On Performance of IRS-Assisted Downlink NOMA Systems With Imperfect SIC

Intelligent reflecting surfaces (IRS) are emerging as a promising technology for forthcoming 6G networks, offering a novel paradigm for regulating wireless channels. The IRS can change the wireless propagation environment, considerably enhancing wireless communication performance. However, the non-orthogonal multiple access (NOMA) technique offers significant spectral efficiency and massive connectivity. The power domain variation NOMA employs superposition-coded symbols with varying powers for multiple-user symbols. This paper investigates an IRS-assisted downlink NOMA system that combines the IRS’s and NOMA’s inherent benefits to increase system performance. To use the inherent advantages of green communication, the base station in such a system uses passive reflectors to connect with multiple users. Furthermore, to gain analytical insights, exact and asymptotic closed-form expressions of the average bit error probability (ABEP) are derived, accounting for error propagation due to successive interference cancellation errors. The ABEP results are simulated for different modulation pairs of NOMA end users over multiple IRS elements. The close agreement between the Monte Carlo simulations with the exact and asymptotic closed-form expressions verifies the accuracy of the analysis. Meanwhile, increasing the reflecting elements and modulation pair improves the system’s performance, making it more suitable for various low-power internet-of-things devices.