【作者】 刘荷莉；

【导师】 宋梅；

【作者基本信息】 北京邮电大学 ， 电子科学与技术， 2013， 硕士

【摘要】 随着网络技术的不断发展,4G LTE技术的研究已逐渐成熟并将很快投入使用。在LTE成为下一代移动通信系统中最重要无线技术的网络中,语音业务将仍然是最重要的业务之一。LTE之所以非常适合承载基于IP的语音数据,是因为它采用了OFDM等技术使得其具有很高的频谱利用率、抗衰落性、高带宽、大容量等优点。由于4G网络的初期部署将采用点状部署的方式,使得在LTE的边界通信质量将大幅度的下降,因此寻找传统的网络进行业务切换,是解决LTE初期部署阶段,业务连续性不符合要求问题的当务之急。本文针对用户环境感知与LTE中语音业务切换方案两大内容进行讨论研究,主要围绕着LTE初期部署时的点状分布情况导致的语音业务不连续问题进行分析,得出将现有业务切换到其他网络为最佳选择的结论。本文的第一个研究方面,环境感知,主要用于决策哪一个网络时最佳目的切换网络,第二个研究点,语音切换小区选择方案改进,是用于对选好目的网络后的切换流程进行改进,二者具有递进关系,可应用于网络切换准备和切换执行阶段。通过对环境感知问题的分析,本文设计出一种利用物理层用户通信信道的时延参数来感知主体所在的无线环境、网络环境或用户环境的方案。这种方案可以基于感知到的信息适当的调整网络配置的参数,能够使用户更好的适应环境的变化,增强网络中的用户体验。这里所提到的用户环境指用户所处的周围地理环境,包括室内或室外,平原或高山等情况。为了实现这一方案,本文采用了VLSI可变长度的结构自适应信道均衡器,该均衡器可以根据用户终端所在环境信息来自适应的改变自身抽头阶数和系数。而通过调研可知，抽头阶数与某环境下的多径时延是密切相关的。通过关联可变长度的结构自适应信道均衡器的阶数与环境延,来智能判决用户终端所处的具体环境,进而实现了为用户终端的网络选择最优化提供判决依据。基于以上分析,本文提出了基于BSAEQ环境感知方案的最优网络选择模型。当LTE用户由于网络的点状部署而需要进行话音业务切换时,利用以上提到的环境感知方法,可以根据用户所处环境帮助其确定新接入的最优网络应该是WLAN抑或2G／3G移动通信网络。当用户切入到2G／3G网络时,应用到LTE向2G／3G网络切换的SRVCC机制。本文深入研究该机制,并重点针对数据域业务与电路域业务协同切换的情况进行研究,分析LTE切换至2G／3G网络时传统小区重选方法的不足之处,并结合PS域核心网知识,综合考虑小区信号强度与SGSN的利用率,对小区重选问题提出改进方案。基于LTE网络的分阶段部署方案,分场景给出语音切换改进方案的设计。通过对所有可能切换场景的分析,将小区选择方案归纳为两种：传统的S准则(S Criteria)和改进的ES (Evolved S Criteria)算法方案。通过两种方案的仿真比较,可以发现ES算法较传统的S准则能更好的适用于接收信号功率较小的情况,并在文章最后给出了SRVCC场景下小区选择改进方案流程。更多还原

【Abstract】 With the continuous development of network technology,4G LTE technology is maturing and will soon be put into use. With the LTE becoming the most important wireless technology in the network of the next-generation mobile communication system, voice service will still be one of the most significant businesses. LTE has the advantages of high spectrum efficiency, anti-fading, high bandwidth, high capacity, so it is very suitable for carrying the IP-based voice data. Since it will adopt dot deployed measure in the initial deployment of4G networks, so the.communication quality in LTE border will decline significantly, and business continuity does not meet the requirements of the subscribers. The best way to solve this problem is to look for a traditional network as the destination network through handover technology.This article focuses on two research aspect, one is which network is the best one to handover, and the other one is how to implement this approach.Through the analysis of the context aware, we designed a scheme which uses the communication multipath delay parameters in physical layer to perceive the location and surrounding of the terminal. In order to achieve this program, we use a VLSI variable length structure adaptive channel equalizer. This equalizer adapts its own taps orders and coefficients to suit and match with the changing of the environmental information. Plenty of researches show that the tap orders of an equalizer and multi-path delay in a certain environment are closely related. By an associated database of tap orders and the multipath delay, the terminal can intelligently specific its surrounding, thereby achieving the optimization of the network selection for terminal. Based on the above analysis, we propose the optimal network selection model based on BSAEQ context-aware programs.When LTE users need to handover to other networks because of going out of the LTE network, the above-mentioned method can help them select the optimal network. Generally, the optimal network will always be WLAN or 2G/3G mobile communication network. When the destination network is2G/3G network, SRVCC mechanism designed for LTE handover to2G/3G network switching should be applied. This article studies SRVCC mechanism and focuses on the collaborative handover scheme containing not only packet system but also circus system service. Considering the inadequacies of the traditional method of cell reselection and the knowledge of the PS in mobile system core network, we combine the utilization of cell signal strength and the SGSN and come up with a novel cell selection scheme called ES (Evolved S Criteria). According to the deployment phases of LTE networks, we analyze all the possible service switching sub-scenes for improved cell selection scheme and classified them into two:the traditional S criterion (S Criteria) and the ES (Evolved S Criteria) algorithm program. By simulation, we can find that ES algorithm is better than the traditional S criterion when applying a small received signal power. In the last of this article, we conclude the final improved cell selection processes in SRVCC scene.更多还原