基于休眠机理的三维小基站蜂窝网络能效优化

摘要

小基站通常部署在写字楼、商贸区等城市密集区域以弥补传统宏基站在覆盖和传输方面的不足.小基站的分布一般是根据高峰时的网络负荷设计的,这必然导致网络负荷较低时的资源浪费.讨论了在平均接入率和信道容量双重约束下基于休眠机理的三维小基站蜂窝网络的能效优化问题.借助泊松点过程理论推导了三维小基站网络下行信道容量和平均接入率的数学表达式.通过分析下行信道容量和平均接入率的单调性得出同时满足传输信道容量和接入率要求的最佳休眠概率.分析了小基站最大用户连接数的最佳值,通过对该参数的合理配置,可以在满足通信指标的前提下最大程度地降低网络能耗.仿真结果表明,设计的基站休眠机理可以使小基站网络的能耗下降约21%.%Wireless cellular networks all over the world are undergoing a profound transformation evolving from voice-oriented to data networks. Larger coverage area, better service quality, and lower energy cost are the key issues in the deployment of cellular networks. To achieve these goals, small cells, such as the femtocells and picocells, have become an important part of the current 4G and future 5G wireless cellular networks. Generally speaking, small cell networks are deployed according to the peak traffic load, which causes energy waste during low traffic periods. Against this background, energy efficiency optimization has become one of the research hotspots in wireless communications. In this paper, we focus on the energy efficiency problem in small cell networks in which a large number of small cells are spatially deployed in dense urban areas such as office buildings and shopping malls. We optimize the energy efficiency through small cell dormant mechanism under the constraints of average connection ratio (ACR) and average downlink channel capacity. First, we derive the mathematical expressions for average downlink channel capacity and ACR in three-dimensional (3D) small cell networks by Poisson point process (PPP) theory. Second, the monotonicities of channel capacity and ACR are analyzed in detail. Then, based on the results of monotonicity analysis, the optimal small cell dormant probability is calculated to satisfy the constraints of ACR and average downlink channel capacity respectively. Finally, we formulate a network energy consumption minimization problem subject to the constraints of ACR and channel capacity to determine the dormant probability. In addition, we also formulate an optimal maximum connection number of small cells, which minimizes the energy consumption subject to the joint constraints of ACR and channel capacity. Numerical results show that our 3D PPP model is more accurate than the traditional two-dimensional (2D) one in both channel capacity and ACR performance, and that the energy consumption of small cell networks can be reduced by about 21% of the total energy consumption with the dormant strategy in this paper. More importantly, the optimal dormant probability and appropriate configuration of the maximal number of connection can be effectively used to design small cell dormant strategy for 3D small cell networks.