Outage Analysis of IRS-assisted UAV NOMA Downlink Wireless Networks

Yuan Liu,Ke Xiong,Yongdong Zhu,Hong-Chuan Yang,Khaled Ben Letaief,Pingyi Fan
DOI: https://doi.org/10.1109/jiot.2023.3323311
IF: 10.6
2023-01-01
IEEE Internet of Things Journal
Abstract:This paper studies an intelligent reflecting surface (IRS)-assisted unmanned aerial vehicle (UAV) network, where the ground users (GUs) desire to receive information from a UAV. Downlink non-orthogonal multiple access (NOMA) is considered typically with two GUs being selected according to whether a Line of Sight (LoS) link between GUs and UAV exists. As the accurate channel information of LoS or Non-Line-of-Sight (NLoS) links for multiple GUs is difficult to acquire, an approximate LoS region-based method is designed to select GUs as an alternative. In order to enhance the communication quality of the far GU, an IRS is deployed to assist the NLoS transmission. For such a system, we evaluate its outage performance in Nakagami-m fading. First, the central limit theorem (CLT) and Laplace Transform (LT) are employed to derive the channel statistics of the UAV-IRS-user link. Then, asymptotic closed-form expressions of the outage probabilities are derived for the selected GUs based on Gaussian-Chebyshev quadrature approximation. Monte Carlo simulations validate the validness of our derived outage probabilities. It shows that the approximate LoS region-based scheme provides similar outage performance laws as the accurate LoS region-based one. Moreover, the outage probabilities of selected GUs in terms of NOMA-based protocol and orthogonal multiple access (OMA)-based protocol are analyzed. Simulation results confirm that the proposed NOMA-based protocol is capable of achieving superior performance compared with the OMA-based protocol by setting power allocation factor and targeted SINR thresholds of near GU and far GU properly. Specifically, when the rate threshold of near GU is relatively large or the rate threshold of far GU is relatively small, the outage performance derived by NOMA-based protocol performs better than OMA-based protocol in most of cases.
What problem does this paper attempt to address?