【作者】 张祖凡；

【导师】 朱维乐；

【作者基本信息】 电子科技大学 ， 信号与信息处理， 2007， 博士

【摘要】 如何提高蜂窝移动通信系统容量,一直受到学术界、工业界的广泛关注。本论文的研究课题就是研究提高蜂窝移动通信系统容量众多方案中的同频干扰和越区切换两个问题。研究表明,蜂窝移动通信系统的基本组成单元—蜂窝小区的几何特性直接影响同频干扰和越区切换两个问题。为此,本论文围绕蜂窝小区几何特性对同频干扰和越区切换两个问题的影响从三个方面展开了深入地研究:(1)基于蜂窝小区的同频干扰模型研究;(2)基于蜂窝小区的越区切换模型研究;(3)基于基站自适应多波束天线的分层蜂窝小区的同频干扰和越区切换问题的研究。本论文主要研究工作与贡献表现如下:1.针对现有扇形蜂窝小区同频干扰计算模型中,通常认为同频干扰小区对有用小区内的所有激活用户产生下行干扰。本文提出了一种结合有用小区和同频干扰小区内用户位置概率分布的同频干扰计算模型,使得同频干扰小区对有用小区内的部分激活用户产生下行干扰。研究结果表明,采用所提出的计算模型可以提高同频干扰计算精确度。2.不同于现有同频干扰计算模型中将用户位置概率分布通常用其所属基站位置作为坐标系的原点表示。本文提出了用户位置概率分布用所属基站(有用信号)或干扰基站(干扰信号)位置分别作为坐标系的原点来表示。理论分析表明,这种模型有利于将用户位置概率分布与同频干扰的计算联系更为紧密。3.针对现有同频干扰计算模型,通常认为用户位置概率分布在小区内是连续的,从而采用连续积分方法求得小区内平均同频干扰。本文提出了将顺序统计量引入到蜂窝移动通信同频干扰计算模型中,顺序统计可以实现较为实际的离散化用户位置概率分布。研究结果表明,用户位置概率分布离散化可以将小区及其同频干扰小区内激活的用户或信道数直接引入到同频干扰计算模型中。4.针对现有用户驻留时间分析主要在圆盘区域内用户运动方向和运动速度下进行。本文提出了一种在扇形区域内用户位置概率分布、运动方向、运动速度下分析驻留时间概率密度函数模型。理论分析和数值计算结果表明,蜂窝小区几何结构和用户位置概率分布影响驻留时间概率密度函数及话务量性能。5.为了改进智能天线技术在蜂窝移动通信中进一步的应用,本文提出并研究了智能蜂窝小区结构中天线波束宽度对驻留时间概率密度函数的影响。研究结果表明,基于智能天线的智能蜂窝小区结构可以增加小区用户驻留时间,改善系统话务量性能。6.与传统的分层小区结构不同,本文提出一种基于基站多波束天线在垂直面分割形成的分层蜂窝小区结构。研究结果表明,新的分层蜂窝小区结构具有提高频谱利用率和灵活地实现动态覆盖等性能。更多还原

【Abstract】 How to improve the capacity of mobile cellular communication systems have been attracting great interests of industry and researchers. This dissertation deals with two major issues of improving the capacity of the system, namely, co-channel interference and handover. Recent research indicates that the geometry of the system fundamental unit, i.e., cellular, has immediate influence upon the two major issues. Therefore, this dissertation will focus on the following three key aspects related to the two issues, i.e. (1) research an improved analysis model of co-channel interference based on cellular geometry; (2) research an improved analysis model of handover based on cellular geometry; (3) research on the co-channel inference and handover of a proposed hierarchical cellular configuration based on adaptive multi-beam base-station antennas.The main research work and contributions of the dissertation include the following aspects:1. Since the existing analysis model for co-channel interference generally considers that the inference power from the interfering cells will interfere all the active users on the downlink, this dissertation proposes an improved analysis model for co-channel interference considering the users’ position probability distribution in the desired and interfering cells, where the inference power from the interfering cells will interfere only partial active users on the downlink. Simulation results show that the improved analysis model can enhance precision in calculation of co-channel inference.2. Different from the existing analysis model for co-channel interference, which normally expresses the users’ position probability distribution referring its home base-station as the coordinates origin. This dissertation proposes an improved expressing model, which refers its home base-station (desired signal) or interfering base-stations (interference signals) as the coordinates origin on the up and down link. Theoretical results show that the proposed model will touch the relations between the co-channel calculation and the users’ position probability distribution.3. Since the existing analysis model for co-channel interference normally get the average co-channel interference by continuous integral over the cells due to the continuous users’ position probability distribution. This dissertation further proposes an improved analysis model for co-channel interference by introducing the order statistics, which will produce a discrete users’ position probability distribution press close to the real occurrence. Theoretical results show that the proposed model can directly introduce the active users or channels in the cells into the co-channel calculation.4. Different from the conventional cell dwell time analysis model considering the users moving direction and velocity in a specified disc region. This dissertation proposes analyzing the probability density distribution of dwell time should consider the users’ position probability distribution and moving direction and velocity in a specified sector region. Theoretical and simulation results show that cellular geometry and the users’ position probability distribution influence the probability density distribution of dwell time and the performance of cell traffic.5. This dissertation proposes and researches the relations between the beam width of the smart antenna and the probability density distribution of dwell time. Simulation results show that smart cellular will increase the dwell time and improve the performance of cell traffic.6. This dissertation proposes a novel hierarchical cellular system. Different from conventional hierarchical cellular system, the proposed one is based on multi-beam base-station antenna splitting in the elevation-radiating plane. The validity of the proposed has been checked through spectrum efficiency and dynamic coverage with simulation results.更多还原