水下瞬态信号特征提取与多分类器融合
|
摘要
水下瞬态信号的特征提取是空投目标分类的重要环节。回顾历史,用于特征提取的信号分析方法有很多:傅立叶变换、短时傅立叶变换、Wigner-Ville变换、小波变换等等。然而它们应用于水下目标特征提取时也存在各自不同的问题。傅立叶分析要求数据必须具有线性、周期性和平稳性的特点;短时傅立叶变换窗口的大小是固定不变的;Wigner-Ville变换会产生严重的交叉项;小波变换的时频窗面积不变,只是形状变化。因此,本文采用希尔伯特-黄变换方法,它是一种更适用于分析非线性、非平稳数据的时频分析方法。
在目标分类领域中,人工神经网络分类器应用最为广泛。利用信号的不同特征或不同结构的人工神经网络分类器可以得到不同的分类结果,往往可以利用这些结果之间的互补性来改善系统的分类效果。多分类器输出向量加权投票算法是一种有效的多分类器融合算法,它具有算法简单、便于处理实际问题的优点。因此,本文采用此方法进行多分类器融合。
本文首先回顾了几种处理非平稳数据的常见方法,并分析了它们的优缺点,之后介绍了希尔伯特-黄变换方法。通过对空投目标入水声信号构成特点的研究,从击水声脉冲信号、“寂静”区间和气泡脉动信号三方面对空投目标入水声信号进行了理论分析和实测数据的特征提取研究。给出了常见人工神经网络的结构特点以及基本原理,并应用单个人工神经网络分类器对空投目标进行分类,分类结果表明不同分类器之间存在互补性。在本文的最后论述了多分类器融合算法,并给出基于多分类器融合算法的目标分类结果,证明了多分类器融合算法在解决目标分类问题时具有优越性。
Feature extraction of underwater transient signal is the key of the airdrop targets classification. Historically, there are many methods of signal analysis which are applied to extract signal characteristics, such as Fourier Transform, Short Time Fourier Transform, Wigner-Ville Transform, Wavelet Transform, and so on. But each of them has difficulties when they are used in the extraction of characteristics of underwater targets. The difficulty of Fourier Transform is that the signal must be linear, strictly periodic and stationary; the difficulty of Short Time Fourier Transform is that size of the window function is unchangeable; the difficulty of Wigner-Ville Transform is that it has severe cross terms; the difficulty of Wavelet Transform is that it can only change the shape of its window. So this paper uses the method of Hilbert-Huang Transform, which is a powerful method for nonlinear and non-stationary time series analysis.
In the area of target classification, the Artificial Neural Network classifier is the most widely used. Different characteristics of signal or different classifiers with different structures will give diffentent classification results, which we may use to improve the performance of the system. The weighted vote method of multi-classifier outputs is an effective multi-classifier fusion algorithm, which has simple terms and is suit to deal with actual problems, so we adopt this method in this paper.
In this paper, we first review the methods for processing non-stationary data which have been used widely and point out their limitations, and then introduce Hilbert-Huang Transform method. Through the study on the structure characteristics of the signal from airdrop target water-entry, we do the research and extract the characteristics from three parts that are pulse signal, "quiet" interval, and fluctuant signal. We have studyed some artificial neural netwoks and used them respectively to classify the targets, and the results have shown that they have the complementarity. In the last of the paper, we have studyed the multi-calssifier fusion algorithm, and have used it to classify the targets, then we get a good result, which proves that the fusion method can be used in classification problems.
在目标分类领域中,人工神经网络分类器应用最为广泛。利用信号的不同特征或不同结构的人工神经网络分类器可以得到不同的分类结果,往往可以利用这些结果之间的互补性来改善系统的分类效果。多分类器输出向量加权投票算法是一种有效的多分类器融合算法,它具有算法简单、便于处理实际问题的优点。因此,本文采用此方法进行多分类器融合。
本文首先回顾了几种处理非平稳数据的常见方法,并分析了它们的优缺点,之后介绍了希尔伯特-黄变换方法。通过对空投目标入水声信号构成特点的研究,从击水声脉冲信号、“寂静”区间和气泡脉动信号三方面对空投目标入水声信号进行了理论分析和实测数据的特征提取研究。给出了常见人工神经网络的结构特点以及基本原理,并应用单个人工神经网络分类器对空投目标进行分类,分类结果表明不同分类器之间存在互补性。在本文的最后论述了多分类器融合算法,并给出基于多分类器融合算法的目标分类结果,证明了多分类器融合算法在解决目标分类问题时具有优越性。
Feature extraction of underwater transient signal is the key of the airdrop targets classification. Historically, there are many methods of signal analysis which are applied to extract signal characteristics, such as Fourier Transform, Short Time Fourier Transform, Wigner-Ville Transform, Wavelet Transform, and so on. But each of them has difficulties when they are used in the extraction of characteristics of underwater targets. The difficulty of Fourier Transform is that the signal must be linear, strictly periodic and stationary; the difficulty of Short Time Fourier Transform is that size of the window function is unchangeable; the difficulty of Wigner-Ville Transform is that it has severe cross terms; the difficulty of Wavelet Transform is that it can only change the shape of its window. So this paper uses the method of Hilbert-Huang Transform, which is a powerful method for nonlinear and non-stationary time series analysis.
