Practical evaluation of on-demand smallcell ON/OFF based on traffic model for 5G cellular networks

Abstract

5G communication networks will bring enhanced mobile broadband services to users and vertical markets supporting very wide range requirements from context-dependent applications. To support such applications in 5G, not only the conventional area traffic capacity but network energy efficiency is also a critical factor since energy consumption of information processing is also becoming an economic issue for operators. Dealing with this problem, our project considers a C-RAN based cloud cooperated HetNet architecture which enables global resource optimization among smallcells to maximize objective functions of interest e.g. energy efficiency. On the other hand, a dynamic traffic model based on that the network can dynamically adapt to the variation in a cost-effective way is also crucial for the design of 5G. This paper develops such traffic model based on realistic measurement data in metropolitan Tokyo. Based on the developed model, a on-demand cell activation / deactivation jointly with user association mechanism is applied to maximize the network's energy efficiency defined as the system rate over the total consumed energy. Numerical results show the effectiveness of the proposed algorithm against dynamic variation of hourly traffic while improving users' satisfaction as compared to conventional homogeneous network. The paper also confirms the superiority of mm-wave smallcell HetNet against conventional microwave HetNet in terms of energy efficiency in bps/W.