基于非参数化概率密度估计的雷达目标识别

赵峰; 张军英; 刘敬; 梁军利

doi:10.3724/SP.J.1146.2006.01925

基于非参数化概率密度估计的雷达目标识别

doi: 10.3724/SP.J.1146.2006.01925

基金项目:

国家自然科学基金(60371044，60574039)和国家部级基金资助课题

计量
- 文章访问数: 3748
- HTML全文浏览量: 95
- PDF下载量: 1093
- 被引次数: 0
出版历程
- 收稿日期: 2006-12-04
- 修回日期: 2007-07-19
- 刊出日期: 2008-07-19

Radar Target Recognition Based on Nonparametric Density Estimation

摘要

摘要: 针对雷达目标识别中，参数化方法估计高分辨距离像的概率密度存在的模型失配问题，提出一种非参数化方法基于累计量的随机学习算法，估计距离像的概率密度。该算法运用多层感知器估计训练样本的分布函数，然后求导得到概率密度。该算法不仅能全面、精确地估计概率密度，而且回避了许多其他非参数方法面临的窗宽敏感性问题。基于外场实测数据的实验证明了该文方法的有效性。
- 雷达目标识别; 高分辨距离像; 密度估计
Abstract: In order to solve the problem of model mismatch when using parametric approach to estimate the density of High-Resolution Range Profile(HRRP) in radar target recognition, a nonparametric methodStochastic Learning of the Cumulative(SLC) is presented for the density estimation of HRRP. SLC uses a multiplayer network to estimate the distribution function of the training samples and obtains density by taking derivative. SLC not only describes the density function more comprehensive and accurately, but also avoids the problem of being sensitive to window width that many nonparametric approaches may suffer. Experimental results using outfield real data demonstrate the validity of the proposed learning algorithm.

HTML全文

参考文献(1)

Xing M D, Bao Z, and Pei B N. Properties of high-resolutionrange profiles [J].Optical Engineering.2002, 41(2):493-504[2]Willis C. Aircraft recognition using high-resolution radarrange profiles [A]. Proc SPIE Int Soc Opt Eng[C]. Orlando,United States, 2003: 200-209.[3]Du L, Liu H W, and Bao Z. Radar HRRP target recognitionbased on higher order spectra [J].IEEE Trans. on SignalProcess.2005, 53(7):2359-2368[4]时宇, 张贤达. 基于局部双谱的高分辨距离像雷达目标识别[J]. 清华大学学报，2002, 42(3): 407-410.Shi Y and Zhang X D. Local bispectra-based high-resolutionradar target recognition with range profiles [J]. Journal ofTsinghua Univ., 2002, 42(3): 407-410.[5]叶炜. 逆合成孔径雷达运动补偿与成像研究[D]. 西安：西安电子科技大学.[J].1996.Ye W. Study of the inverse synthetic aperture radar imagingand motion compensation [D]. Xian: Xidian Univ.1996,:-[6]刘宏伟, 杜兰, 袁莉. 雷达高分辨距离像目标识别研究进展[J].电子与信息学报.2005, 27(8):1328-1334浏览[7]Kim K T, Seo D K, and Kim H T. Efficient radar targetrecognition using the MUSIC algorithm and invariantfeatures [J].IEEE Trans. on Antennas Propagation.2002,50(3):325-337[8]Webb A R. Gamma mixture models for target recognition [J].Pattern Recognition.2000, 33(12):2045-2054[9]Copsey K and Webb A R. Bayesian Gamma mixture modelapproach to radar target recognition [J].IEEE Trans. onAerospace and Electronic Systems.2003, 39(4):1201-1217[10]Jacobs S P. Automatic target recognition usinghigh-resolution radar range profiles [D]. Washington:Washington University, 1999.[11]Heiden R V and Groen F C A. The box-cox metric for nearestneighbor classification improvement [J].Pattern Recognition.1997, 30(2):273-279[12]Steinberg B D. Microwave Imaging with Large AntennaArrays: Radio Camera Principle and Technique [M]. NewYork: John Wiley and Sons, 1983: 746-852.[13]Duda R O, Hart P E, and Stork D G著. 李宏东, 姚天翔译.模式分类 [M].北京：机械工业出版社, 2003: 132-163.[14]张炤, 张素, 章琛曦. 基于支撑向量机的概率密度估计方法[J].系统仿真学报, 2005, 17(10): 2355-2357.Zhang Z, Zhang S, and Zhang C X. Density estimation basedon support vector machines [J]. Journal of System Simulation,2005, 17(10): 2355-2357.[15]Malik M I and Amir A. Density estimation and randomvariate generation using Multilayer networks [J].IEEE Trans.on Neural Networks.2002, 13(3):497-520[16]Fukunage K. Introduction to Statistic Pattern Recognition[M]. Boston MA: Academic, 1990: 368-424.[17]Skolnik M. Introduction to Radar Systems [M]. New York:McGraw-Hill, 1980: 523-614.[18]Haykin S著. 叶世伟, 史忠植译. 神经网络原理 [M].北京：机械工业出版社, 2004: 110-181.

施引文献

资源附件(0)

访问统计