雷达信号脉内调制方式识别与特征分析
|
摘要
随着新体制雷达的迅猛发展,空间电磁环境的日益复杂,雷达信号脉内调制方式识别和特征分析变得愈加困难。本文首先以小波包变换和瞬时频率分析为基础,研究了两种雷达信号调制方式识别的算法,然后从雷达信号调制参数和雷达信号包络特征两个方面,研究了雷达信号脉内特征分析的相关算法。主要工作如下:
1.研究了基于小波包的雷达信号脉内调制方式识别算法。首先深入研究小波包变换的基本理论,利用最优小波基技术,提取了含有信息量大的小波包重构系数。针对小波包重构系数维数过高和如何构造识别特征的问题,给出了两种特征构造的方法:第一、计算小波包重构系数的能量,以能量的统计特征为识别特征;第二、对小波包重构系数进行奇异值分解,并取前6个大的奇异值做为识别特征。详细介绍了BP网络模型的参数设置问题,并建立了相应的BP分类器。最后通过实验分析了基于两种构造特征的识别算法的性能。
2.研究了基于瞬时频率的雷达信号调制方式识别算法。针对在低信噪比条件下瞬时频率统计特征稳定性差而导致识别率低的问题,给出了一种基于瞬时频率图的雷达信号调制方式识别的方法。在深入分析雷达信号瞬时频率特性的基础上,给出了一种将瞬时频率转化为二值图像的方法,提取了瞬时频率二值图像的矩特征,并进行了归一化处理,消除了采样频率和信号长度对特征的影响。建立了两种级联分类器并设计了相应的BP神经网络分类器,最终通过仿真实验分析了该方法的性能。
3.研究了线性调频信号参数的贝叶斯估计方法和相关接收提取相位编码子码宽度的方法。推导了一种线性调频信号参数的贝叶斯估计模型,并将其推广到正弦信号载频的贝叶斯估计中。通过MCMC方法对贝叶斯估计进行优化计算,解决了贝叶斯估计方法计算量大的问题;研究了一种结合随机游走抽样和独立马尔科夫抽样的混合抽样方法,提高了MCMC算法的收敛速度;利用相关接收法提取子码宽度,采用平滑,子码宽度微调和实虚部结果相平均等方法,提高了子码宽度的估计精度。
4.研究了雷达信号包络特征分析的相关算法。深入研究了希尔伯特变换提取包络的算法,针对其抗噪性能差,提取的包络误差大等缺点,研究了复小波变换提取雷达信号的包络算法。对提取的包络进行平滑处理,减小噪声对包络的影响。提取了复小波变换后几层的包络特征,并使用PCA技术对包络特征进行了优化处理,减小了包络特征的维数并增加了包络特征的稳定性。
Radar signal intra-pulse modulation recognition and analysis have encountered challenges, as new system radars have rapidly developed and electromagnetism environment in space was more and more complex. First, two algorithms of modulation recognition is researched based on wavelet packet and BP neural network. Then, intra-pulse character of radar signal is researched from radar signal modulation parameter extraction and radar signal inter-pulse analysis. My main work is summarized as follow:
1. Radar signal intra-pulse modulation recognition algorthms based on wavelet packet is studied. Wavelet packet translation is studied deeply, using optimization algorithm of wavelet packet decomposed, reconstruct coefficients which contained large information was extracted. To solve the mesh of the wavelet packet reconstruct coefficients too high and how to construct the recognition feature, two method are proposed: First one, the energy of wavelet packet reconstruct coefficient is calculated, statistical features of energy are recognition feature; Another one, feature selecting based on SVD from Reconstruct coefficients of wavelet packet is proposed, six big singular values was recognition feature. How to setup parameter of BP model is introduced, and establish the BP classifier. The performance of two kinds of features is analyzed by simulating.
2. Radar signal intra-pulse modulation recognition algorthms based on instantaneous frequency is studied. To solve the lower recognition rate issue, a radar signal intra-pulse modulation recognition algorthms based on instantaneous frequency image is proposed. The instantaneous frequency was studied deeply, a method of transforming the instantaneous frequency into binary image is proposed. The moment feature of the binary image is extracted, and normalizing the moment feature eliminates the affect of sampling frequency and the length of signal. Two classifiers of and a BP classifier is established, finally the performance of this method is analyzed by simulating.
3. Bayesian parameter estimation algorithm of LFM signal and code width of PSK signal based on Correlation receive algorithm is studied. The model of Bayesian estimation is studied for estimating the modulation parameter of LFM signal and the frequency of sine signal. The MCMC algorithm is introduced to calculate the Bayesian estimation for eliminating the computation burden. To improve the convergence speed of MCMC, a hybrid sampling between Rand Walks and Independent Markov Chain is proposed. Correlation receive algorithm is studied for estimating the width of code. The precision of code width is improved by smoothing, fine-tuning the width of code and averaging the real part’result and imaginary part’result.
