【作者】 黄宇；

【导师】 王文博；

【作者基本信息】 北京邮电大学 ， 通信与信息系统， 2014， 博士

【摘要】 传统通信网络采用单层宏基站负责覆盖,随着业务量的爆炸式增长,宏基站的数量将很快饱和,导致基站间干扰和无线网络功耗不断增加,再通过增加宏基站密度的方式来提升网络容量变得十分不经济。近年来绿色节能通信技术得到了广泛而深入的研究,使用新型低功耗的微基站与宏基站组成的异构无线网络,在满足高速业务传输和无缝连接需求的同时,大大减少了不必要的能量消耗,从而成为下一代绿色无线通信网络的发展方向。在异构无线网络架构下,高能效网络资源控制技术作为提高网络能量效率的重要手段,因其性能优异、实现灵活和可操作性强等特点,逐渐成为节能技术研究的重点。本论文在深入分析业务在时间、空间和内容维度上的不均匀性现象的基础上,对网络能效与资源控制之间的关联关系进行理论研究,从而揭示出业务不均匀性对高能效网络资源控制技术的影响和制约,为在下一代无线网络中应用该技术提供必要的理论支持。本论文首先对业务在各个维度上的不均匀程度进行量化建模,并研究了宏基站和微基站共存的异构无线网络能量效率。在此基础上针对业务空间上的不均匀性,研究高能效地对无线网络中的站点密度、发射功率和传输带宽资源进行控制和部署。然后进一步根据业务发起时间的不均匀性,提出高能效的基站开关休眠以及覆盖范围缩放的控制策略,最后结合业务在内容上的不均匀性,设计高能效的单播组播混合传输模式控制策略,满足低能耗和低延时的传输需求。论文的主要工作和创新点在于：1.业务不均匀性的量化建模目前学术界缺乏业务模型可以对业务各个维度上的不均匀程度进行衡量,本文首次建立了业务不均匀性的量化模型：首先针对业务在时间、空间和内容各个维度上的不均匀性进行了统一模型,提出业务聚集曲线对业务的不均匀性进行直观描述,并定义业务聚集系数对各个维度上的业务不均匀性进行定量度量。然后推导得到业务在多种经典非均匀概率分布下的业务聚集系数表达式。最后,给出实际通信网络普遍存在的业务在热点和非热点区域不均匀分布的业务聚集系数表达式。该模型是对传统业务不均匀性研究的补充和完善,为在无线网络中利用业务不均匀现象降低能耗提供了理论支撑。2.网络能效、网络资源、业务不均匀性之间关联关系研究传统网络能效理论分析缺乏考虑业务不均匀性的问题,本文针对业务不均匀性对网络能效的影响展开研究,首先建立了宏基站和微基站共存的异构无线网络模型,并分析了同频和异频情况下的空间平均速率表达式,然后推导出适用于单层和双层无线网络能效表达式,从理论上给出无线网络的能效与传统点对点传输能效的差异。所得表达式包含了不同类型基站的发射功率、固定功率以及站点密度,应用场景广泛。在此基础上再进一步分析了不同的业务不均匀程度对高能效网络资源控制的影响,并推到出相应的网络能效表达式。从而在理论上得到三者的关联关系,论证了业务不均匀性在高能效网络资源控制方面存在巨大研究价值和节能潜力。3.针对业务空间不均匀性的高能效基站密度功率带宽控制研究传统针对业务均匀分布场景的网络资源控制理论研究,无法发掘业务空间不均匀性带来的节能潜力,本文在所提业务不均匀性量化模型的基础上,结合对异构无线网络能效分析,提出能效最优的宏基站和微基站的发射功率和站点密度联合优化配置策略。然后再利用部分频率复用以及载波聚合技术,提出让微基站使用宏基站广域覆盖所不适合使用的频段的带宽分配策略,兼备了发射功率低和不对宏基站用户产生干扰的优点,从而进一步论证了网络资源控制技术在提高网络能效方面的可行性和必要性。这些方案均能准确适配业务的空间不均匀性,因此可以直接应用于实际网络的规划和优化。4.针对业务时间不均匀性的高能效基站开关及覆盖范围缩放研究受限于以往无法对业务不均匀性进行量化分析,传统基站开关及覆盖缩放技术的理论分析与算法设计不能有效地适应业务随时间的变化。本文基于所提的业务不均匀性量化模型,从理论上解决了基站最优开关策略与最优覆盖大小的问题：首先针对局部区域的业务量高低随时间变化的问题,提出了高能效基站开关策略,然后针对高业务量区域的面积大小随时间变化的情况,给出了高能效的基站覆盖缩放策略及最优覆盖范围表达式。该方案在保证业务连续性和服务质量的前提下,能根据不同区域业务量高低和面积大小随时间的变化,实时对基站的工作状态进行精确控制。5.针对业务内容不均匀性的高能效传输模式控制研究大量的视频业务需求会严重消耗网络资源和加剧能量消耗,而视频业务在内容维度上的不均匀性意味着存在小部分业务拥有非常大的需求量,因此如果能够把请求量很大且内容相同的业务合理地聚合起来进行组播传输就能够在很大程度上节约网络的资源和能耗。本文首先针对非实时业务,提出在一段时间窗口内,把内容相同的业务请求收集起来后,再高能效地进行单播组播混合传输的方案。为了增强节能效果,进一步给出一种多小区联合组播的传输方案,不但能够收集到更多的内容相同的业务请求,而且通过多天线技术提高传输信号质量。最后利用增补传输单播数据流的方法克服了组播传输带来的延时问题,设计了一种实时高能效的单播组播混合传输方案,同时具备低功耗和零延时的优势。更多还原

