On The Cellular Signaling Based Transport Trajectory Matching Technologies In Urban Area

Over the past decades, the rapid growth of mobile communication technology has prompted a number of emerging technologies. Signaling data of cellular phones has the advantages of low cost, large amount of data, wide coverage and so on. It has become very valuable information in the field of urban traffic applications. This paper focuses on the evaluation of the travel path of mobile phone users in the traffic system. By using call detail records (CDR), the route of traveling can be determined. This information will provide a vital foundations for urban traffic forecasting and planning, the development of traffic management measures, and even for real-time route guidance. Specifically, the mainly study contents of the paper are as follows:In this paper, by using the handover location data, we propose a travel path matching algorithm based on mobile handover location technology. Firstly, due to the wireless location technology effect, the location data may present erroneous data, we adopt appropriate technologies for data preprocessing, includes the sliding window method process repeated location data, the speed limit threshold method process drift data, the adjacent base station handover time threshold method process Ping-pong handover data. Secondly, due to the coarse shape of the cellular and the different distribution of base stations, we use Voronoi tessellation modeling cellular network. The cellular handover sequence are converted into Voronoi cells handover sequence, thus determining the candidate routes, which covers all handover Voronoi cells. Then, due to the scales of the Voronoi cells a plethora of possible routes could be found. In order to generate the best match route, the map matching algorithm based on global minimum Euclidean distance is developed. Finally, China Telecom real CDR and synthetic data are used for simulation tests. The effectiveness of the proposed algorithm are improved by experiment and comparative analysis.In this paper, the extension of the idea of handover data matching algorithm, we propose a new travel path hybrid matching algorithm based on mobile handover location technology. The proposed method consists of two steps:1) determining the best match link of each handover,2) through best match link connection to determine travel path. Firstly, the link which covers corresponding handover Voronoi cells as the candidate links of each handover. Secondly, due to the fuzzy uncertainty of the determining the best match link of each handover, the problem is transformed into the multiple attribute decision making problem. An algorithm for fuzzy multiple attribute decision making based on preference relationship is proposed. By establishing evaluation value model to describe the decision making problem of the best match link of each handover, determining the best match link of each handover. Then, there are two key problems with the best match link connection to determine travel path:1) there are many links between two adjacent continuous handover,2) matching results of the multi-attribute decision matching algorithm based on fuzzy preference relationship may be mistaken. In order to solve the problems, map matching historical information and speed information limits of the road are used to identify the best route between two successive handover. Eventually generating a complete handover corresponding travel path. Finally, China Telecom real CDR and synthetic data are used for simulation tests. The effectiveness of the proposed algorithm are improved by experiment and comparative analysis.