Outage Performance with Deep Learning Analysis for UAV-Borne IRS Relaying NOMA Systems with Hardware Impairments.

While intelligent reflecting surfaces (IRSs) and non-orthogonal multiple access (NOMA) techniques have shown great potential to boost the spectral and energy efficiency for future wireless networks, unmanned aerial vehicles (UAVs) are committed for enhancing the wireless connectivity with fast and flexible deployment. In this regard, we study an integration of an IRS in UAV-enabled wireless relaying system using NOMA transmissions. We also count on the impacts of residual hardware impairments (HIs) in user devices and imperfect successive interference cancellation (SIC) in NOMA, which are inevitable in practical system implementation. We analyze the system performance by deriving the closed-form expressions of outage probability (OP) and system throughput over the line-of-sight (LoS) Rician fading channels for the aerial links. We further pursue asymptotic OP analysis to reveal useful insights on the achievable diversity order. Above all, we present a deep neural network (DNN) framework for OP prediction with a short execution time under the dynamic stochastic environment. Our results validate the theoretical proposition and accentuate the performance advantages of the proposed UAV-borne IRS relaying NOMA system.