Base station energy saving based on dynamic programming combined with cell clustering

With continuous growth of the wireless cellular networks, the network operators take the energy-efficiency into their consideration as a major design goal lately. It also receives a growing number of attentions from engineers and researchers, due to the increasing awareness of environmental and economic issues. In this paper, we develop a theoretical framework for base station (BS) energy saving that encompasses cell clustering and dynamic BS sleeping operation. The combination of these two aspects allows for a more flexible and accessible tradeoff between cellular networks performance and energy consumption. Our clustering strategy based on the geographic information and traffic load is able to degrade the overall scale of this problem. Moreover, the BS sleeping policy could be found quickly by the dynamic programming (DP) algorithm with a state-reduced solution space based on the distances between BSs and the utilizations of BSs. Specifically, we minimize the total cost of the whole network as much as possible with a low complexity of the algorithm and little additional signaling, thus our policy is easy to implement in practice. The simulation results under practical configurations demonstrate that the proposed dynamic algorithms can effectively reduce energy consumption at a low complexity.