基于EMD算法的手机切换模式判定与道路匹配方法研究
发布时间：2018-12-12 00:07
【摘要】：近年来随着智能交通的推进,手机切换定位技术的应用也在逐步开展,但目前技术应用仍面临着一系列有待研究解决的问题,手机切换道路匹配也是其中的一个关键性难题。论文研究基于以往方法的不足,创新性地将近年来广泛应用于不同模式识别的EMD(Earth Mover's Distance)算法思想引入不同道路手机切换序列的区分中,因为手机切换序列本身也可以作为一种分布模式,而且内在包含切换基站、切换时间的参数,与EMD算法模型需要是吻合的。而面对同样是手机切换定位技术应用中遭遇的一大难题——数据来源问题,采用仿真建模的手段,在深入剖析手机切换机理和手机切换道路匹配问题产生的原理基础上,建立集成交通与无线通信双重网络的手机切换数据仿真平台,作为基础数据来源的重要依托。 EMD算法源于交通运输问题的最优方案选取,数学模型原理为线形规划问题,其求解方法较为成熟。将算法引入手机切换的模式判定,对参数进行有效对应和计算,是算法应用于问题解决的基本思路。鉴于结论的挖掘需要建立在一定量的数据处理和分析基础之上,借助Matlab语言平台对算法计算进行编程,在节省大量人工计算工作量的基础上更加高效、准确地得到数据分析的结果。为了便于从大量数据处理结果中挖掘重要结论和手机切换道路匹配方法的研究,论文设计了算法应用于手机切换序列区分的实验方案,有针对性地将数据分组处理,首先从训练组数据的结果分析中得到结论：相同道路上手机切换序列对之间EMD值远小于不同道路上切换序列对之间的MED值,据此提出手机切换序列道路匹配方法——EMD距离矩阵。在此基础上用测试组数据进行了手机切换道路匹配,结果验证了方法的可行性。
[Abstract]:In recent years, with the advance of intelligent transportation, the application of mobile phone switching and location technology is gradually developing. However, at present, the application of mobile phone is still facing a series of problems to be solved, and the mobile phone handover road matching is also a key problem. Based on the shortcomings of previous methods, this paper innovatively introduces the idea of EMD (Earth Mover's Distance) algorithm, which has been widely used in different pattern recognition in recent years, into the differentiation of different road mobile phone switching sequences. Because the mobile phone handover sequence itself can be used as a distribution mode, and the intrinsic parameters of the handover base station and handover time are consistent with the EMD algorithm model. In the face of the problem of data source, which is also a difficult problem in the application of mobile phone switching and location technology, the method of simulation modeling is used to deeply analyze the mechanism of mobile phone switching and the principle of mobile phone handover road matching. A mobile phone switching data simulation platform integrating traffic and wireless communication networks is established, which is an important support for the basic data source. The EMD algorithm originates from the optimal scheme selection of transportation problem. The mathematical model is linear planning problem, and its solution method is more mature. It is the basic idea to apply the algorithm to solve the problem by introducing the algorithm into determining the mode of mobile phone switching and effectively matching and calculating the parameters. In view of the fact that the mining of conclusion needs to be based on a certain amount of data processing and analysis, the algorithm calculation is programmed with the help of Matlab language platform, and it is more efficient on the basis of saving a lot of manual calculation work. Get the results of the data analysis accurately. In order to mine important conclusions from a large number of data processing results and study the method of mobile phone handoff path matching, this paper designs an experimental scheme of mobile phone handover sequence differentiation, which can deal with the data grouping. First of all, the conclusion is drawn from the result analysis of the training group data: the EMD value between the mobile phone handover sequence pairs on the same road is much smaller than the MED value between the mobile phone handover sequence pairs on different roads. Based on this, a path matching method of mobile phone switching sequence, called EMD distance matrix, is proposed. On this basis, the mobile phone switch path matching is carried out with the test group data, and the results verify the feasibility of the method.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495;TN929.53
本文编号：2373492
[Abstract]:In recent years, with the advance of intelligent transportation, the application of mobile phone switching and location technology is gradually developing. However, at present, the application of mobile phone is still facing a series of problems to be solved, and the mobile phone handover road matching is also a key problem. Based on the shortcomings of previous methods, this paper innovatively introduces the idea of EMD (Earth Mover's Distance) algorithm, which has been widely used in different pattern recognition in recent years, into the differentiation of different road mobile phone switching sequences. Because the mobile phone handover sequence itself can be used as a distribution mode, and the intrinsic parameters of the handover base station and handover time are consistent with the EMD algorithm model. In the face of the problem of data source, which is also a difficult problem in the application of mobile phone switching and location technology, the method of simulation modeling is used to deeply analyze the mechanism of mobile phone switching and the principle of mobile phone handover road matching. A mobile phone switching data simulation platform integrating traffic and wireless communication networks is established, which is an important support for the basic data source. The EMD algorithm originates from the optimal scheme selection of transportation problem. The mathematical model is linear planning problem, and its solution method is more mature. It is the basic idea to apply the algorithm to solve the problem by introducing the algorithm into determining the mode of mobile phone switching and effectively matching and calculating the parameters. In view of the fact that the mining of conclusion needs to be based on a certain amount of data processing and analysis, the algorithm calculation is programmed with the help of Matlab language platform, and it is more efficient on the basis of saving a lot of manual calculation work. Get the results of the data analysis accurately. In order to mine important conclusions from a large number of data processing results and study the method of mobile phone handoff path matching, this paper designs an experimental scheme of mobile phone handover sequence differentiation, which can deal with the data grouping. First of all, the conclusion is drawn from the result analysis of the training group data: the EMD value between the mobile phone handover sequence pairs on the same road is much smaller than the MED value between the mobile phone handover sequence pairs on different roads. Based on this, a path matching method of mobile phone switching sequence, called EMD distance matrix, is proposed. On this basis, the mobile phone switch path matching is carried out with the test group data, and the results verify the feasibility of the method.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495;TN929.53
【相似文献】
相关期刊论文 前2条
1 刘海波;玄志武;;利用微分算子增强EMD算法频带分解能力[J];振动与冲击;2013年18期
2 曹鹏举;魏文斌;李晓峰;刘昕晖;;一种运用EMD算法的装载机动态称重系统[J];工程机械;2007年05期
相关硕士学位论文 前1条
1 刘中游;基于EMD算法的手机切换模式判定与道路匹配方法研究[D];西南交通大学;2014年
，本文编号：2373492
本文链接：https://www.wllwen.com/kejilunwen/jiaotonggongchenglunwen/2373492.html
