基于相关系数的雷达高分辨距离像分帧方法

A Frame Segmentation Method for Radar HRRPs Based on Correlation Coefficient

高分辨距离像(HRRP)的方位敏感性造成HRRP数据是很长的统计特性连续变化的序列。从HRRP序列的统计特性出发,提出一种基于相关系数的HRRP分帧方法。外场实测数据的分帧结果表明,该方法可准确划分HRRP统计特性的变化,分帧结果与实际飞行轨迹符合;且采用时域模板匹配法(TMM)、谱域最短距离分类器和支撑向量机(SVM)分类器的识别结果均表明,相比目前基于散射点模型的均匀分帧方法,本文方法可有效提高识别率。

Target-aspect sensitivity makes High Resolution Range Profiles （HRRP） become a long sequence with statistical characteristic changing continuously. A novel frame segmentation method based on correlation coefficient is presented according to the statistical characteristic changes of HRRP sequence. Experimental results for measured data show that the presented method can exactly divide the statistical characteristic changes of HRRPs and the resulting frames are coincident with flying trajectory. Template Matching Method （TMM） classifier in time domain, shortest distance classifier and Support Vector Machine （SVM） classifier in frequency domain are used to evaluate recognition performances. Comparing with the current uniform frame segmentation method based on scattering center model, the presented method efficiently improves the recognition rates.