J4 ›› 2011, Vol. 38 ›› Issue (6): 22-29.doi: 10.3969/j.issn.1001-2400.2011.06.004

• 研究论文 • 上一篇    下一篇

雷达高分辨距离像分帧新方法

王鹏辉;杜兰;刘宏伟;李彦兵;吴兆平   

  1. (西安电子科技大学 雷达信号处理国家重点实验室,陕西 西安  710071)
  • 收稿日期:2010-09-02 出版日期:2011-12-20 发布日期:2011-11-29
  • 通讯作者: 王鹏辉
  • 作者简介:王鹏辉(1984-),男,西安电子科技大学博士研究生,E-mail: wangpenghui@mail.xidian.edu.cn
  • 基金资助:

    国家自然科学基金资助项目(60901067,61001212);新世纪优秀人才支持计划资助项目(NCET-09-0630);长江学者和创新团队发展计划资助项目(IRT0954);国家部委预研基金和中央高校基本科研业务费专项基金联合资助项目

New frame segmentation method for radar HRRPs

WANG Penghui;DU Lan;LIU Hongwei;LI Yanbing;WU Zhaoping   

  1. (National Key Lab. of Radar Signal Processing, Xidian Univ., Xi'an  710071, China)
  • Received:2010-09-02 Online:2011-12-20 Published:2011-11-29
  • Contact: WANG Penghui

摘要:

针对雷达高分辨距离像的姿态敏感性问题,提出了一种新的分帧方法.首先分析了距离像频谱幅度的统计特性并假设其服从联合高斯分布,之后使用概率主分量分析模型对各个子帧内频谱幅度建模并用期望最大化算法求取模型参数,最后依次合并Kullback-Leibler距离最小的相邻子帧来完成分帧.实测数据的分帧结果与识别率均表明了该分帧方法的有效性.

关键词: 雷达目标识别, 高分辨距离像, 目标姿态敏感性, 概率主分量分析, Kullback-Leibler距离

Abstract:

To deal with the target-aspect sensitivity of radar High Resolution Range Profiles (HRRP), a novel frame segmentation method is proposed. Firstly, the statistical characteristic of HRRPs' frequency spectrum amplitudes is analyzed and assumed to be joint Gaussian distribution. Then Probabilistic Principle Component Analysis (PPCA) is employed to model each sub-frame and the Expectation Maximization (EM) algorithm is adopted to estimate the model parameters. Finally, the adjacent sub-frames which have the minimum Kullback-Leibler (KL) distance are merged iteratively to finish the frame segmentation. Both segmentation results and recognition performance of measured data verify the effectiveness of this method.

Key words: radar target recognition, high resolution range profile(HRRP), target-aspect sensitivity, probabilistic principle component analysis(PPCA), Kullback-Leibler distance

中图分类号: 

  • TN959.1+7
Baidu
map