Rate-Energy Optimization for Hybrid-Powered Full-Duplex Relays in Cognitive C-NOMA With Impairments

Abstract

This paper investigates the tradeoff between sum-rate and energy efficiency in cognitive cooperative non-orthogonal multiple access (C-NOMA) with spectrum sharing. In C-NOMA networks for this study, we consider a multi-antenna secondary source and full-duplex relays equipped with power splitting (PS)-based radio frequency (RF) and renewable energy harvesting capabilities. Practical aspects such as full-duplex self-interference, non-linear RF energy harvesting (EH), hardware impairments, imperfections in channel state information and successive interference cancellation are incorporated into the model. We formulate two nonconvex mixed-integer nonlinear programming objective functions to maximize sum-rate and energy efficiency of the C-NOMA network, meeting IoT service requirements. These functions consider transmit power, PS factor for RF harvesting, NOMA power coefficients, qualityof- service, and EH constraints. Specifically, we propose the practical cognitive C-NOMA with EH (PCCN-EH) algorithm. It identifies and selects the best channel gain for relays from the multiple antennas at the secondary source. Relying on particle swarm optimization, we optimize transmit powers, PS factor, and power coefficients. We propose a novel relay selection scheme employing the relay optimization circle. Furthermore, we examine the computational complexity of the PCCN-EH algorithm, demonstrating a manageable complexity with efficiency and feasibility for practical deployment. Through extensive simulations, the proposed PCCN-EH algorithm demonstrates significant performance in sum-rate and energy efficiency across various scenarios, showing remarkable results against benchmarks.