Research And Application Of Radar Data Processing Algorithm Based On Adaptive DBSCAN

When the target is tracked by modern radar,the corresponding target track can be generated only after radar signal processing and radar data processing.With the increasing number of radar tracking targets and the irregularity of clutter and the trend of radar networking,a large number of point data may appear in radar admittance,which brings great processing load to radar data processing,resulting in a decrease in radar tracking efficiency and low track quality.With the project ‘Based on the resolution of the angle(azimuth,pitch angle)of threecoordinate air traffic control primary radar' as the background,the following work is done around the optimization of the radar data processing algorithm:(1)The radar target tracking filter is deeply studied,and summary of the advantages and disadvantages of several typical algorithms in the field of radar target tracking algorithms and the scope of application(2)The DBSCAN clustering algorithm and the adaptive DBSCAN algorithm based on the dynamic selection of clustering parameters are deeply studied.(3)Starting from the point of view of reducing the candidate points of input to radar data processing system and using the characteristics of point measurement data as time series data,a radar data processing algorithm based on adaptive DBSCAN,that is,DBSCANKALMAN hybrid algorithm,is proposed.The algorithm consists of two parts: point trace data refinement processing stage and target tracking stage.Min Pts is determined statically according to the flight characteristics of radar target tracking target,Eps is determined dynamically according to the data distribution of different time slices.And through the experiment of MATLAB,after analyzing the experimental results,it is found that the real target points that are incorrectly marked as noise points account for 7.8%,the cases that are correctly marked as noise points account for 92.2%,and the cases that are correctly marked as target points account for 85.7%,the error was marked as the target point accounted for 14.3%.Experiments verify that the hybrid algorithm is effective on the real data set admitted at an airport in Nanchong,that is,he quality of track tracked is obviously improved,the short track is basically disappeared,the number of false tracks is reduced,and the track bifurcation is improved.(4)DBSCAN-KALMAN hybrid algorithm is applied to the project of "softwareization of primary radar tracking system of three-coordinate air traffic control primary radar",and the system is deployed in an open area of Nanchong airport to track the actual target and observe the tracking effect.It can be proved that the DBSCAN-KALMAN hybrid algorithm is effective in real scene,that is,the large area of discrete noise points is reduced,The decrease in the number of false track,tracking quality improvement,tracking efficiency.