基于上下行解耦接入架构的高能效异构蜂窝网络研究

发布时间：2018-06-23 04:02

本文选题：异构蜂窝网络 + 上下行解耦接入架构　；参考：《电子科技大学》2017年硕士论文

【摘要】：随着互联网、物联网产业的迅猛发展,移动通信系统的数据业务需求呈现出爆炸式的增长。为了在有限的频谱资源下实现网络容量的扩充,异构蜂窝网络应运而生。异构蜂窝网络在传统单层宏蜂窝的基础上,通过在热点及盲点区域叠加其他类型的低功率、低成本设备,实现多类型网络基础设施的综合服务。尽管低功率小基站的引入降低了单个基站的平均能耗,随着基站数量的不断增加,网络的整体能耗开销日益增大。尤其在网络密集化不断加剧的第五代移动通信系统(5G)中,高能效的异构蜂窝网络研究已经成为工业界和学术界关注的热点。本文主要从异构蜂窝网络的接入架构出发,理论分析并设计一种新型的高能效异构蜂窝网络接入架构,上下行解耦接入架构。传统蜂窝网络用户的上行和下行通信接入相同的基站,通常为地理位置最近,也就是接收信号最强的基站,这种上下行耦合的接入方式在异构蜂窝网络中并不适用。一方面,异构蜂窝网络中不同类型基站的发射功率差异较大,地理位置最近的基站未必会提供最强的下行接收信号;另一方面,电池供电的终端用户更倾向上行接入地理位置最近的基站来节约能耗。基于上行和下行不同关联基站的需求,上下行解耦接入架构受到越来越多的关注。本文在已有研究基础上,对该架构的应用场景、面临的挑战、存在的研究难点及解决方案等进行了细致分析,应用随机几何理论和排队论理论研究该架构所面临的关键问题,包括上下行解耦接入架构带来的网络性能增益,该架构的普适性分析,该架构下高能效异构蜂窝网络设计及部署等。本文的主要研究成果如下:1)理论分析了上下行解耦接入架构为异构蜂窝网络带来的性能增益。上下行解耦接入架构与传统接入架构相比,直观上可以提高上行传输的性能,但同时面临着信号处理方面的额外开销,包括信令同步认证、信号转交等。本文应用随机几何理论,对上下行解耦接入架构进行网络建模,从理论上定量分析该架构与传统接入架构相比带来的网络性能增益。研究表明,上下行解耦接入架构为异构蜂窝网络带来了基站的负载均衡、系统频效和能效的提升以及公平的用户服务特性等性能增益,为该架构的工程实际应用奠定了理论基础。2)上下行解耦接入架构适用于多类型的网络场景。本文分析了异构蜂窝网络中的两种典型用户分布场景,包括用户随机均匀分布以及热点地区用户成簇分布,分别采用了泊松点过程和泊松簇过程进行建模。研究表明上下行解耦接入架构可以在两种用户接入场景中提供网络性能增益。尽管用户成簇分布模型加大了理论分析的难度,该研究一定程度上为上下行解耦接入架构应用场景的普适性提供了理论支持。3)理论分析并设计了上下行解耦接入架构的高能效休眠机制。尽管上下行解耦接入架构可以为异构蜂窝网络带来为一定程度上的能效提升,传统高能效休眠机制在这一新型接入架构上将进一步提高系统能效。本文采用随机几何理论对二层异构蜂窝网络的休眠模式进行建模。研究表明,已有上下行非解耦接入架构中采用的休眠机制在新型架构中的直接应用,将过度关闭基站,降低用户服务质量,造成网络实际能效的降低。根据上下行解耦接入架构的特性,我们设计了新型休眠机制,有效节能的同时保障了网络服务。4)设计基于动态用户行为的上下行解耦接入架构的高能效休眠机制。考虑用户业务的动态特性,根据排队论理论,从节约系统能耗和降低服务时延两个角度,折中分析并设计休眠机制,给出了连续休眠时长,即休眠窗口的理论上界和下界。同时分析了上行和下行业务量对休眠窗口大小的影响,为实践中根据业务量设计合适的上下行解耦接入架构休眠策略提供了理论指导。
[Abstract]:With the rapid development of the Internet and the Internet of things industry, the demand for data services in mobile communication systems is increasing. In order to expand the network capacity under the limited spectrum resources, heterogeneous cellular networks emerge as the times require. Other types of low power, low cost devices implement integrated services for multi type network infrastructure. Although the introduction of low power and small base stations reduces the average energy consumption of a single base station, with the increasing number of base stations, the overall energy consumption of the network is increasing. Especially in the fifth generation mobile communication system, which is intensifying in the network denseness. In 5G, high-energy heterogeneous cellular networks have become the focus of industry and academia. Based on the access architecture of heterogeneous cellular networks, this paper analyzes and designs a new type of high-energy heterogeneous cellular network access architecture, uplink and downlink decoupling architecture. The uplink and downlink of traditional cellular network users Communication access to the same base station, usually the nearest geographic location, is the strongest base station receiving the signal, which is not applicable in heterogeneous cellular networks. On the one hand, the transmission power of different types of base stations in heterogeneous cellular networks is much different, and the nearest base station in geographic location may not provide the strongest bottom On the other hand, the battery powered terminal users are more inclined to access the nearest geo location base station to save energy. Based on the requirements of the uplink and downlink base stations, the up-downlink decoupling access architecture is being paid more and more attention. This paper is facing the challenge of the application scenario of the architecture on the basis of the existing research. A detailed analysis of the existing research difficulties and solutions is carried out. The key problems faced by the stochastic geometry theory and queuing