides of the STAR-RIS. In this system, the users first harvest the radio-frequency energy from the AP in the downlink, then adopt the harvested energy to transmit information to the AP in the uplink concurrently. Two policies are considered for the proposed system. The first one assumes that the time-switching protocol is used in the downlink while the energy-splitting protocol is adopted in the uplink, which is referred to as time-switching and energy-splitting policy (TEP). The second one assumes that the energy-splitting protocol is utilized in both the downlink and uplink, which is referred to as double energy-splitting policy (EEP). The outage probability, sum throughput, and average age of information (AoI) of the proposed system with TEP and EEP are investigated over Nakagami-
$m$
fading channels. In addition, we also analyze the outage probability, sum throughput, and average AoI of the STAR-RIS aided wireless-powered time-division-multiple-access (TDMA) system. Simulation and numerical results demonstrate that the proposed system with TEP and EEP yields better performance than baseline schemes. Although the proposed system sacrifices the outage probability and average AoI performance compared to STAR-RIS aided wireless-powered TDMA systems, it significantly enhances the sum throughput. Furthermore, we design a genetic-algorithm based time allocation and power allocation (GA-TAPA) algorithm to maximize the sum throughput and ensure a certain average AoI. Simulation results indicate that the proposed GA-TAPA algorithm can further improve the sum throughput of the proposed system.