Users’ Mobility Analysis Based On Massive Data Of CDRs

With the development of mobile communication technology, mobile phones are widely popular, resulting in a flood of mobile communication data. Wherein the call detail records (CDR) contains user’s call-related information, such as call time, duration, and the communication cell connected. Combining with the position information of base stations, CDR substantially reflects the user’s activity trajectories. Massive mobile trajectories contain the temporal and spatial distribution characteristics of human behavior, from which we can dig people’s moving mode, understand the dynamics of human behavior, and meanwhile we can understand the lifestyle, environment, traffic conditions, and the level of development of the area where users reside. Further, we can then provide references for urban construction, road planning and social development.Since recently studies on users’ mobility based on Call Detail Records (CDR) mainly use metrics in1D space, such as the travel distance and the radius of gyration which cannot exactly describe the scope of users’mobility, the Area of the Convex Hull Covering a user’s daily Trajectory (ACHCT) was applied to investigate users’ mobility scale in2D space, and the mobility vector was introduced to study the mobility of the crowd. First, a method was designed to set up2D Cartesian coordinates based on latitude and longitude coordinates. The method applied the Mercator projection and the Haversine formula to calculate the bearing and distance between scattering points, based on which the coordinates of points in the plane coordinates were determined. Then, based on the coordinates, the convex hulls covering users’daily trajectories were calculated and the distribution of the areas of all convex hulls was analyzed. Finally, the mobility vectors of agglomerated de-identified callers were accumulated respectively in different time segments and the changes in a day were analyzed. Experiments show that, within the scale of180km, the average deviations of bearing angle and distance calculated with the new coordinates are0.037degrees and0.102%, compared with those calculated with Mercator projection and Haversine formula. The new coordinates can maintain the distance and bearing between points well. ACHCT follows a power-law distribution and has a strong correlation with the travel distance. The changes of the crowd’s mobility vector of show the tidal phenomenon of the crowd’s travel and give a new sight to discover the correlation between areas where users reside and those nearby. At the same time, based on CDR we studied the mobility pattern of Beijing residents and found them living a busier life than those in Portugal and Ivory Coast.Land use classification is an essential issue in urban planning. Currently the land use can be distinguished according to the physical characteristics and community features, while the former can be done by remote sensing and image processing technology. Community features has rarely been applied to the land classification. By creating time series of mobile data, TAO PEI, etc. from MIT had done a cluster analysis on change patterns of the amount of call in order to classify the Singapore regional land use, which proposed a new perspective for understanding the land use. Inspired by their work, this paper established user mobility feature models and comparatively analyzed the models’ performance in different functional areas, and then summed up the characteristics of users’ mobility in a particular functional area. The findings provide the basis for recognizing urban functional areas based on user mobility and understanding the land use.Finally, the article verified the conjecture that "the movement trajectory of a short time interval, a short distance and a special direction can better match to the roads, by testing the match rate to urban roads of trajectories of different distances, time intervals and directions. The finding provides a basis for identifying the road network based on trajectories described by CDRs.