theory are studied, including the network performance gain, the universality analysis of the architecture, the design and deployment of the high energy efficiency heterogeneous cellular network, and so on. The main research results are as follows: 1) the performance gain of the upstream and downlink decoupling access architecture for heterogeneous cellular networks is analyzed theoretically. The upstream and downlink decoupling access architecture can improve the performance of the uplink transmission directly compared with the traditional access architecture, but at the same time, it faces the additional overhead of signal processing, including signaling synchronization authentication, signal transfer. In this paper, we use random geometry theory to model the up-downlink decoupling access architecture, and quantitatively analyze the network performance gain of the architecture compared with the traditional access architecture. The research shows that the up-downlink decoupling access architecture brings the load balance of the base station, the system frequency efficiency and the energy efficiency of the heterogeneous cellular network. And the performance gain of fair user service characteristics, which lays a theoretical foundation for the practical application of the architecture.2), the downlink decoupling access architecture is suitable for multiple types of network scenes. This paper analyzes two typical user distribution scenes in heterogeneous cellular networks, including the uniform distribution of users and the clustering of users in hot areas. The research shows that the upstream and downlink decoupling access architecture can provide network performance gain in two user access scenarios. Although the user cluster distribution model increases the difficulty of theoretical analysis, it is a certain extent for the application scenario of the upstream and downlink decoupling access architecture. The adaptability provides theoretical support for.3) theoretical analysis and the design of high energy efficiency dormancy mechanism for the upstream and downlink decoupling access architecture. Although the upstream and downlink decoupling access architecture can bring a certain degree of energy efficiency for heterogeneous cellular networks, the traditional high efficiency dormancy mechanism will further improve the system energy efficiency in this new type of access architecture. This paper uses random geometry theory to model the dormancy mode of two layer heterogeneous cellular networks. The study shows that the direct application of dormancy mechanism used in the upstream and downlink non decoupling access architecture will close the base station, reduce the user's service quality and reduce the actual energy efficiency of the network. The characteristics of the architecture, we designed a new dormancy mechanism, effectively saving energy and guaranteeing the network service.4) to design the high energy efficiency dormancy mechanism based on the up-downlink decoupling access architecture based on dynamic user behavior. Considering the dynamic characteristics of the user service, according to the queuing theory, it is reduced from two angles of saving the energy consumption of the system and reducing the service delay. The mechanism of dormancy is analyzed and designed, and the theoretical upper and lower bounds of the dormancy window are given. The influence of the upstream and downlink traffic on the size of the dormancy window is analyzed, which provides a theoretical guidance for the design of a proper downlink decoupling architecture in practice.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5

【参考文献】