基于高分辨一维距离像的雷达目标识别方法研究
|
摘要
本文应用现代信号处理技术和现代模式识别技术的方法主要研究了基于高分辨雷达一维距离像的雷达目标识别问题，研究的目的在于降低高分辨雷达目标识别算法的复杂性和提高目标识别的准确性，为高分辨雷达智能信号处理的发展提供研究基础。主要内容如下：
首先对当前基于高分辨一维距离像进行雷达目标识别的技术做了回顾和阐述，指出了雷达目标识别研究过程典型的误区及关键问题。
研究了高分辨雷达目标一维距离像的原理及仿真方法并对高分辨雷达一维距离像的姿态敏感性进行了分析。分析了高分辨雷达杂波的特性，研究了高分辨雷达杂波仿真的经典方法，提出了将神经网络方法应用到高分辨雷达杂波仿真中去，给出了经验迭代方法及自然梯度方法在雷达杂波仿真中的具体应用，并分析了这两种方法较经典方法的改进之处。
在研究了基于高分辨雷达一维距离像的雷达目标三维成像方法的基础上，介绍了基于独立分量分析(ICA)方法的雷达目标三维结构信息的特征提取方法；讨论了这种方法的局限性，给出了基于聚类分析方法和基于量化统计的雷达目标三维成像的改进算法，给出仿真结果并进行比较分析。
最后，总结了本文的研究工作，指出了需要进一步解决的问题。
In this paper, the problem of recognition of radar target using high resolution range profiles is mainly studied by the means of techniques in modern signal process and in modern pattern recognition. The aim is to make the arithmetic of high resolution radar target recognition have less complexity and identify targets more accuracy. In the paper, there are:
At first, it makes a brief introduction to the techniques in recognition of radar target using high resolution range profiles. Some representative misleading areas and key points in the process of the study on recognition of radar target are pointed out.
The principle and simulation method of high resolution range profiles is studied,
and the pose sensibility of range profiles is investigated. After introducing the characteristics and classic simulation method of high resolution radar clutter, a neural network type empirical iteration method and a neural network Trained by natural gradient algorithm is proposed to simulate correlated non-Gaussian radar clutter by zero-memory nonlinearities (ZMNL).
In the paper, a method is presented which gets radar target high dimensional structure information from one dimensional range profiles based on Independent Component Analysis (ICA). Then after discussing the limitations of ICA, two modified algorithms of radar target 3D imaging using clustering analysis and quantization statistical analysis are proposed. The emulation results are presented, analyzed and compared.
In conclusion, the main works of the dissertation are summarized and the future research areas are pointed out.
首先对当前基于高分辨一维距离像进行雷达目标识别的技术做了回顾和阐述，指出了雷达目标识别研究过程典型的误区及关键问题。
研究了高分辨雷达目标一维距离像的原理及仿真方法并对高分辨雷达一维距离像的姿态敏感性进行了分析。分析了高分辨雷达杂波的特性，研究了高分辨雷达杂波仿真的经典方法，提出了将神经网络方法应用到高分辨雷达杂波仿真中去，给出了经验迭代方法及自然梯度方法在雷达杂波仿真中的具体应用，并分析了这两种方法较经典方法的改进之处。
在研究了基于高分辨雷达一维距离像的雷达目标三维成像方法的基础上，介绍了基于独立分量分析(ICA)方法的雷达目标三维结构信息的特征提取方法；讨论了这种方法的局限性，给出了基于聚类分析方法和基于量化统计的雷达目标三维成像的改进算法，给出仿真结果并进行比较分析。
最后，总结了本文的研究工作，指出了需要进一步解决的问题。
In this paper, the problem of recognition of radar target using high resolution range profiles is mainly studied by the means of techniques in modern signal process and in modern pattern recognition. The aim is to make the arithmetic of high resolution radar target recognition have less complexity and identify targets more accuracy. In the paper, there are:
At first, it makes a brief introduction to the techniques in recognition of radar target using high resolution range profiles. Some representative misleading areas and key points in the process of the study on recognition of radar target are pointed out.
