Extraction And Application Of Urban Travel Trajectory Based On WiFi Probe Data

The research of urban travel trajectory mining provides an important reference for urban planning decision-making,traffic management,facility operation and other work.Traditional travel behavior survey methods mainly include household visit,questionnaire and survey based on GPS data.In recent years,with the rapid development of the city and the increasing richness of residents’ life,the traditional methods of residents’ travel survey have become more and more difficult and the data coverage has become narrower and narrower.At the same time,due to its powerful functionality and convenience,mobile intelligent terminal has been rapidly popularized,which basically everyone has one and takes it with them wherever they go.Therefore,WiFi probe data,as a kind of terminal detection record data,can provide a new data source for urban travel trajectory mining.Based on the above understanding,this paper proposes the urban travel trajectory extraction and application research based on WiFi probe data.(1)Data preprocessing.In order to improve the quality of data input,aiming at the problems of WiFi probe data,data preprocessing is carried out to ensure the efficient operation of the extraction algorithm.In addition,in order to obtain the speed of each road section in the road network in different periods,the coarse value of road section speed extracted from WiFi probe data is filtered,and then combined with the road network vector data extracted from satellite images and data provided by relevant departments,the road network time and space data is obtained.(2)Travel trajectory extraction based on WiFi probe data.Firstly,according to the general definition of "one trip",the WiFi probe data is processed into a minimum terminal trip detection record segment that meets the definition of "one trip",and then the signal strength RSSI value is used to extract the candidate point set on the road network for each record.Secondly,an algorithm based on local evaluation is designed to gradually select the best matching points from the candidate points.Then,for the non adjacent missing trajectories in the sequence of matching points,the path search algorithm based on depth first and TOPSIS decision method are used to reconstruct the optimal supplementary path.Finally,the WiFi probe data collected in Dongguan City is used to test the feasibility of the method.(3)Travel characteristics analysis based on WiFi probe data.Based on the trajectory extracted from the previous step,the characteristics of residents’ trips in the region are studied.The similarities and differences of daily trips and the time distribution rules of the working days,rest days and holidays in the region are analyzed and compared.The influence of COVID-19 on city economy is also analyzed from the total travel volume between January 2020 and July.The changes of traffic flow and road congestion index of each road section in the whole day and morning and evening peak hours are analyzed and visualized in Arc GIS.This paper takes WiFi probe data as the research object,from which the extraction of urban travel trajectory and the analysis of travel characteristics are realized.The research results provide a new solution for urban trajectory mining,and also provide a certain data basis for urban planning decision-making.