| In recent years,with the continuous acceleration of China's urbanization process and the increasing diversification of residents' transportation activities,it has brought enormous challenges to transportation planning and demand management.By accurately grasping the current situation of urban traffic,analyzing the internal operating mechanism and accurately recognizing people's travel,we can scientifically and rationally carry out traffic planning and design,and guide the orderly and normal development of the traffic system.However,the collection of traditional residents' travel information mainly relies on manual questionnaires.For a long time,the subject's subjective memories are more,the survey cost is higher,the organization is difficult to implement,and the quality and accuracy of the data are difficult to achieve precision planning needs.With the boom of big data and the advent of the Internet era,mobile phone data as a mobile probe can better record the travel situation of residents,and the coverage is wide.The cost of collecting large sample data is low,and the data can be dynamically updated in real time.With the development and popularization of 4G-LTE mobile communication networks,the accuracy of mobile phone data positioning and tracking investigations have been greatly improved.Although many cities have tried and implemented practices,the accuracy of the analysis results has been a difficult question to answer.At present,the research on mobile phone signaling data mostly focuses on data mining,but the signaling data is applied to traffic data collection.Channel fading is an important factor affecting the quality of traffic information such as travel endpoints and OD,which is directly related to traffic information.The accuracy of the extraction needs to be studied in depth.This paper analyzes the effect of signaling endpoint data and travel OD on signaling data under different channel fading.Based on the mastery of the basic principles of wireless communication and the statistics of the actual signaling data communication event interval,a traffic-communication simulation platform is constructed.By designing traffic combination travel simulation experiments and data preprocessing in different scenarios,the signaling data under different channel fading is obtained.According to the data characteristics and research purposes,the density-based spatiotemporal clustering algorithm(ST-DBSCAN)is used to extract the travel endpoint information of travel data under different channel fading,and the feasibility of the algorithm is verified by taking the measured data as an example.The identified travel endpoints are mapped into the traffic cell,and the identified position coordinates are compared with the traffic cell area by using the corresponding relationship between the points and the faces,and the edge point and internal point identification accuracy in the urban central area and the urban suburban area are found.There are differences,the error rate of the cell edge point is above 90%,and the error rate of the internal point of the city center is about 8%,and the probability of the point jump within the urban suburb is between 4%and 18%.In order to further explore the distance error between the identification point and the real point,using the advantage of the traffic-communication simulation platform,the identified travel end point is compared with the set standard point,and the distance deviation value in the central area of the city is found to be in the range of 100-200m.Fluctuation,the distance deviation value is the smallest when the shadow fading is 5db.The urban suburbs fluctuate within the range of 450-650m,and satisfy the law that the distance deviation value first decreases and then increases as the communication disturbance increases,and the distance deviation value is the smallest when the shadow fading is 6db.The conclusion provides a certain degree of reference value and significance for the extraction of mobile phone signaling data OD precision and travel endpoint error. |
PDF Full Text Request