一种新的雷达高分辨距离像平移不变特征

刘敬; 张军英; 赵峰

doi:10.3724/SP.J.1146.2007.00024

一种新的雷达高分辨距离像平移不变特征

doi: 10.3724/SP.J.1146.2007.00024

基金项目:

国家自然科学基金(60574039，60371044)资助课题

计量
- 文章访问数: 3073
- HTML全文浏览量: 86
- PDF下载量: 880
- 被引次数: 0
出版历程
- 收稿日期: 2007-01-05
- 修回日期: 2007-06-28
- 刊出日期: 2008-08-19

A New Time-Shift Invariant Feature of Radar HRRPs

摘要

摘要: 针对雷达自动目标识别(RATR)中高分辨距离像(HRRP)的平移敏感性，从HRRP中提取出一种新的距离像平移不变特征：距离像幅度谱差分。理论分析的结果表明距离像幅度谱差分特征比距离像幅度谱特征更适合于RATR。采用外场实测数据，分别训练了最短距离分类器和SVM分类器，两种分类器的识别结果均表明，相比幅度谱特征，幅度谱差分特征可提高数据可分性并显著提高识别率。
- 雷达自动目标识别; 幅度谱差分; 平移敏感性; 高分辨距离像
Abstract: Radar High-Resolution Range Profile (HRRP) is very sensitive to time-shift; therefore, HRRP-based Radar Automatic Target Recognition (RATR) requires efficient time-shift invariant features. A new time-shift invariant feature, i.e., amplitude spectrum difference of HRRP, is extracted from HRRP to solve the time-shift sensitivity. The result of theoretical analysis shows that, as a time-shift invariant feature, amplitude spectrum difference is more suitable for HRRP-based RATR than amplitude spectrum is. Shortest distance classifier and Support Vector Machine (SVM) classifier are designed to evaluate the recognition performance. Experimental results for measured data show that, comparing with amplitude spectrum, amplitude spectrum difference improves recognition performance remarkably.

HTML全文

参考文献(1)

Webb A R. Gamma mixture models for target recognition[J].Pattern Recognition.2000, 33(12):2045-2054[2]Copsey K and Webb A R. Bayesian Gamma mixture modelapproach to radar target recognition[J]. IEEE Trans. on AES,2003, 39(4): 1201-1217.[3]Jacobs S P and Osollivan J A. Automatic target recognitionusing high-resolution radar range profiles [D]. WashingtonUniversity, 1999.[4]Van der Heiden R and Groen F C A. The Box-cox metric fornearest neighbour classification improvement [J]. PatternRecognition, 1997, 30(2): 273-279.[5]Xing Mengdao and Bao Zheng. The properties of rangeprofiles of aircraft[J]. Chinese Journal of Electronics, 2002,11(1): 1-6.[6]Rothwell E J, Cheng K M, and Nyquist D P. Anadaptive-window-width short time Fourier transform forvisualization of radar target substructure resonances [J].IEEE Trans. on Antenna and Propagation.1998, 46(9):1393-1395[7]赵群. 基于高分辨一维距离像的雷达目标识别与检测.[博士论文], 西安: 西安电子科技大学, 1995.Zhao Qun. Radar target recognition and detection based onhigh resolution one dimensional range profiles.[Ph.D.dissertation], Xian: Xidian university, 1995.[8]时宇，张贤达. 基于局部双谱的高分辨距离像雷达目标识别.清华大学学报(自然科学版)[J], 2002, 42(3): 407-410.Shi Yu and Zhang Xianda. Lacal bispectra-based highresolutionradar target recognition with range profiles[J].Journal of TSinghua University (Science and Technology),2002, 42(3): 407-410.[9]Pei Bingnan and Bao Zheng. Logarithm bispectrum basedapproach to high radar range profile for automatic targetrecognition[J].Pattern Recognition.2002, 35(11):2643-2651[10]廖学军. 基于高分辨距离像的雷达目标识别. [博士论文], 西安: 西安电子科技大学, 1999.Liao Xuejun. Radar target recognition base on high resolutionrange profiles. [Ph.D.dissertation], Xian: Xidian university,1999.[11]Zyweck A and Bogner R E. Radar target classification ofcommercial aircraft. IEEE Trans. on AES, 1996, 32(2): 598-606.[12]Zhang Junying, Wang Yue Joseph, Khan J, and Clarke R.Gene selection in class space for molecular classification ofcancer[J].Science in China Series F: Information Sciences.2004, 47(3):301-314

施引文献

资源附件(0)

访问统计