Research On Track Analysis Method Based On Radar Data

Track cluster analysis is the core problem of track data analysis,through clustering algorithm for clustering analysis of two radar data in track and extract the center track based on the cluster,extracting center track on the basis of clustering,that is,the prevailing traffic flow,the analysis to extract the prevailing traffic flow was applied to improve the terminal area control procedure.This paper starts with the historical data of the flight track,analysis of the characteristics of track data,abnormal data processing and feature extraction,then clustering analysis was carried out on the aircraft flight arrival and departure track data using different clustering analysis,an example is given at last,clustering results show the average track,the method is feasible and effective.In this paper,the factors affecting the clustering effect are analyzed,and a new route clustering model is set up and a new clustering method is proposed.The main part of the thesis is as follows:1.the traditional clustering algorithm can only deal with a small number of data sources,so that the clustering effect can not truly reflect the macro characteristics of the data flow.A new model based on the database to solve the problem of cluster simulation is proposed.The experimental results show that the time time of the track clustering simulation is reduced from the hour level to the second level,and the track distribution is obtained.The characteristics and rules of the aircraft are clear,which proves that the new model is feasible and advantageous for extracting macroscopical data from massive data.2.the traditional clustering algorithm only considers the clustering of the three-dimensional coordinates of the track point in the whole cluster analysis process,without considering the influence of the flight direction change and the high descent value on the clustering results and the lack of time information in the clustering process.In this paper,the track clustering based on the LOFC time window segmentation algorithm is proposed.In the class study,two radar data are selected as the simulation analysis object.The simulation results verify the feasibility and superiority of the new algorithm for the cluster analysis of track points and the identification and elimination of outliers.3.in view of the existing problems of the center track extraction method in the current track clustering analysis,In this paper,the extraction and analysis method of central trackbased on characteristic track simplification model is proposed,which effectively solves the problem of missing turning points which often occurs when processing center track extraction problem.It is proved that the new algorithm is effective and feasible for the extraction of the center track of processing center.