A stochastic geometry based two-stage energy consumption minimization strategy via sleep mode with QoS constraint

With the ever-increasing demand for data traffic, high dense cellular networks are gaining extensive attention. But at the same time dense deployment of small cells results in high energy consumption and operating cost. Sleep mode for high dense cellular networks is emerging as a promising way to meet green communication. The aim of this paper is to use the two-stage strategy for reducing the energy consumption of high dense deployed cellular networks by turning some base stations (BSs) into sleep mode. The distribution of BSs and users is modeled as Poisson point process. By using the tools of stochastic geometry, we derive the network load and overload possibility of BSs. Based on the derived results, in the first stage, we randomly turn some BSs from the dense networks into sleep mode so as to reduce the density and power consumption of BSs while maintaining the quality of service of users. In the second stage, some of the remaining BSs are further turned into sleep mode by means of the proposed selection algorithm. Our two-stage strategy can significantly simplify the complexity of finding a relatively good result. Numerical results validate the analysis and show that the proposed strategy can significantly reduce energy consumption.