User Travel Behavior Research Based On Mobile Phone Signaling Data

The research on the resident travel behavior is of great significance to the urban traffic planning and layout,and rational utilization of urban traffic space.Due to the high cost and long cycle of investigation,the traditional method of resident travel survey is difficult to obtain real-time and extensive travel information.With the continuous development of the growing popularity of mobile phone and mobile communication network,when the mobile phone is connected to the mobile network,it will produce a large number of signaling data,which records the location information of mobile phone and the time.Therefore,it is possible to obtain the resident’s travel behavior through the mobile phone.The use of mobile phone signaling data to study the resident travel behavior has the advantages of low cost and wide coverage,which makes up for the shortcomings of the traditional user travel survey methods.In this thesis,we study the resident travel behavior through the mobile signaling data,including the analysis of spatial and temporal distribution characteristics of travel,travel purpose judgment and trip mode identification.Specific work contents are as follows.1.This thesis analyzes the advantages and disadvantages of DBSCAN density clustering algorithm,and proposes an improved DBSCAN density clustering algorithm to partition the traffic area.Due to the uneven distribution density of base stations in adjacent areas,adopting the general DBSCAN density clustering algorithm can not accurately identify the adjacent and different density of the region,which affect the clustering results.Therefore,an improved DBSCAN clustering algorithm is proposed in this thesis.The main idea of this thesis is to set up a density threshold as the condition of cluster expansion.Thus,an improved DBSCAN clustering algorithm is proposed in this thesis.The main idea of this thesis is to set up a density threshold as the condition of cluster expansion.2.Traffic community semantic processing.Carry cluster partition traffic area into the semantic processing with POI data,according to the proportion of POI data types of each traffic area,the theme is given to the community to facilitate the purpose of the user travel.3.Analysis of Temporal and Spatial Distribution.According to the mobile phone user’s trajectory,counts the trip proportion per hour,in hours,analyses the characteristics of the user travel time.Then,combines with the division of traffic area,the traffic volume of each traffic area is calculated,analyses the spatial feature of user travel.4.Travel mode identification.In order to improve the accuracy of recognition,on the one hand,fuzzy identification according to the membership function of each travel mode.On the other hand,combined with Baidu maps API route track and signal trajectory matching,to further identify the way the mobile phone users travel.5.Case Study.Analysis above resident travel behavior analysis model combined with specific examples.