A Recognition Method of Aerospace Time-Sensitive Targets Based on Radar High Resolution Range Profile

Abstract

Aiming at the problem of aerosapce time-sensitive target recognition, a new sequential feature extraction method for radar high resolution range profile (HRRP) is proposed, and a multi classifier based on decision tree and support vector description (SVDD) method is designed. Firstly, the method estimates the radial size of the target based on time-series HRRP, uses sequential pulse accumulation to perform sliding window processing on the size estimation results, and obtains the four statistical characteristics of mean, range, median, and end mean of the radial size in each window. Then, The four types of features are spliced to obtain more robust high-dimensional features. Finally, the proposed multi classification SVDD method based on decision tree (Multi-SVDD-DT) is used to classify the obtained high-dimensional features. The simulation data experiments of four types of aircraft show that the proposed method can extract the robust features of the target, and can effectively complete the recognition task of aerospace time-sensitive targets.

Keywords

• |time-sensitive target|high resolution one-dimensional range profile|radical size estimation|multi-classifier design|decision tree|target recognition|radar