Dynamic Computation and Radio Resource Allocation in Downlink Green C-RAN Systems

Abstract

Cloud radio access network (C-RAN) is a promising solution for future wireless network due to easy cooperation among remote radio heads (RRHs) and concentrated management in baseband unit (BBU) pool. Physical servers in the BBU provides computation ability in the form of virtual machines(VMs) to manage data of different users. In this paper, we focus on a scenario where each server dynamically provides computation capability according to users'(UE) quality of service (QoS) requirement, which is more practical. By jointly optimizing joint beamforming, RRH selection and dynamic computation allocation under the constraint of limited fronthaul and computation capacity, we minimize the network power consumption (NPC) in both RRHs and BBU pool. The formulated problem is non-convex and difficult to solve in general. We propose a simple two-step iterative algorithm to solve the problem. Firstly, we solve the common wireless optimization problem without considering computation allocation. Then by fixing the transmitting beamforming, we solve the computation resource allocation problem, which is a 0-1 multiple knapsack problem (MKP). Simulation results suggests that our proposed dynamic resource allocation algorithm can improve the system performance and save energy.