A Clustering-Based Energy Saving Scheme for Dense Small Cell Networks

Abstract

The network traffic of mobile wireless communication grows exponentially in recent years. In order to serve intense traffic areas, such as train stations, airports, or hot-spot buildings, the investigation on 5G networks for the next-generation mobile telecommunications has been triggered. One promising solution is the dense small cell networks which consist of a large number of small cells. Base stations (BSs) are one of the major contributors to the power consumption in cellular networks. To address the power consumption issue, the BSON/OFF control mechanism is proposed by the 3GPP. Some previous solutions for BSON/OFF control try to save energy by switching off as many BSs as possible. In this paper, we aim to minimize the total power consumption of a dense small cell network while offering QoS guarantees to all its user equipment (UEs). The BSON/OFF control for minimal total power consumption of an entire network is an NP-hard problem. To this problem, we propose a clustering-based scheme for BSON/OFF control which in the past was applied to mitigate interference. The proposed scheme is a three-phase approach, consisting of BS clustering algorithm, intra-cluster BSON/OFF control algorithm, and inter-cluster BSON/OFF control algorithm. By the clustering technique, we reduce an NP-hard problem to a number of manageable sub-problems. For the possibility of further power saving, we also consider the ON/OFF control of BSs residing on the boundaries of clusters. The time complexity is analyzed. Finally, experimental results reveal that the proposed scheme outperforms existing approaches in power saving and is capable of retaining QoS guarantees for all UEs.