Target detection trace condensation method based on peak value clustering

Abstract

The invention discloses a target detection trace condensation method based on peak value clustering, relates to the technical field of signal and information processing, and considers the characteristic that data points of the same category in a CFAR detection trace result are gathered towards peak value points. Meanwhile, the algorithm also considers the characteristic that different types of data points in a CFAR detection trace result have obvious differences. The algorithm is based on a peak clustering plot condensation algorithm for automatically searching peak data points, the Euclidean distance of all collected data is calculated according to distance-Doppler two-dimensional information by sampling plot data of a CFAR detection result, the peak data points of each category are searched by using amplitude information, and clustering is carried out; the method can overcome the difficulty that a plurality of targets which are adjacent to each other and have large plot data amplitude and density distribution differences are difficult to correctly classify, and can finish the plot condensation process at a higher speed on the basis of not increasing hardware structures and expenditures.

Description

technical field [0001] The invention belongs to the technical field of signal and information processing, and in particular relates to a method for aggregating target detection points and traces based on peak clustering. Background technique [0002] Such as figure 1 The typical radar signal processing flow is shown. The main task of the radar system is to determine whether the target exists, and to accurately detect and track multiple batches of targets. With the increase of the radar range and the improvement of the radar resolution accuracy, the radar The amount of trace data collected also increases and trace expansion occurs for the same target. In order to reduce the amount of data to be processed and ensure the accuracy and reliability of track information, the point-trace aggregation processing system with good performance becomes Particularly important. The radar signal processing flow includes multiple links. After the constant false alarm detection (CFAR) proces...

AI Technical Summary

Problems solved by technology

However, due to the global parameters set by this method, it is difficult to correctly classify multiple target data points with different density distributions and relatively large amplitude differences. Although the modified DBSCAN point trace aggregation algorithm can solve the above problems, due to This algorithm has too many iterations and runs slowly. Therefore, how to find ideas from a new direction to improve the effect of point trace aggregation is a huge challenge.

Images