Map matching of mobile probes based on handover location technology

With the rapid development of wireless location technologies, mobile terminals as probes have been regarded as a feasible and low cost method for real-time and wide-area city traffic information collection. If mobile probes deployed in the suburban areas, it will be a feasible supplementary way of real-time traffic information collection in China. This paper tries to investigate map matching problem of mobile probes for further traffic information extraction in the suburban areas. In this paper, handover location technology was firstly used to collect location information for a test mobile probe in field runs on a stretch of highway in the inner suburban area in Beijing. GPS data were also collected in field tests. A map matching process is proposed to find the most matched path for the test mobile probe, including two map matching algorithms, the first one of which is used for determining the matched link for each handover in a sequence of handover and the second one of which is used for determining the matched path for the whole sequence of handover. In the first map matching algorithm, frequency of handover on a candidate link, distance between the matched position and starting node of a candidate link, and geographic topological relation were combined by D-S evidence theory to determine the matched link for each handover. In the second map matching algorithm, historical matching results, real time traveling time information, drivers´ route selection behavior were combined together to find the matched path for the test mobile probe. Finally, map matching performance was verified and discussed by field data collected from test runs.