In the area of target classification, the Artificial Neural Network classifier is the most widely used. Different characteristics of signal or different classifiers with different structures will give diffentent classification results, which we may use to improve the performance of the system. The weighted vote method of multi-classifier outputs is an effective multi-classifier fusion algorithm, which has simple terms and is suit to deal with actual problems, so we adopt this method in this paper.
In this paper, we first review the methods for processing non-stationary data which have been used widely and point out their limitations, and then introduce Hilbert-Huang Transform method. Through the study on the structure characteristics of the signal from airdrop target water-entry, we do the research and extract the characteristics from three parts that are pulse signal, "quiet" interval, and fluctuant signal. We have studyed some artificial neural netwoks and used them respectively to classify the targets, and the results have shown that they have the complementarity. In the last of the paper, we have studyed the multi-calssifier fusion algorithm, and have used it to classify the targets, then we get a good result, which proves that the fusion method can be used in classification problems.
引文
[1]胡广书.现代信号处理教程.第1版.清华大学出版社,2004:52-53页
[2]皇甫堪,陈建文,楼生强.现代数字信号处理.第1版.电子工业出版社,2004:392-395页
[3]章新华.水下目标自动识别的核心技术.兵工学报.1998,19(3):275-277页
[4]马艳,李志舜.基于连续小波变换的水下目标特征提取与分类.系统工程与电子技术.2003,25(3):375-378页
[5]高源.水雷目标识别中的数据融合技术.哈尔滨工程大学硕士论文.2002:20-21页
[6]谭善文.多分辨希尔伯特-黄(Hirbert_Huang)变换方法的研究.重庆大学博士论文.2001:25-26页
[7]张曼,程家兴,樊甫华,刘慧婷.利用Hilbert变换提取信号瞬时特征参数的问题研究.电视技术.2003,(4):44-48页
[8]高云超.希尔伯特.黄变换方法的仿真研究.哈尔滨工程大学硕士论文.2005:8-9页
[9]Norden E.Huang,Zheng Shen,Steven R.Long et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear non-stationary time series analysis.Proc.R.Soc,1998..A454:903-995P
[10]熊学军,郭炳火,胡筱敏,刘建军.EMD方法和Hilbert谱的应用与探讨.黄渤海海洋.2002,20(2):12-21页
[11]李迎,王帮峰,孙亚飞,施益峰.经验模式分解与Hilbert谱的分析与应用.传感器技术,2005,24(5):71-76页
[12]P.J.Oonincx,J.-P.Hermand.EMPIRICAL MODE DECOMPOSITION OF OCEAN ACOUSTIC DATA WITH CONSTRAINT ON THE FREQUENCY RANGE.Proceedings of the Seventh European Conference on Underwater Acoustics.Delft,The Netherlands.2004:1-6P
[13]钟佑明,金涛,秦树人.希尔伯特-黄变换中的一种新包络线算法.数 据采集与处理.2005,20(1):13-17页
[14]杨世锡,胡劲松,吴昭同等.基于高次样条插值的经验模态分解方法研究.浙江大学学报(工学版).2004,38(3):267-270页
[15]黄大吉,赵进平,苏纪兰.希尔伯特.黄变换的端点延拓.海洋学报.2003,25(1):1-11页
[16]邓拥军,王伟,钱成春,王忠,戴德君.EMD方法及Hilbert变换中边界问题的处理.科学通报.2001,46(3):257-263页
[17]张郁山,梁建文,胡聿贤.应用自回归模型处理EMD方法中的边界问题.自然科学进展.2003,13(10):1054-1059页
[18]龙思胜,张铁宝,龙峰.希尔伯特-黄变换中拟合过冲和端点飞翼的原因及解决办法.地震学报.2005,27(5):561-568页
[19]鲍筱玲,汤渭霖.击水声特性研究.哈尔滨工程学院学报.1981,(1):27-39页
[20]曲宝纯.入水研究-入水现象.船舶科学技术.1984,(2):6-13页
