Sum rate maximization for RIS-assisted NOMA-enabled underlay AmBC-CR networks

Abstract

Ambient backscatter communications (AmBC) have introduced a revolutionary approach to wireless communication that utilizes ambient radio frequency (RF) signals in contrast to self-generated RF signals. This makes AmBC a compelling candidate for future Internet-of-Things (IoT) applications, particularly considering the constraints of energy and spectrum resources. In this work, we investigate a cognitive radio (CR) based underlay network of backscatter devices (BDs) that communicate with the backscatter receiver (BR) by utilizing non-orthogonal multiple access (NOMA). Due to the dual fading effect in backscatter networks, the backscatter link is weakened which degrades the achievable performance. In order to tackle this issue, reconfigurable intelligent surface (RIS) is exploited in such networks. Therefore, to maximize the sum rate of the RIS-assisted NOMA-enabled underlay AmBC-CR (RIS-NUAC) network, subject to individual BDs’ data rate constraint and primary receiver (PR) interference threshold constraint, we jointly optimize the phase shift matrix (PSM) at the RIS and the power reflection coefficients (PRCs) at the BDs. To solve the resultant non-convex optimization problem, we propose a sub-optimal solution based on an alternating optimization (AO) algorithm by applying semidefinite relaxation (SDR) and rank one approximation. Based on simulation results, the proposed scheme demonstrates superior performance over baseline schemes, including NOMA with random phases, NOMA without RIS, orthogonal multiple access (OMA) with optimal phases, and OMA without RIS.