UE Throughput Guaranteed Small Cell ON/OFF Algorithm with Machine Learning

Tae-Yoon Park, Jong-Won Han, and Een-Kee Hong

108 223.223-229

10.1109/JCN.2020.000020

Abstract :Ultra-dense network (UDN) has been considered a promising solution to improve the signal quality of cell edge UEs and enhance the serving coverage. However, a large number of small cells causes severe inter-cell interference and high energy consumption, and efficient small cell management is an important issue. In this paper, a UE throughput guaranteed small cell on/off algorithm in UDN environment is proposed to solve the problem of UE throughput reduction due to small cell off process. The proposed small cell on/off algorithm with machine learning is executed by the following processes: First, the network attributes that affect UE throughput are analyzed. Second, the correlation between UE throughput and network attributes is evaluated through multiple linear regression analysis. Third, through understanding the correlation between UE throughput and network attributes, we determine the proper criteria for small cell on/off process. Simulation results show that the proposed small cell on/off algorithm can improve the total network energy efficiency as well as efficiently ensure sufficient UE throughput. Compare to the conventional algorithm, the proposed algorithm shows more than 75% improvements of average network energy efficiency. whole network. 

Index terms :Cell densification, energy efficiency, machine learning, small cell ON/OFF, throughput, UDN.