基于随机网络演算的高速铁路无线网络性能评估

发布时间：2018-01-07 15:01

本文关键词：基于随机网络演算的高速铁路无线网络性能评估　出处：《北京交通大学》2015年硕士论文　论文类型：学位论文

【摘要】：摘要：近年来,为了更好地满足移动通信用户日益增长的需求,高速铁路下一代宽带移动通信系统LTE-R (Long Term Evolution for Railway)的研究得到了广泛关注。为了更好的检验LTE-R的系统性能,保证高速铁路(High-Speed Railway, HSR)能安全可靠的运行,我们必须在铁路系统开始运营前,对它进行准确的系统性能评估,从而优化使其性能满足一定的要求指标。LTE-R是准4G技术LTE在高速铁路上的一个应用,它承载了LTE的关键技术,比如：全IP包交换,这意味着所有已有的以及新兴的铁路业务都将通过包交换方式传输。以往对于高速铁路性能评估的研究较多采用仿真方法,这一方法固然可以较为准确的分析性能,但却耗时较长、成本较高,应用起来具有局限性。而随机网络演算是一种先进的网络性能分析理论工具,可以用来解决现代包交换网中的性能保证,并且耗时和成本较低,准确性较高,可以较为容易的进行性能预测,这就为分析LTE-R系统的网络性能开辟了一条新途径。本文旨在采用随机网络演算的理论方法分析高速铁路无线网络性能,利用实测数据构造高速铁路专用的业务模型,构造综合考虑大尺度衰落和小尺度衰落的信道模型,基于业务模型和信道模型得到理论性能结果。并依据MATLAB平台仿真结果对随机网络演算理论方法的准确性进行验证,同时分析高速铁路场景下的时延等网络性能指标。本文的创新点主要体现在以下几点：一、构造刻画高速铁路无线信道的大尺度及小尺度衰落的分组级马尔可夫模型。将高速列车的移动模型建模半马尔可夫过程,并将无线信道的瞬时数据速率建模为半马尔可夫调制过程,同时考虑AMC选择对性能的影响。基于半马尔可夫调制过程,得到高速铁路通信系统的随机服务曲线。二、构造合理的高速铁路专用业务模型。通过对收集到的列车控制实测业务数据进行分析,采用随机到达曲线模型对列车控制业务进行建模。三、对比并分析CCDF随机网络演算和MGF随机网络演算。采用随机网络演算方法求得列车控制业务的积压和时延边界,并将两种方法下得到的时延边界进行对比,分析它们各自的优缺点。四、验证并分析随机网络演算的准确性。将时延边界的理论结果与仿真结果对比,证明随机网络演算方法的可行性与正确性,同时依据边界结果对下一代高速铁路移动通信系统LTE-R进行性能分析。
[Abstract]:Abstract : In recent years , in order to meet the increasing demands of mobile communication users , the research of LTE - R ( Long Term Evolution for Railway ) of the next generation broadband mobile communication system of high - speed railway has been paid much attention . In order to better check the system performance of LTE - R , it is necessary to accurately evaluate the system performance before the railway system begins to operate . The LTE - R is an application of LTE in high - speed railway . The LTE - R is a key technology of LTE , such as the whole IP packet switching , which means that all existing and emerging railway traffic will be transmitted in a packet - switched manner . This paper analyzes the performance of high - speed railway network by using the theory of random network calculus , which can be used to analyze the performance of high - speed railway . The innovation points of this paper are mainly embodied in the following points : Firstly , the large - scale and small - scale Markov model of high - speed railway wireless channel is constructed . The mobile model of high - speed train is modeled as semi - Markov process , and the instantaneous data rate of the wireless channel is modeled as a half - Markov modulation process , and the influence of AMC selection on performance is considered . Based on the semi - Markov modulation process , a random service curve of high - speed railway communication system is obtained . secondly , constructing a reasonable service model of the high - speed railway , and carrying out analysis on the collected train control measured business data , and adopting a random arrival curve model to model the train control service . Thirdly , the CCDF random network calculus and the MGF random network calculus are analyzed and analyzed . The integrated pressure and time delay boundary of the train control service are obtained by the random network calculation method , and the time delay boundary obtained by the two methods is compared , and their advantages and disadvantages are analyzed . Fourthly , the accuracy of random network calculus is verified and analyzed . The theoretical results of the time delay boundary are compared with the simulation results to prove the feasibility and the correctness of the random network calculation method . At the same time , the performance analysis of the next generation high - speed railway mobile communication system LTE - R is carried out according to the boundary results .

【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U285;TN929.5

【共引文献】