首页 | 本学科首页   官方微博 | 高级检索  
     

一种基于HMM的雷达目标识别方法
引用本文:陈江峰,裴炳南. 一种基于HMM的雷达目标识别方法[J]. 郑州大学学报(理学版), 2003, 35(1): 52-56
作者姓名:陈江峰  裴炳南
作者单位:郑州大学信息工程学院,郑州,450052
基金项目:国家自然科学基金资助项目,编号 6983 10 40 .
摘    要:根据雷达目标散射点的一般模型和特定条件的简化模型;提出用RELAX方法从高分辨雷达回波提取目标散射点分布的位置信息作为目标HRRR的特征向量;利用HRRR对雷达视角敏感这一特点,用隐马尔可夫过程表征多视角雷达回波序列,获得目标距离-方位两维信息,用若干HMM子过程构成的模式链表征一个飞行目标的飞行姿态变化,从而采用基于隐马尔可夫模型的分类器实现目标类属和方位的自动识别和分类,实测数据的计算机仿真结果表明,这一方法的平均识率为99.80%和81.2%。

关 键 词:HMM 雷达目标识别 隐马尔可夫模型 RELAX方法 高分辨距离像 散射点模型 雷达回波序列
文章编号:1671-6841(2003)01-0052-05
修稿时间:2002-04-28

Approach Based on HMM Framework to Identify Radar Target
Chen Jiangfeng,Pei Bingnan. Approach Based on HMM Framework to Identify Radar Target[J]. Journal of Zhengzhou University(Natrual Science Edition), 2003, 35(1): 52-56
Authors:Chen Jiangfeng  Pei Bingnan
Abstract:Based on general models of the radar target scatter and predigested models under the special conditions, an approach is proposed to produce radar vectors from high resolution radar profiles, here the method of vector quantization uses RELAX algorithm. The hidden Markov model is used as the classification algorithm. HMM is used to describe multi-azimuth radar profiles to obtain the distance-azimuth information of radar target through that radar HRRP is sensitive to radar azimuth. The usultsot computer simulation shows that the identification accuracy adapted method is very high. The average identifying rates are 99.80% and 81.20% in different data.
Keywords:target identify  hidden Markov models  RELAX arithmetic  high resolution radar profiles  scatter models
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号