Performance Analysis of IRS-Assisted NOMA Networks With Randomly Deployed Users

Abstract

In this article, we investigate the intelligent reflecting surface (IRS)-assisted downlink nonorthogonal multiple access (NOMA) networks, where the users are randomly deployed in a disk. The randomly deployed users are divided into two groups, named center group and edge group, and then two distance-dependent user selection schemes are proposed. The developed closed-form expressions of outage probability and average rate for the proposed two user selection schemes in IRS-assisted NOMA networks are derived, as well as the asymptotic results at high signal-to-noise ratio (SNR) and the number of the reflecting number of the IRS trends to infinity, respectively. The asymptotic analysis results of outage probability for cell-edge user decay as $\frac{\ln \mathrm{SNR}}{\mathrm{SNR} ^{K+1}}$ , and the achieved diversity gain is $K+1$ , at high SNR region, where $K$ denotes the number of elements of the IRS. Furthermore, as $K \rightarrow \infty$ , the asymptotic average rate of cell-edge user only depends on the power allocation factor of cell-center user. In addition, compared to IRS-assisted orthogonal multiple access networks, the IRS-assisted NOMA networks can always realize lower outage performance and higher average rate. Finally, Monte–Carlo simulation results are provided to verify the accuracy of the developed analytical results for the proposed two user selection schemes.