Joint Fronthaul Multicast Beamforming and User-Centric Clustering in Downlink C-RANs

The cloud radio access network (C-RAN) has been deemed a cost-effective architecture for exploiting the capacity benefit of densely deployed radio access points. The low-latency fronthaul data transmission from the central processor to small-cell base stations (SBSs) is a key requirement in C-RANs for which conventional wired fronthaul links will be cost-prohibitive and also inconvenient. Therefore, scalable and low-cost wireless fronthaul solutions have drawn much attention in both industry and academia. In this paper, we propose adopting the multicast beamforming strategy over fronthaul links to deliver each user's message to a cluster of SBSs selected according to the user-centric clustering scheme, which then adopts the joint beamforming technique to cooperatively transmit the signal to the target users. Some approximate techniques are applied to obtain a tractable formulation for this mixed integer nonlinear programming problem, and an iterative algorithm based on the block coordinate update method is proposed accordingly. Then, a binary search based algorithm is developed to preserve the sparsity of beamformers due to the relaxation of the discrete clustering function with the continuous exponential function. Extensive simulation results are provided to show the performance of the proposed algorithms in terms of convergence, power consumption, and weighted sum rate.