【Abstract】 Green wireless communication is considered as the most promising method for reducing the increasing energy consumption, which has been researched extensively and intensively recently. Wireless resources includes BS density, transmit power and frequency bandwidth. As an effective method to improve the energy efficiency of wireless network, energy-efficient resources management of wireless networks has rapidity become a hot issue in green communications. The heterogeneous network is a new trend of wireless network development, which including the complementary low-power nodes providing very high data rates in hotspot region and the traditional macro nodes providing wide-area coverage in non-hotspot region. Such new architecture can reduce the increasing power consumption of network while meet the increasing data rates and the seamless connection of traffic demands. Our research mainly includes the analysis the heterogeneity of traffic demands in the time/spatial/content dimensions, the quantitative relationship between the energy-efficient resources management and energy efficiency of wireless network, the influence on the heterogeneity of traffic demands and thus provides necessary theoretical support to apply this energy-efficient resources management in next generation wireless networks. In this dissertation, we present a quantitative model for heterogeneity of traffic demands. Based on this model, the energy-efficient two-tier deployment and configuration is proposed to adapt the heterogeneity of traffic demands in the spatial dimension, where the base station (BS) density, BS transmit power and frequency bandwidth are taken into account. We present closed-form formulas which establish the quantitative relationship between heterogeneity of traffic demands and energy-efficient network resources allocation. In order to adapt the heterogeneity of traffic demands in the time dimension, we further present three energy-efficient control strategies of micro BSs, including micro BS sleep control, coverage expansion control and coverage shrinking control. Last we propose three energy-efficient multicast schemes in order to adapt the heterogeneity of traffic demands in the content dimension, which can potentially lead to significant power savings. The main study points are shown as follows.1. Quantitative model for heterogeneity of traffic demandsThe existing traffic models are unable to characterize the heterogeneity of traffic demands and thus we develop the quantitative model for heterogeneity of traffic demands in the time/spatial/content dimensions. First, we propose a traffic heterogeneity curve to describe the heterogeneity phenomena of traffic demands and then define a traffic heterogeneity coefficient to quantitative characterize the heterogeneity of traffic demands. Secondly, we calculate the traffic heterogeneity coefficient for several classic heterogeneity probability distributions. Last, we derive the traffic heterogeneity coefficient for the scenario where the traffic demands are non-uniformly distributed in hotspot regions and non-hotspot regions. This research can be regarded as the improvements for the traditional study of heterogeneity of traffic demands and provides theoretical support for the analysis and utilizing heterogeneity of traffic demands in the rest of paper.2. Research on theoretical relationship between energy efficiency, resources and heterogeneity of traffic demandsTraditional theoretical analysis of energy efficiency of wireless network is unable to consider the impact of heterogeneity of traffic demands. We first develop the heterogeneous networks model including both macro BSs and micro BSs and drive the expressions for outage probability, average spatial rate and network energy efficiency, which shows the diffidence energy efficiency of point-to-point transmit and that of networks. These expressions includes the BS density, BS transmit power and frequency bandwidth and thus very flexible in different scenarios. Furthermore, we analyze the impact of different degree of heterogeneity of traffic demands on energy efficiency of wireless network and resource and prove the significant potential of energy-efficient resource management in saving network energy consumption.3. Research on energy-efficient resource allocation networks for spatial heterogeneity of traffic demandsTraditional energy-efficient resource allocation is unable to consider the impact of heterogeneity of traffic demands. Based on the proposed quantitative model for heterogeneity of traffic demands, we develop the optimal BS transmit power and BS density configuration for wireless networks. Secondly, we design the optimal bandwidth allocation strategies for fractional frequency reuse and carrier aggregation scenarios. The micro BSs can use the frequency band for hotspot regions coverage which is not suitable for macro BSs in wide-area coverage. This technique needs lower transmit power and will not generate the interference to the users severing by macro BS and further prove the significant feasibility of energy-efficient resource management in saving network energy consumption and thus can be directly deployed in real wireless networks.4. Research on energy-efficient BS sleep control, coverage expansion and shrinking control networks for time heterogeneity of traffic demandsTraditional BS sleep control, coverage expansion and shrinking control are not precisely designed because lack of heterogeneity of traffic demands. In order to meet the fluctuation of traffic demands in time dimension, we develop the BS sleep control strategy based on the heterogeneity of traffic demands. Due to the energy ineffective BS coverage for hotspot regions, we proposed the BS coverage expansion and shrinking control. This technique can further decrease the power consumption of BS while guaranteeing the Quality of Service (QoS) and the seamless connectivity of traffic demands.5. Research on energy-efficient BS unicast and multicast mode control networks for content heterogeneity of traffic demandsIn recent years, the widespread requirement of wide-band multimedia communication, especially Video-on-Demand, has led to rapidly rising energy consumption and wireless network resource. The heterogeneity of traffic demands in the time/spatial/content dimensions means a large number of users request the same traffic within short time interval thus the transmission technique can aggregate the traffic demands for the same content of traffic can significantly improve the energy efficiency of wireless network. We propose an energy-efficient transmission scheme where in each time-window the transmitter analyzes the similarity of users’traffic requests and the similar traffics will be transmitted by multicast mode while the other traffics will be transmitted using unicast mode. Secondly we extent this scheme to adept the multi-cell and multi-antenna scenario which can aggregate more traffic demands for the same content and provide higher multicast data rate. Last, in order to overcome the transmit delay in previous two schemes, we propose an energy-efficient multicast scheme with patching stream which enable the transmitter to deliver both patching streams through unicast transmission guaranteeing the real-time demand and shared streams through multicast transmission guaranteeing high energy efficiency.更多还原