Robust transmission for multi-user OFDM-based IRS-assisted cognitive radio networks

Abstract

In cognitive radio networks (CRNs) employing underlay multi-user scenarios, intelligent reflecting surface (IRS) plays a crucial function in augmenting total transmission rate for cognitive radio (CR) users, while adhering to restrictions caused by the interference from secondary users (SUs) to primary users (PUs). However, due to the passive characteristics of IRS, it operates in a non-cooperative relationship with both PU and SU network, resulting in delayed information acquisition and channel errors. Therefore, in this research, considering channel uncertainties, limited interference, and interference among SUs, we put forward a robust power and subcarrier allocation strategy for orthogonal frequency division multiplexing (OFDM) with IRS-assisted CRNs in a multi-user scenario. Formulated as a resource allocation (RA) problem in an underlay spectrum sharing mode for multi-user CR, the joint optimization problem encompasses subcarrier allocation and power allocation for SUs, a phase shift matrix for IRS, and minimum rate constraints. We transform the problem using worst-case criterion into a deterministic convex optimization problem to handle the constraints of bounded channel uncertainties. By utilizing matching theory, variable substitution, scaling methods, convex-concave programming (CCP), and alternating optimization techniques, the converted non-convex problem is transformed into a convex one. Simulation results showcase that the proposed algorithm achieves significant gains in transmission rate and exhibits robustness.