Beamforming Design and RRH Selection for Energy-Efficient C-RAN with ISAC

Abstract

In this work, we investigate an energy-efficient design in Cloud Radio Access Network (C-RAN) system with integrated sensing and communication (ISAC), where geographically distributed remote radio heads (RRHs) are coordinated by a central controller to simultaneously support communication users and sensing targets. Although activating all RRHs in a C-RAN system can significantly improve both communication and sensing performance by leveraging spatial degrees of freedom, the resulting network densification inevitably increases transport energy consumption and fronthaul overhead. To address this issue, we jointly design RRH mode selection and transmit beamformingto achieve high energy efficiency while satisfying the requirements of both communication users and sensing target detection. Instead of using discrete variables to represent the operation mode of RRHs and introducing a complicated mixed-integer nonlinear programming problem, we introduce working mode sets of RRHs and decouple the original problem into two components: beamforming design for a fixed RRH set and active RRH set selection. Based on this reformulation, we develop a majorization-minimization (MM)-based algorithm that exploits the group sparsity of the beamforming vectors together with a geometry-aware RRH ordering scheme to achieve efficient RRH selection. Numerical results demonstrate that our proposed method outperforms the baseline schemes. In specific, our proposed method can achieve more than 15% reduction in system power consumption compared with the fully active RRH scheme.