The principle and simulation method of high resolution range profiles is studied,
and the pose sensibility of range profiles is investigated. After introducing the characteristics and classic simulation method of high resolution radar clutter, a neural network type empirical iteration method and a neural network Trained by natural gradient algorithm is proposed to simulate correlated non-Gaussian radar clutter by zero-memory nonlinearities (ZMNL).
In the paper, a method is presented which gets radar target high dimensional structure information from one dimensional range profiles based on Independent Component Analysis (ICA). Then after discussing the limitations of ICA, two modified algorithms of radar target 3D imaging using clustering analysis and quantization statistical analysis are proposed. The emulation results are presented, analyzed and compared.
In conclusion, the main works of the dissertation are summarized and the future research areas are pointed out.
引文
[1] 周代英．雷达目标—维距离像识别研究．电子科技大学博士论文，1998
[2] D. R. Wehner. High Resolution Radar. Norwood. MA: Artech House. Inc., 1987
[3] 黄德双．高分辨率雷达智能信号处理技术．北京：机械工业出版社，2001
[4] 王晓丹，王积勤．雷达目标识别技术综述．现代雷达，2003，25(5)：22-26
[5] Bell M B, Gyubbs R A. J EM modeling and measurement for radar target identification. IEEE Trans on AES, 1993, 29(1):12-13
[6] 黄为倬．目标识别与逆合成孔径雷达．现代雷达，1988，10(6)：108-122
[7] Blaricum ML Van, Mittra R. A technique for extracting the poles and residues of a system directly from its transient response. IEEE Trans. on AP, 1975, 23(6):77-781
[8] 廖学军．基于高分辨距离像的雷达目标识别．西安：西安电子科技大学博士论文，1999
[9] 张仲明．基于高分辨—维距离像的雷达目标识别方法研究．国防科学技术大学，2004．
[10] 闫锦，黄培康．高距离分辨像雷达目标识别．航天电子对抗，2004(2)：36-41
[11] H. Li,S. Yang. Using range profiles as feature vectors to identify aerospace object. IEEE Trans. on AP, 1993,41(3):261-280
[12] S. Hudos, Psalltis. Correlation filters for aircraft identification from range profiles. IEEE Trans. on AES,1993,29(3):741-748
[13] 苗雨，姜文利等．基于高分辨距离像的雷达目标识别特征分析．国防科技大学学报，2001(5)：93-97
[14] 边肇祺，张学工等编著．模式识别(第二版)[M]．北京：清华大学出版社，2000．
[15] Lindsay I Smith. A tutorial on Principal Components Analysis[J]. February 26, 2002.
[16] Anil K. Jain, Robert P.W. Duin, Jianchang Mao. Statistical Pattern Recognition: A Review. IEEE Transactions on Pattern Analysis and Machine Inteligence [J], 2000.
[17] A. K. Shaw, R. Vashist, et.al. HRR-ART using eigen-template with noisy observation in unknown target scenario [J]. SPIE, 2000, 4053:467-478
[18] A. Zyweck, R.E. Bogner. Radar target classification of commercial aircraft [J]. IEEE Trans. on AES, 1996,32(2): 589-606
[19] R. Carriere, R.L. Moses.High resolution radar modeling using a modified Prony estimator [J]. IEEE Trans. on AP, 1992, 40:13-18
[20] 姜卫东．光学区雷达目标结构成像的理论及其在雷达目标识别中的应用．国防科技大学博士论 文，2000．
[21] K.T. Kim, D. K. Seo, et.al. Efficient radar target recognition using the MUSIC algorithm and invariant features [J]. IEEE Trans. on AP, 2002,50(3):325-337
[22] 刘静，李兴国，顾玉辉．频率步进雷达关键技术的研究．制导与引信，2003，24(4)．
[23] 古鲁[美]等著．电磁场与电磁波．机械工业出版社，2005
[24] 黄培康．雷达目标特性．电子工业出版社，2005
[25] Hurst M P, Mittra R. Scattering center analysis prony's method. IEEE Trans. On A. P., 1987, 35(3):986-988
[26] Moses R L, Carriere R. Parametric modeling of radar targets using canonical scattering centers. Report 719269-13. Dec., 1988, ESL, The Ohio State University.
