Joint Baseband and Radio Resource Allocation for 5G Network Slicing in H-CRANs

Abstract

Network slicing is one of the key technologies of the fifth generation (5G) mobile communications for supporting various use cases with diverse quality of service (QoS) require-ments. As each network slice is tailored to meet a specific set of QoS requirements, intelligent resource allocation becomes vital for network slicing. However, intelligent resource allocation that jointly considers QoS and energy efficiency for network slicing has not been well-studied. Therefore, in this paper, under the architecture of heterogeneous cloud radio access network (H-CRAN), we propose an intelligent, multi-objective resource allocation scheme that takes into account the aforementioned performance aspects for 5G network slicing. This resource allocation scheme encompasses allocation of baseband unit (BBU) pool resources among remote radio heads (RRHs), and resource block (RB) allocation among user equipment (UEs) for all RRHs. In particular, the BBU resource allocation and the RB allocation are performed using a greedy algorithm and deep reinforcement learning algorithm respectively. Simulation results show that the proposed scheme can achieve better QoS provisioning and comparable energy efficiency as compared with the baseline scheme.