Multi-Objective Resource Allocation in Density-Aware Design of C-RAN in 5G

In this paper, a multi-objective resource allocation algorithm in a novel density-aware design of virtualized software-defined cloud radio access network (C-RAN) is proposed. We consider two design modes based on the average density of users: 1) high-density mode when a large number of low-cost remote radio heads (RRHs) without baseband processing capability are controlled by one single base station and 2) low-density mode when a small number of RRHs with baseband processing capability are deployed. In high-density mode, the challenge of front-haul capacity limitation is tackled via separating control plane and data plane in a heterogeneous structure. Besides, the fully centralized processing and management, and energy-efficient use of infrastructure in low traffic time by turning off RRHs are achieved. In the low-density mode, the transmission delay due to the large distance between the sparse RRHs and cloud unit, is more critical. This practical issue is handled by sharing the baseband processing and resource management among these units in a hierarchical structure. This resulting heterogeneous /hierarchical virtualized software-defined cloud-RAN (HVSD-CRAN) offers various tradeoffs in resource management objectives such as throughput and delay versus power and cost. Consequently, we resort to multi-objective optimization theory to propose a resource allocation framework in HVSD-CRAN.