User fairness-enhanced beamforming for IRS-assisted cognitive radio networks based on prior judgment mechanism

Abstract

This research proposes a novel approach, namely the user fairness-enhanced beamforming (UFEB) scheme, to improve the user fairness in intelligent reflecting surface (IRS) assisted multiuser cognitive radio networks (CRNs). The proposed scheme takes into account the bounded channel state information (CSI) error of the primary user (PU) related channel and establishes a sum α-fair utility maximization problem, where α denotes the fairness adjustment index. This problem optimizes the system for fairness and rate, subject to constraints, such as the quality-of-service requirement of secondary users (SUs), limited interference on PUs, maximum transmit power of cognitive base station (CBS), and unit modulus of the IRS. Firstly, we design a prior judgment mechanism that makes preliminary assessments of the optimized requirements of the system in terms of fairness and rate, whose rationality can be proved in the simulation results. Then, with the prior selection of α, by exploiting the successive convex approximation and Taylor series expansion, we present an alternating iterative algorithm to solve the α-fair utility maximization problem. Simulation results show that the prior selection of α can accurately reflect the difference in the optimization requirements for sum rate and fairness of the system after the beamforming of CBS and the phase of IRS are determined, and validate the effectiveness and superiority of our proposed UFEB scheme in improving the fairness among SUs.