一种基于雷达高分辨距离像的空天时敏目标识别方法

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

针对空天时敏目标识别问题,提出了一种新的针对雷达高分辨距离像(HRRP)的序贯特征提取方法,并设计了一种基于决策树和支持矢量描述(SVDD)方法的多分类器。该方法首先基于时序HRRP估计目标的径向尺寸,利用序贯脉冲积累对尺寸估计结果进行滑窗处理,获取各个窗内径向尺寸的均值、极差、中值以及结尾均值四种统计特征;然后,将得到的四种特征进行拼接,从而获取更加鲁棒的高维特征;最后,使用基于决策树的多分类SVDD方法(Multi-SVDD-DT)对获取的高维特征进行分类。四类飞机的测量数据实验表明,本文所提方法可以提取出目标的稳健特征,能够有效完成空天时敏目标的识别任务。

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.