超蜂窝网络的高能效基站部署及其动态控制  

Energy efficient network planning and dynamic control for hyper-cellular network

作　　者：周盛[1,2] 张珊[3] 牛志升 

机构地区：[1]清华大学电子工程系 [2]清华大学信息科学与技术国家实验室(筹) [3]Department of Electrical and Computer Engineering, University of Waterloo

出　　处：《中国科学：信息科学》2017年第11期1566-1582,共17页Scientia Sinica(Informationis)

基　　金：国家重点基础研究发展计划(批准号:2012CB316000);国家自然科学基金(批准号:61461136004;61571265;91638204)资助项目

摘　　要：在控制覆盖和业务覆盖分离的超蜂窝网络架构下,负责数据传输的业务基站可根据业务量的动态变化灵活休眠或唤醒,或将业务转移到相邻业务基站或上层控制基站,以有效降低网络能耗.而如何优化超蜂窝网络部署,并根据用户服务质量要求和网络业务量变化动态调整业务基站部署密度、休眠方式及无线资源分配,是超蜂窝网络中十分重要且具有挑战性的问题.本文总结了该领域的一系列研究成果:在高能效网络部署方面,面向控制基站和业务基站功能及拓扑结构的差异性采用不同的网络模型进行分析,基于随机几何理论分析给出了业务基站和控制基站的最优部署密度;在高能效基站动态控制方面,提出了基于负载转移的高能效动态基站休眠和频谱资源分配机制,可在保证网络覆盖及用户服务质量的前提下有效降低网络能耗.Hyper cellular network architecture is proposed to separate control coverage and traffic coverage, to realize flexible and energy-efficient network operations. Specifically, the traffic base stations（BSs）, which only handle the data services, can be dynamically switched on/off and offload traffic to adjacent traffic base stations or control BSs for energy saving, according to the network load dynamics. With this new feature leveraged, it is crucial and challenging to revisit the problems of network planning, and dynamic BS sleeping and wireless resource allocation, based on the variation of the network traffic load. For the network planning, based on the stochastic geometry theory, the optimal densities of the control BSs and traffic BSs are derived, with respect to their different network functions and topology features. For energy efficient network control, dynamic BS sleeping and spectrum resource allocation mechanisms are proposed and optimized based on traffic offloading, which can substantially reduce the network energy consumption, with the guarantees on the network coverage and user quality of service.

关 键 词：超蜂窝网络 动态休眠 随机几何 负载转移 能效优先 频谱资源管理 

分 类 号：TN929.5[电子电信—通信与信息系统]