4. The algorithm of envelop feature analysis of radar signal is studied. The performance of Hilbert transform and complex wavelet transform for extracting envelop is studied deeply. In order to eliminating the affect of noise, complex wavelet transform is used to extract envelop. To eliminate the affect of noisy, envelop is smoothed. Finally the envelop feature is extracted from the last layers of complex wavelet transform, and the PCA method is used to eliminate the redundancy of the envelop feature and enhance the stability of the envelop feature.
1.研究了基于小波包的雷达信号脉内调制方式识别算法。首先深入研究小波包变换的基本理论,利用最优小波基技术,提取了含有信息量大的小波包重构系数。针对小波包重构系数维数过高和如何构造识别特征的问题,给出了两种特征构造的方法:第一、计算小波包重构系数的能量,以能量的统计特征为识别特征;第二、对小波包重构系数进行奇异值分解,并取前6个大的奇异值做为识别特征。详细介绍了BP网络模型的参数设置问题,并建立了相应的BP分类器。最后通过实验分析了基于两种构造特征的识别算法的性能。
2.研究了基于瞬时频率的雷达信号调制方式识别算法。针对在低信噪比条件下瞬时频率统计特征稳定性差而导致识别率低的问题,给出了一种基于瞬时频率图的雷达信号调制方式识别的方法。在深入分析雷达信号瞬时频率特性的基础上,给出了一种将瞬时频率转化为二值图像的方法,提取了瞬时频率二值图像的矩特征,并进行了归一化处理,消除了采样频率和信号长度对特征的影响。建立了两种级联分类器并设计了相应的BP神经网络分类器,最终通过仿真实验分析了该方法的性能。
3.研究了线性调频信号参数的贝叶斯估计方法和相关接收提取相位编码子码宽度的方法。推导了一种线性调频信号参数的贝叶斯估计模型,并将其推广到正弦信号载频的贝叶斯估计中。通过MCMC方法对贝叶斯估计进行优化计算,解决了贝叶斯估计方法计算量大的问题;研究了一种结合随机游走抽样和独立马尔科夫抽样的混合抽样方法,提高了MCMC算法的收敛速度;利用相关接收法提取子码宽度,采用平滑,子码宽度微调和实虚部结果相平均等方法,提高了子码宽度的估计精度。
4.研究了雷达信号包络特征分析的相关算法。深入研究了希尔伯特变换提取包络的算法,针对其抗噪性能差,提取的包络误差大等缺点,研究了复小波变换提取雷达信号的包络算法。对提取的包络进行平滑处理,减小噪声对包络的影响。提取了复小波变换后几层的包络特征,并使用PCA技术对包络特征进行了优化处理,减小了包络特征的维数并增加了包络特征的稳定性。
Radar signal intra-pulse modulation recognition and analysis have encountered challenges, as new system radars have rapidly developed and electromagnetism environment in space was more and more complex. First, two algorithms of modulation recognition is researched based on wavelet packet and BP neural network. Then, intra-pulse character of radar signal is researched from radar signal modulation parameter extraction and radar signal inter-pulse analysis. My main work is summarized as follow:
1. Radar signal intra-pulse modulation recognition algorthms based on wavelet packet is studied. Wavelet packet translation is studied deeply, using optimization algorithm of wavelet packet decomposed, reconstruct coefficients which contained large information was extracted. To solve the mesh of the wavelet packet reconstruct coefficients too high and how to construct the recognition feature, two method are proposed: First one, the energy of wavelet packet reconstruct coefficient is calculated, statistical features of energy are recognition feature; Another one, feature selecting based on SVD from Reconstruct coefficients of wavelet packet is proposed, six big singular values was recognition feature. How to setup parameter of BP model is introduced, and establish the BP classifier. The performance of two kinds of features is analyzed by simulating.
2. Radar signal intra-pulse modulation recognition algorthms based on instantaneous frequency is studied. To solve the lower recognition rate issue, a radar signal intra-pulse modulation recognition algorthms based on instantaneous frequency image is proposed. The instantaneous frequency was studied deeply, a method of transforming the instantaneous frequency into binary image is proposed. The moment feature of the binary image is extracted, and normalizing the moment feature eliminates the affect of sampling frequency and the length of signal. Two classifiers of and a BP classifier is established, finally the performance of this method is analyzed by simulating.
3. Bayesian parameter estimation algorithm of LFM signal and code width of PSK signal based on Correlation receive algorithm is studied. The model of Bayesian estimation is studied for estimating the modulation parameter of LFM signal and the frequency of sine signal. The MCMC algorithm is introduced to calculate the Bayesian estimation for eliminating the computation burden. To improve the convergence speed of MCMC, a hybrid sampling between Rand Walks and Independent Markov Chain is proposed. Correlation receive algorithm is studied for estimating the width of code. The precision of code width is improved by smoothing, fine-tuning the width of code and averaging the real part’result and imaginary part’result.