[27] 杜兰，保铮，邢盂道．飞机目标的雷达—维距离像特性研究，西安电子科技大学学报．2001，28(增刊)：14-19
[28] X.Frank Xu. Nonlinear analysis of heterogeneous materials with gmc technique. 16th ASCE Engineering Mechanics Conference, University of Washington, Seattle, July 16-18,2003.
[29] Wen Wei, Jerry M. Mende, A Fuzzy Logic Method for Modulation Classification in Nonideal Environments, IEEE TRANSACTIONS ON FUZZY SYSTEMS, VOL. 7, NO. 3, JUNE 1999
[30] 刘年宝，周铨兴，钱崇智等．低信噪比下的杂波建图研究．上海航天，1994，6
[31] 张胜付，赵蕙昌，李官发等．海杂波的数学模型及其计算机模拟．南京理工大学学报，1995，19(2)
[32] 陈彦辉，谢维信．地海杂波的随机分形模型．电子科技大学学报，2000，29(3)：239-242
[33] 刘灵，吴曼青，洪一．基于ZMNL的K杂波建模与仿真．合肥工业大学学报，2006，29(5)：544-563
[34] 吕雁，史林，杨万海．SIRP法相干相关K分布雷达杂波的建模与仿真．现代雷达，2002，24(2)：13-16
[35] 许稼，卢凌，颜南霞．一种基于球不变随机过程的雷达K分布杂波模拟方法．武汉交通科技大学学报，2000，24(5)：469-472
[36] Marier L James. Correlated K-distributed clutter generation for radar detection and track. IEEE Trans. On AES, 1995, 31(2): 568-580.
[37] 朱灿焰，何佩琨，毛二可．雷达杂波相关功率谱特性的AR模型及其模拟．华东交通大学学报，1998，15(3)：50-55
[38] 刘宏伟．雷达目标识别技术．雷达信号处理国防科技重点实验室讲座 PPT，2004
[39] 李晓辉．基于非线性方法的雷达目标识别研究．国防科技大学博士论文，2004
[40] 张恂．雷达目标的高分辨参数建模及其在自动目标识别中的应用．长沙：国防科技大学博士论 文，1997．
[41] Erkki Ojia, A class of neural networks for independent component analysis. IEEE Trans. Neural networks, 8(3):486-504,1997
[42] Bell Anthony J, Sejnowski Terrence J. An information-maximization approach to blind separation and blind deconvolution. Neural Computation [J], 1995,7(6):1004-1034.
[43] P.O. Hoyer and A. Hyvarinen. Independent Component Analysis Applied to Feature Extraction from Colour and Stereo Images[J]. Network: Computation in Neural Systems, 11 (3): 191-210, 2000.
[44] J. Hurri, A. Hyv"arinen, J. Karhunen, and E. Oja. Image feature extraction using independent component analysis[J]. In Proc. NORSIG'96, pages 475-478, Espoo,Finland, 1996.
[45] P. O. Hoyer and A. Hyv"arinen. Feature extraction from colour and stereo images using ICA[J]. In Proc. Int. Joint Conf. on Neural Networks (IJCNN2000), Como,Italy, 2000.
[46] A. Hyv"arinen and P. O. Hoyer. Emergence of phase and shift invariant features by decomposition of natural images into independent feature subspaces[J]. Neural Computation, 12(7):1705-1720, 2000.
[47] R. Vig'ario, A. Hyv"arinen, and E. Oja. ICA fixed-point algorithm in extraction of artifacts from EEG[J]. In Proc. NORSIG'96, pages 383-386, Espoo, Finland,1996.
[48] Space Time Iterative Receivers for Wireless Communications - Master's Thesis. http://www.ittc.ku.edu/～rvc/html/thesis.html#Introduction
[49] Alan Oursland, Jdah De Paula, Nasim Mahmood, Case Studies of Independent Component Analysis, https://webspace.utexas.edu/deoaula/work/cs383c/index.html.