[21]路龙龙.空投鱼雷入水技术研究.西北工业大学硕士论文.2006:22-24页
[22]Stefannia Colonnese,Gaetano Sxarano.Transient signal detection using higher order moments.IEEE Transaction on Signal Processing.1999,47(2):515-520P
[23]Claire Chichereau Jean Louis.Short delay detection of a transient in additive Gaussian noise via higher order statistic test.IEEE Digital Signal Workshop Proceedings,New York,NY,USA,1996:299-302P
[24]Wang Zhen,Willett P.A performance study of some transient detectors.IEEE Transaction on Signal Processing.2000,48(9):2682-2685P
[25]Kirsteins I P,Mehta S K.Power-law processors for detection unknown signals in colored noise.IEEE Proc.ICASSP.1997,1:483-486P
[26]WANG Zhen,Willett P.All-purpose and plug-in power-law detectors for transient signals.IEEE Transaction on Signal Processing.2001,49(11):2454-2466P
[27].王燕麟.基于匹配滤波器的LFM信号高阶谱检测.声学与电子工程.1997,(4):1-8页
[28]吴国清,陈永强等.水声瞬态信号短时谱形态及谱相关法检测.声学学报.2000,26(6):510-515页
[29]孙鹏,刘平香等.瞬态信号检测方法的研究.舰船科学技术.2005,27(3):72-76页
[30]M.Strasberg.Gas Bubbles as Soruces of Sound in Liquids.JASA.1956,28(1):20-26P
[31]刘伯胜,雷家煜.水声学原理.第1版.哈尔滨工程大学出版社,1993:6-7页
[32]谭善文,秦树人,汤宝平.Hilbert-Huang变换的滤波特性及其应用.重庆大学学报.2004,27(2):9-12页
[33]胡桥,郝保安,吕林夏等.经验模式能量熵在水声目标检测中的应用.水下目标分类识别技术.2007,26(5):181-183页
[34]杨建刚.人工神经网络实用教程.第1版.浙江大学出版社,2001:12-13页
[35]楼顺天,施阳.基于MATLAB的系统分析与设计.神经网络.第1版.1998:6-9页
[36]CHEN S,COWAN CFN.GRANT P M.Orthogonal least squares learning algorithm for radial basis function networks.IEEE TransNeural Networks.1991(2):302-309P
[37]ROBERT J,SCHILLINGJAMES,CARROLL J.Approximation of nonlinear systems with radial basis function neural networks.IEEE Transactions on Neural Networks.2001,12(1):21-28P
[38]黄涛,夏佩伦.潜艇多信息源目标融合识别模型.兵工自动化.2004,23(3):3-5页
[39]Suen C Y,Nadal C,Mai T A etal.Recognition of totally unconstrained handwriting numerals based on the concept of multiple experts.Frontiers in Handwrinting Recognition,Montreal,Canada:International Workshop on Frontiers in Handwrinting Recognition,1990:131-143P
[40]景志宏,赵谊虹,程国华等.基于多神经网络分类器的目标识别仿真实验研究.系统仿真学报.2003,15(3):441-443页
[41]张亚军.基于神经网络数据融合的水下目标检测识别研究.西北工业大学硕士论文.2006:54-55页
[2]皇甫堪,陈建文,楼生强.现代数字信号处理.第1版.电子工业出版社,2004:392-395页
[3]章新华.水下目标自动识别的核心技术.兵工学报.1998,19(3):275-277页
[4]马艳,李志舜.基于连续小波变换的水下目标特征提取与分类.系统工程与电子技术.2003,25(3):375-378页
[5]高源.水雷目标识别中的数据融合技术.哈尔滨工程大学硕士论文.2002:20-21页
[6]谭善文.多分辨希尔伯特-黄(Hirbert_Huang)变换方法的研究.重庆大学博士论文.2001:25-26页
[7]张曼,程家兴,樊甫华,刘慧婷.利用Hilbert变换提取信号瞬时特征参数的问题研究.电视技术.2003,(4):44-48页
[8]高云超.希尔伯特.黄变换方法的仿真研究.哈尔滨工程大学硕士论文.2005:8-9页
[9]Norden E.Huang,Zheng Shen,Steven R.Long et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear non-stationary time series analysis.Proc.R.Soc,1998..A454:903-995P
[10]熊学军,郭炳火,胡筱敏,刘建军.EMD方法和Hilbert谱的应用与探讨.黄渤海海洋.2002,20(2):12-21页
[11]李迎,王帮峰,孙亚飞,施益峰.经验模式分解与Hilbert谱的分析与应用.传感器技术,2005,24(5):71-76页
