Energy Minimization via BS Selection and Beamforming for Cloud-RAN under Finite Fronthaul Capacity Constraints

We consider a downlink cloud radio access network (C-RAN) in which all the base stations (BSs) are connected to a central processor (CP) via digital fronthaul links with finite capacities. The problem of joint base station selection and beamforming design to coordinate interference for power minimization is studied. We formulate the problem of minimizing the network power, which includes the transport link power consumption, as an l(0) minimization problem for the network beamforming design. By assuming that the fronthaul capacity is finite, the fronthaul capacity constraints and the objective function of the network power both include l(0) norms, which are non-convex. To solve the l(0) minimization problem, we utilize the smooth function method in which the Majorization-Minimization (MM) algorithm iteratively solves a difference of convex (DC) problem after approximating the l(0) norm in the objective with a logarithmic function. We then utilize the reweighted l(1) norm technique to reformulate the per-BS fronthaul constraints into convex reweighted l(1) forms and solve the power minimization problem through an iterative algorithm.