[50] 孙即祥等．现代模式识别[M]．长沙：国防科技大学出版社，2001．
[51] 涂志江，刘国岁．基于多特征和多阶段的雷达距离像识别．兵工学报，2000，21(3)：233-236
[52] 姜卫东，陈曾平，庄钊文，郭桂蓉．杂波环境下雷达目标频域响应的散射中心估计方法．红外与毫米波学报，2001，20(2)：111-116
[53] A．费利那，F．A．斯塔德．杂波环境中的滤波．雷达数据处理，内部参考，2004．8：227
[2] D. R. Wehner. High Resolution Radar. Norwood. MA: Artech House. Inc., 1987
[3] 黄德双．高分辨率雷达智能信号处理技术．北京：机械工业出版社，2001
[4] 王晓丹，王积勤．雷达目标识别技术综述．现代雷达，2003，25(5)：22-26
[5] Bell M B, Gyubbs R A. J EM modeling and measurement for radar target identification. IEEE Trans on AES, 1993, 29(1):12-13
[6] 黄为倬．目标识别与逆合成孔径雷达．现代雷达，1988，10(6)：108-122
[7] Blaricum ML Van, Mittra R. A technique for extracting the poles and residues of a system directly from its transient response. IEEE Trans. on AP, 1975, 23(6):77-781
[8] 廖学军．基于高分辨距离像的雷达目标识别．西安：西安电子科技大学博士论文，1999
[9] 张仲明．基于高分辨—维距离像的雷达目标识别方法研究．国防科学技术大学，2004．
[10] 闫锦，黄培康．高距离分辨像雷达目标识别．航天电子对抗，2004(2)：36-41
[11] H. Li,S. Yang. Using range profiles as feature vectors to identify aerospace object. IEEE Trans. on AP, 1993,41(3):261-280
[12] S. Hudos, Psalltis. Correlation filters for aircraft identification from range profiles. IEEE Trans. on AES,1993,29(3):741-748
[13] 苗雨，姜文利等．基于高分辨距离像的雷达目标识别特征分析．国防科技大学学报，2001(5)：93-97
[14] 边肇祺，张学工等编著．模式识别(第二版)[M]．北京：清华大学出版社，2000．
[15] Lindsay I Smith. A tutorial on Principal Components Analysis[J]. February 26, 2002.
[16] Anil K. Jain, Robert P.W. Duin, Jianchang Mao. Statistical Pattern Recognition: A Review. IEEE Transactions on Pattern Analysis and Machine Inteligence [J], 2000.
[17] A. K. Shaw, R. Vashist, et.al. HRR-ART using eigen-template with noisy observation in unknown target scenario [J]. SPIE, 2000, 4053:467-478
[18] A. Zyweck, R.E. Bogner. Radar target classification of commercial aircraft [J]. IEEE Trans. on AES, 1996,32(2): 589-606
[19] R. Carriere, R.L. Moses.High resolution radar modeling using a modified Prony estimator [J]. IEEE Trans. on AP, 1992, 40:13-18
[20] 姜卫东．光学区雷达目标结构成像的理论及其在雷达目标识别中的应用．国防科技大学博士论 文，2000．
[21] K.T. Kim, D. K. Seo, et.al. Efficient radar target recognition using the MUSIC algorithm and invariant features [J]. IEEE Trans. on AP, 2002,50(3):325-337
[22] 刘静，李兴国，顾玉辉．频率步进雷达关键技术的研究．制导与引信，2003，24(4)．
[23] 古鲁[美]等著．电磁场与电磁波．机械工业出版社，2005
[24] 黄培康．雷达目标特性．电子工业出版社，2005
[25] Hurst M P, Mittra R. Scattering center analysis prony's method. IEEE Trans. On A. P., 1987, 35(3):986-988
[26] Moses R L, Carriere R. Parametric modeling of radar targets using canonical scattering centers. Report 719269-13. Dec., 1988, ESL, The Ohio State University.
[27] 杜兰，保铮，邢盂道．飞机目标的雷达—维距离像特性研究，西安电子科技大学学报．2001，28(增刊)：14-19
[28] X.Frank Xu. Nonlinear analysis of heterogeneous materials with gmc technique. 16th ASCE Engineering Mechanics Conference, University of Washington, Seattle, July 16-18,2003.
