【作者】 刘鹏宇；

【导师】 艾渤；

【作者基本信息】 北京交通大学 ， 通信与信息系统， 2014， 博士

【摘要】 随着人们生活节奏的不断加快和路面运行车辆的爆发式增长,公路交通拥堵和公路、铁路交通事故频发成为困扰整个社会的难题。智能交通系统中的车载互联网能够将公路交通中的汽车和高速铁路中的列车分别连接起来,使每辆车维持一个一定范围内的车辆运动状态数据库,以预防碰撞事故发生。同时,车载互联网还能有效调度公路车辆运行,为车辆提供合理且最优的运行路线,以缓解部分地区的拥堵局面。可以说,车载互联网为现代交通运输智能化管理提供了解决方案,而这一切有赖于一个切实有效的车载无线通信系统。众所周知,无线信道建模工作是无线系统设计的重中之重。因此,我们必须广泛而深入地开展车对车无线信道建模工作。本文整体研究思路是将车对车通信细分为汽车对汽车通信和高速铁路列车对列车通信两个分支。首先结合车对车通信所独具的高速移动、高多普勒频移和低天线高度等特点,建立了车对车宽带无线信道建模理论和方法。研究了大尺度路径损耗和阴影衰落,小尺度均方根时延扩展、多径延迟线模型、多径莱斯K因子和多径包络相关性等宽带无线信道建模指标。改进了无线信道冲激响应表达式,以突显车对车通信中的多径生灭特性。提出了适于车对车无线信道建模的统计概率分布,以反映多径包络的小尺度快速时变特性。同时,阐释了改进的非几何统计和二维几何统计两种建模方法在车对车无线信道建模中的深刻内涵和内在联系。通过为模型中的发射机、接收机、静态散射体和漫散射体赋予高度维度,本文首次将基于计算机仿真的汽车对汽车非几何统计性建模方法从二维空间拓展至三维空间。其次,在汽车对汽车无线信道建模方面,本文以无线电波传播环境为驱动,划分了汽车对汽车无线传播场景,建立了汽车对汽车无线传播场景四级架构。以此为基础,遴选车辆视距传播、车辆部分视距传播(斜坡和立交桥)、车辆非视距传播和车辆停车场环境等四大典型汽车对汽车无线通信传播环境,采用改进的非几何统计性建模方法,首次在这些场景中从大尺度路径损耗,小尺度均方根时延扩展、多径延迟线模型、多径莱斯K因子、多径相关性和多径生灭性等宽带无线信道建模指标出发,提出对应的宽带无线信道模型,以探求视距出现程度对汽车对汽车无线信道特征的影响。同时,搭建部分视距场景中斜坡和立交桥环境无线传播信道仿真平台以验证实测信道模型的准确性。首次将绕射损耗计算引入汽车对汽车几何统计仿真平台中,从高度维度入手建立由桥体和坡体遮挡引起的阴影衰落损耗模型,从而扩大了非几何统计性建模方法的可用范围。最后,对于高速列车对列车无线信道建模,本文划分了高速铁路列车对列车无线传播场景。为解决同轨列车在间隔10Km情况下的直接通信问题,搭建了基于邻轨列车中继多跳协作的高速铁路列车对列车物理层无线通信模型,推导了该无线通信模型的误码率和中断概率解析表达式。发现了高速铁路城区高架桥周边分布散射体的二元属性,阐释了城区高架桥场景波导效应成因,并由此出发采用三维几何统计性建模方法,搭建了城区高架桥场景下列车对列车宽带无线信道仿真平台。研究了高速列车对列车无线信道大尺度路径损耗,小尺度均方根时延扩展、莱斯K因子、多径延迟线模型等宽带无线信道特征。更多还原

【Abstract】 With the accelerating pace of life and increasing number of vechicles on the roads and tracks, congestion on the road and traffic accidents frequently occurring urgingly need solving. Internet of vehicles in the intelligent transportation systems connect the vehicles on the roads and tracks, respectively, and thus makes each vehicle maintain a database of vehcle-movement state in certain communication range. The vehicle can immediately compute the distance between each other to avoid accidents. Also, internet of vehicles can contribute to selection of optimal route to alleviate the congestion on the road. Thus, internet of vehicles provides solution to modern intelligent transportation systems, which is basically based on the effective and reliable vehicular wireless communication system. As known to us, wireless channel modeling is crutial to design of wireless communication system. So it is indispensable for us to deeply and thoroughly study the vehicular radio channel characterization.The overall structure of this paper is that we divide the vehicle-to-vehicle communication into two parts, i.e. car-to-car communication and train-to-train communication. Considering characterizatics of vehicle-to-vehicle communication including high-speed, high Doppler frequency shift, and low antenna-height, this paper presents channel modeling theory and approach for vehicle-to-vehicle communication, studys large-scale path loss and shadow loss, and small-scale root-mean square-delay spread, tapped-dealy line model, Ricean K factor and multipath envelope correlation. In order to highlight the multipath birth-and-death in the vehicle-to-vehicle channel modeling, we improve the equation of the channel impulse response. We propose relevant probability distributions for the vehicle-to-vehicle channel modeling. We also demonstrate the similarity and difference of modified non-geometrical stochastic and two-dimension geometry-based stochastic channel modeling approaches in the vehicle-to-vehicle channel modeling.By adding height to the transmitter, receiver, static scatterers and diffuse scatterers, we first extend the two-dimension model to thee-dimenstion space.Secondly, motivated by the radio propagation environment, we construct four-level channel architecture for the car-to-car channel modeling. Car line-of-sight propagation, car non-of-sight propagation, car partial non-of-sight propagation (slope and overpass) and car parking-garage environment are seleted for car-to-car wideband channel modeling to explore the influence of extent of the occurring of line-of-sight to the channel characterization. We also use the geometry-based stochastic modeling approach to build simulation platform of slope and overpass scenario. In this platform, from the height dimension, we first consider diffraction in the geometry-based stochastic channel modeling approach.Finnaly, this paper builds the four-level channel architecture for the high-speed train-to-train channel modeling. In order to solve the problem of communication between trains on the same track10km apart, we develop a train-to-train communication model in physical layer using relay of mobile train on the neighboring track for cooperative communication, and deduce the analytical expression of bit-error rate and interrupt probability. We point out that the binary characterization of scatterers around tracks of the high-speed railway in the urban overpass scenario, and illustrate the waveguide effect of radio propagation in this area. Based on this finding, we study the train-to-train wireless wideband channel characterization using geometry-based stochastic channel modeling approach in the aspect of path loss, root-mean square delay spread, tapped-delay line model and Ricean K factor.更多还原