一种雷达高分辨距离像模板库建立方法  

A Method of HRRP Template Library Creating

作　　者：王奇 管志强 饶起 籍林峰 

机构地区：[1]中国船舶重工集团第724研究所,南京210015

出　　处：《科学技术与工程》2015年第5期267-271,共5页Science Technology and Engineering

摘　　要：对于非合作目标而言,预警雷达单航路数据的数据率低,能够提供的信息有限,无法达到建立完备模板库要求。针对这一问题,提出了一种将多条航路数据融合的模板库建立方法。首先对雷达高分辨距离像(HRRP)进行预处理,然后对单航路数据分帧建模,最后给出一种改进Jensen-Shannon divergence(JSD)计算子帧统计分布的相似性,将多条航路数据的分帧结果合并获得模板库。实验结果表明,提出的方法能够建立冗余度较低且包含全角域信息的模板库,并且具有较强的在线学习能力。For non-cooperative target,single route data of early warning radar has a low data rate,and cannot provide enough information to create a complete template library. To solve this problem,a new method is presented to create a complete template library via integrating multiple routes data. Firstly,the Radar High Range Resolution Profiles（ HRRP） is preprocessed. Then every single route data is frame segmented and modeled. Finally,an improved JSD（ Jensen-Shannon divergence） is used to measure the similarity of statistical distribution in order to integrate multiple routes data and get the template library. Experimental results show that the proposed method can build a low redundancy template library which contains full angular domain information,and has a strong online learning ability.

关 键 词：雷达目标识别 高分辨距离像 模板库建立 Jensen-Shannon散度 

分 类 号：TN959.1[电子电信—信号与信息处理]