[29] Wen Wei, Jerry M. Mende, A Fuzzy Logic Method for Modulation Classification in Nonideal Environments, IEEE TRANSACTIONS ON FUZZY SYSTEMS, VOL. 7, NO. 3, JUNE 1999
[30] 刘年宝，周铨兴，钱崇智等．低信噪比下的杂波建图研究．上海航天，1994，6
[31] 张胜付，赵蕙昌，李官发等．海杂波的数学模型及其计算机模拟．南京理工大学学报，1995，19(2)
[32] 陈彦辉，谢维信．地海杂波的随机分形模型．电子科技大学学报，2000，29(3)：239-242
[33] 刘灵，吴曼青，洪一．基于ZMNL的K杂波建模与仿真．合肥工业大学学报，2006，29(5)：544-563
[34] 吕雁，史林，杨万海．SIRP法相干相关K分布雷达杂波的建模与仿真．现代雷达，2002，24(2)：13-16
[35] 许稼，卢凌，颜南霞．一种基于球不变随机过程的雷达K分布杂波模拟方法．武汉交通科技大学学报，2000，24(5)：469-472
[36] Marier L James. Correlated K-distributed clutter generation for radar detection and track. IEEE Trans. On AES, 1995, 31(2): 568-580.
[37] 朱灿焰，何佩琨，毛二可．雷达杂波相关功率谱特性的AR模型及其模拟．华东交通大学学报，1998，15(3)：50-55
[38] 刘宏伟．雷达目标识别技术．雷达信号处理国防科技重点实验室讲座 PPT，2004
[39] 李晓辉．基于非线性方法的雷达目标识别研究．国防科技大学博士论文，2004
[40] 张恂．雷达目标的高分辨参数建模及其在自动目标识别中的应用．长沙：国防科技大学博士论 文，1997．
[41] Erkki Ojia, A class of neural networks for independent component analysis. IEEE Trans. Neural networks, 8(3):486-504,1997
[42] Bell Anthony J, Sejnowski Terrence J. An information-maximization approach to blind separation and blind deconvolution. Neural Computation [J], 1995,7(6):1004-1034.
[43] P.O. Hoyer and A. Hyvarinen. Independent Component Analysis Applied to Feature Extraction from Colour and Stereo Images[J]. Network: Computation in Neural Systems, 11 (3): 191-210, 2000.
[44] J. Hurri, A. Hyv"arinen, J. Karhunen, and E. Oja. Image feature extraction using independent component analysis[J]. In Proc. NORSIG'96, pages 475-478, Espoo,Finland, 1996.
[45] P. O. Hoyer and A. Hyv"arinen. Feature extraction from colour and stereo images using ICA[J]. In Proc. Int. Joint Conf. on Neural Networks (IJCNN2000), Como,Italy, 2000.
[46] A. Hyv"arinen and P. O. Hoyer. Emergence of phase and shift invariant features by decomposition of natural images into independent feature subspaces[J]. Neural Computation, 12(7):1705-1720, 2000.
[47] R. Vig'ario, A. Hyv"arinen, and E. Oja. ICA fixed-point algorithm in extraction of artifacts from EEG[J]. In Proc. NORSIG'96, pages 383-386, Espoo, Finland,1996.
[48] Space Time Iterative Receivers for Wireless Communications - Master's Thesis. http://www.ittc.ku.edu/～rvc/html/thesis.html#Introduction
[49] Alan Oursland, Jdah De Paula, Nasim Mahmood, Case Studies of Independent Component Analysis, https://webspace.utexas.edu/deoaula/work/cs383c/index.html.
[50] 孙即祥等．现代模式识别[M]．长沙：国防科技大学出版社，2001．
[51] 涂志江，刘国岁．基于多特征和多阶段的雷达距离像识别．兵工学报，2000，21(3)：233-236
[52] 姜卫东，陈曾平，庄钊文，郭桂蓉．杂波环境下雷达目标频域响应的散射中心估计方法．红外与毫米波学报，2001，20(2)：111-116
[53] A．费利那，F．A．斯塔德．杂波环境中的滤波．雷达数据处理，内部参考，2004．8：227