[12]P.J.Oonincx,J.-P.Hermand.EMPIRICAL MODE DECOMPOSITION OF OCEAN ACOUSTIC DATA WITH CONSTRAINT ON THE FREQUENCY RANGE.Proceedings of the Seventh European Conference on Underwater Acoustics.Delft,The Netherlands.2004:1-6P
[13]钟佑明,金涛,秦树人.希尔伯特-黄变换中的一种新包络线算法.数 据采集与处理.2005,20(1):13-17页
[14]杨世锡,胡劲松,吴昭同等.基于高次样条插值的经验模态分解方法研究.浙江大学学报(工学版).2004,38(3):267-270页
[15]黄大吉,赵进平,苏纪兰.希尔伯特.黄变换的端点延拓.海洋学报.2003,25(1):1-11页
[16]邓拥军,王伟,钱成春,王忠,戴德君.EMD方法及Hilbert变换中边界问题的处理.科学通报.2001,46(3):257-263页
[17]张郁山,梁建文,胡聿贤.应用自回归模型处理EMD方法中的边界问题.自然科学进展.2003,13(10):1054-1059页
[18]龙思胜,张铁宝,龙峰.希尔伯特-黄变换中拟合过冲和端点飞翼的原因及解决办法.地震学报.2005,27(5):561-568页
[19]鲍筱玲,汤渭霖.击水声特性研究.哈尔滨工程学院学报.1981,(1):27-39页
[20]曲宝纯.入水研究-入水现象.船舶科学技术.1984,(2):6-13页
[21]路龙龙.空投鱼雷入水技术研究.西北工业大学硕士论文.2006:22-24页
[22]Stefannia Colonnese,Gaetano Sxarano.Transient signal detection using higher order moments.IEEE Transaction on Signal Processing.1999,47(2):515-520P
[23]Claire Chichereau Jean Louis.Short delay detection of a transient in additive Gaussian noise via higher order statistic test.IEEE Digital Signal Workshop Proceedings,New York,NY,USA,1996:299-302P
[24]Wang Zhen,Willett P.A performance study of some transient detectors.IEEE Transaction on Signal Processing.2000,48(9):2682-2685P
[25]Kirsteins I P,Mehta S K.Power-law processors for detection unknown signals in colored noise.IEEE Proc.ICASSP.1997,1:483-486P
[26]WANG Zhen,Willett P.All-purpose and plug-in power-law detectors for transient signals.IEEE Transaction on Signal Processing.2001,49(11):2454-2466P
[27].王燕麟.基于匹配滤波器的LFM信号高阶谱检测.声学与电子工程.1997,(4):1-8页
[28]吴国清,陈永强等.水声瞬态信号短时谱形态及谱相关法检测.声学学报.2000,26(6):510-515页
[29]孙鹏,刘平香等.瞬态信号检测方法的研究.舰船科学技术.2005,27(3):72-76页
[30]M.Strasberg.Gas Bubbles as Soruces of Sound in Liquids.JASA.1956,28(1):20-26P
[31]刘伯胜,雷家煜.水声学原理.第1版.哈尔滨工程大学出版社,1993:6-7页
[32]谭善文,秦树人,汤宝平.Hilbert-Huang变换的滤波特性及其应用.重庆大学学报.2004,27(2):9-12页
[33]胡桥,郝保安,吕林夏等.经验模式能量熵在水声目标检测中的应用.水下目标分类识别技术.2007,26(5):181-183页
[34]杨建刚.人工神经网络实用教程.第1版.浙江大学出版社,2001:12-13页
[35]楼顺天,施阳.基于MATLAB的系统分析与设计.神经网络.第1版.1998:6-9页
[36]CHEN S,COWAN CFN.GRANT P M.Orthogonal least squares learning algorithm for radial basis function networks.IEEE TransNeural Networks.1991(2):302-309P
[37]ROBERT J,SCHILLINGJAMES,CARROLL J.Approximation of nonlinear systems with radial basis function neural networks.IEEE Transactions on Neural Networks.2001,12(1):21-28P
[38]黄涛,夏佩伦.潜艇多信息源目标融合识别模型.兵工自动化.2004,23(3):3-5页
[39]Suen C Y,Nadal C,Mai T A etal.Recognition of totally unconstrained handwriting numerals based on the concept of multiple experts.Frontiers in Handwrinting Recognition,Montreal,Canada:International Workshop on Frontiers in Handwrinting Recognition,1990:131-143P
[40]景志宏,赵谊虹,程国华等.基于多神经网络分类器的目标识别仿真实验研究.系统仿真学报.2003,15(3):441-443页
[41]张亚军.基于神经网络数据融合的水下目标检测识别研究.西北工业大学硕士论文.2006:54-55页