基于非泊松点过程的异构蜂窝网络建模与能效优化

With the rapid growth of mobile traffic, the energy consumption problem of mobile network is outstanding day by day. Considering that heterogeneous cellular network is the architecture of the next generation mobile communication, this project tries to model the energy efficiency of heterogeneous cellular network and design several optimization schemes for it. Most of the previous works build the analytic model based on Poisson point process. However, due to the heterogeneity and spatial correlation of base station distribution, these kinds of models do not match with the actual networks. Firstly, the measured data of base station distribution are obtained. Secondly, the measured data of base station are fitted with non Poisson point processes, and the network model for a specific area is built based on the best matching point process. Based on the network model, the relationship between performance and energy consumption of heterogeneous cellular network is analyzed, and the energy efficiency of heterogeneous cellular network is modeled. Finally, several energy efficiency optimization schemes adopting beamforming, power control, node deployment and base station sleep will be proposed, and the joint optimization strategy will also be proposed. Through the study of this project, the advanced theory and technology which can be used to guide the planning and design of heterogeneous cellular networks are expected to form.

随着移动业务量的快速增长，蜂窝网络的能耗问题日渐突出。异构蜂窝网络是下一代移动通信（5G）的网络架构，本项目旨在研究并优化多层异构蜂窝网络的能量效率。现有的研究基于泊松点过程建立分析模型，由于基站分布的异构性和空间相关性，这类模型与现实网络不够匹配。本项目首先获得基站分布的实测数据，然后以非泊松点过程去拟合实测基站分布数据，针对特定区域寻找最匹配的点过程，并采用匹配点过程建立异构蜂窝网络分析模型。在此基础上，研究异构蜂窝网络性能与能量消耗的内在关系，构建能量效率的数学模型，从波束成形、功率控制、节点部署、小基站动态休眠等方面对能量效率进行优化，并给出联合优化策略。通过本项目的研究，期望能获得对异构蜂窝网络规划与设计具有指导意义的先进理论和技术。

随着移动业务量的快速增长，蜂窝网络的能耗问题日渐突出。异构蜂窝网络是5G的网络架构，本项目旨在研究并优化多层异构蜂窝网络的能量效率。大部分现有研究基于泊松点过程建立分析模型，由于基站分布的异构性和空间相关性，这类模型与现实网络不够匹配。本项目主要研究以下内容：.1、基于非泊松模型的异构蜂窝网络性能。研究非泊松模型在异构蜂窝网络中的应用，分析了宏基站与微基站的空间相关性，分别利用泊松硬核过程、Matérn硬核过程和泊松洞过程对异构蜂窝网络进行建模，对信干比增益、网络覆盖率和能量效率进行分析，并与泊松模型下的相关性能进行了对比。.2、基于三维PPP模型的异构蜂窝网络性能分析。针对小基站网络立体分布特性，提出基于三维泊松点过程（3-D Poison Point Process，3-DPPP）的小基站网络模型。根据3-DPPP模型下小基站的空间分布特性推导了TDD制式下小基站网络上下行覆盖概率、上行平均信道容量、上下行频谱效率的表达式。研究结果表明，项目提出的模型更接近实际，能够为覆盖概率和系统容量提供更贴近的边界值。.3、基于泊松模型的异构蜂窝网络能量效率优化。建立基于泊松过程的二维空间两层异构蜂窝网络的能量效率数学模型，提出从微基站发射功率、微基站部署密度、中继休眠策略等方面对能量效率进行优化，并给出相应的优化算法。研究结果表明，通过调整微基站密度和发射功率，可以有效提升网络能量效率，并且联合优化密度和发射功率所取得的优化效果更明显。.4、基于非泊松模型的异构蜂窝网络能量效率优化。利用Matérn硬核过程、β-Ginibre点过程，考虑宏基站为单天线或者多天线，建立两层异构蜂窝网络模型，推导出能量效率的数学模型，并且从调整微基站发射功率的角度对能量效率进行优化。研究结果表明，采用非泊松模型的网络，其能量效率要高于采用泊松模型的网络，这些结论可以用于指导实际的网络部署。