Semi-Dynamic Clustering for Large-scale Distributed-Antenna System

Large-scale distributed-antenna system (L-DAS) can further reduce the access distance between user equipments (UEs) and remote antenna units (RAUs), which brings a low propagation loss and high multiplexing gain. However, the interference caused by the overlapping coverage areas of distributed RAUs will severely degrade the sum-rate of the system. Although full cooperation transmission and dynamic clusterinng can effectively suppress inter-RAU interference, it is difficult to share the instantaneous channel state information (CSI) for the L-DAS with the limited-capacity backhaul. In this paper, to reduce the backhaul burden and mitigate interference effectively, a semi-dynamic clustering algorithm with bidirectional selection is proposed by using statistical CSI. Simulation results show that the achievable sum-rate of our proposed algorithm overwhelm that of the static clustering one, as well as, is close to that of the dynamic clustering algorithm using instantaneous CSI.