An Energy-Effective Network Deployment Scheme for 5G Cloud Radio Access Networks

Inspired by the idea of green and flexible access networks, the Cloud Radio Access Network (C-RAN) has been proposed by network operators together with infrastructure vendors as one promising 5G network architecture. In C-RAN, the light Remote Radio Heads (RRHs) installed with antennas are densely deployed and connected to the baseband unit (BBU) pool through fibers. Under dense C-RAN architecture with a large number of RRHs, a critical issue is introduced which is how to select appropriate RRHs to adapt to the temporal and spatial data dynamics in order to improve the energy efficiency. In this paper, we propose a novel energy-effective network deployment (EEND) scheme with traffic demand satisfaction. The BBU is empowered with the ability to respond to the varying traffic demand by selecting a certain subset of RRHs. The network deployment problem is decomposed into two sub-optimal problems: RRH-traffic demand node association and active RRH set determination. The first sub-optimal problem is modelled as a multiple-choice multidimensional knapsack problem and solved by Lagrange multipliers. In order to solve the second sub-optimal problem, we deactivate the underutilized RRHs based on sleeping techniques. We adopt Genetic Algorithm (GA) as the comparison scheme and numerical results demonstrate that the proposed scheme outperforms the GA scheme in terms of energy saving.