Robust Resource Allocation Design for Energy-Efficient Active IRS-Aided C-RSMA Systems

This paper investigates robust resource allocation design for active intelligent reflecting surface (IRS)-aided cognitive rate-splitting multiple access (C-RSMA) systems. In particular, an active IRS is deployed to shape a favorable wireless communication environment for enhancing the system performance. We aim to maximize the system energy efficiency by jointly optimizing the common rate allocations for the users, the transmit beamforming vectors at the coordinated base stations, and the active beamforming matrix at the IRS. We formulate the design as a non-convex optimization problem taking into account the discrete nature of the IRS elements and the transmit power budget constraints of the base stations as well as the active IRS. To tackle the non-convex design problem, a computationally effective iterative suboptimal algorithm is proposed by exploiting the block coordinate descent method, the generalized S-Procedure, the successive convex approximation, and the Dinkelbach’s approach. Simulation results reveal a non-trivial tradeoff between the system energy efficiency and the number of the IRS elements. Moreover, our results unveil that active IRS elements equipped with limited bit-resolution of discrete amplifiers and phase shifters is sufficient to achieve a significant gain in system energy efficiency.