Exploiting Active Intelligent Reflecting Surface for Wireless Powered Nonorthogonal Multiple Access

Abstract

In this paper, we propose a wireless powered communication network (WPCN) employing active intelligent reflection surface (IRS) and nonorthogonal multiple access (NOMA) to support multiple battery-free devices. We aim to maximize the sum rate by incorporating practical issues such as amplifying power constraints of active IRS and a nonlinear energy harvesting (EH) model. Since jointly optimizing IRS beamforming in wireless power transfer (WPT), IRS beamforming in wireless information transfer (WIT), and time allocation (TA) is computationally challenging, we resort to an alternating optimization (AO) algorithm in a computationally efficient manner. Additionally, we propose a one-shot (OS) optimization approach for shared amplitude control that provides closed-form expressions for IRS beamforming in WPT and WIT to be computed at a low latency. This OS solution can also serve as the initial point of the proposed AO algorithm and is used for throughput analysis in line-of-sight channels. It is shown that the proposed AO algorithm reduces the complexity of the benchmark and our proposed OS approach, despite its low computational complexity, provides a performance comparable to that of the AO solution. It is also shown that the amplifying power limit of active IRS elements affects the performance severely while the effect of nonlinear EH is not so prominent. Furthermore, the results from analysis and simulation elucidate favorable system configurations for the active IRS-aided WPCN with NOMA when compared with its passive IRS counterpart.