4. The algorithm of envelop feature analysis of radar signal is studied. The performance of Hilbert transform and complex wavelet transform for extracting envelop is studied deeply. In order to eliminating the affect of noise, complex wavelet transform is used to extract envelop. To eliminate the affect of noisy, envelop is smoothed. Finally the envelop feature is extracted from the last layers of complex wavelet transform, and the PCA method is used to eliminate the redundancy of the envelop feature and enhance the stability of the envelop feature.
引文
[1]熊群力,陈润生,杨小牛等.综合电子战[M].北京:国防工业出版社,2008.(25-28).
[2] Liu W K, Zhu D J, Zhang C H. The extraction of modulationcharacteristics of radar signal using wavelet transform[C]. Proceedings of the 4th International Conference on Signal Processing, 1998, 288-291.
[3] P. Prakasam and M.Madheswaran, Digital Modulation Identification Model Using Wavelet Transform and Statistical Parameters[J], Journal of Computer Systems, Networks and Communications, 2008, 1-8.
[4]田野,雷英杰,李军.基于小波变换的雷达信号特征提取方法[J].空军工程大学学报,2006,7(5):22-24.
[5]柳征,姜文利,周一宇.基于小波包变换的辐射源信号识别[J].信号处理,2005,21(5):460-464.
[6] Milne PR,Pace PE.Wigner distribution and analysis of FMCW and P-4 polyphase LPI waveforms[J]. 2002 IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP’02), Orlando, EL, 2002, 39(4):44-47.
[7] Ananthram Swami, Brian M.Sadler, Hierarchical Digital Modulation Classification Using Cumulants[J], IEEE TRANSACTIONS ON COMMUNICATIONS, VOL.48, NO.3, MARCH 2000 .
[8] Wei Dai, Wouzheng Wang,Jing Wang, Joint Power Estimation and Modulation Classification Using Second-and Higher Statistics [C], Wireless Communications and Networking Conference, 2002, 155-158.
[9] Ataollah Abrahamzadeh, Seyed Alireza Seyedin and Mehdi Dehghan. Digital Signal Type Identification Using an Efficient Identifier[C], EURASIP Journal on Advances in Signal Processing, EURASIP Journal on Advances in Signal Processing, 2007, 1-9
[10] JPY Lee. A multi-channel digital receiver: intra-pulse analysis and direction finding[J]. Processdings of IEEE Pacific Rim Conference on Communications, Computers and Signal Processing. 1999,589-592.
[11]蔡权伟.多分量信号的信号分量分离技术研究[D].成都:电子科技大学,2006.
[12]冀贞海,朱伟强,赵力.基于时频分布图像和主分量分析的脉内调制识别算法研究[J].电路与系统学报,2009,2(14):22-26.
[13]张葛祥,胡来招,金炜东.基于熵特征的雷达辐射源信号识别[J].电波科学学报,2005, 20(4):440-445.
[14]褚鹏飞,唐斌,祝俊.基于特征参数的雷达信号识别算法及DSP实现[J].电讯技术,2009,2(49):5-8.
[15]毕大平,董灰,姜秋喜.基于瞬时频率的脉内调制识别技术[J] .电子对抗技术,2005,3(2):6-9.
[16]普运伟,金炜东,胡来招.基于瞬时频率二次特征提取的辐射源信号分类[J].西南交通大学学报,2007,34(3):373-379.
[17] Xia X G.Discrete Chip-Fourier transform and its application to chip rate estimation[J].IEEE Trans. Signal Processing, 2000, 48(11):3122-3133.
[18] G.F.Boudreax-Bartels, Time-Varying signal processing using the Wigner distribution, In Advances of Geophysical Data Processing, Vol 2, M.Simaan, Ed.Greenwich,1985.
[19] Barbarossa S. Analysis of multi comonent LFM signals by a combined Wigner-Hough transform[J]. IEEE Trans Signal Processing, 1995, 43(6):1511-1515.
[20] Wang M S, Chan A K, Chui C K. Linear Frequency modulated signal detection using Radon-Ambituty transform[J]. IEEE Trans on Signal Processing, 1998, 46(3):571-586.
[21]李文臣,王雪松.线性调频参数估计方法的数学统一[J].信号处理,2009, 42(8):1292-1297.
[22] T.Jabatzoglou, Fast maximum likelihood joint estimation of frequency and frequecy rate[J], IEEE Trans On AES, 1986, 22(6):708-715.
[23] Supratim Saha and Steven M.Kay, Maximum likehood parameter estimation of super imposed chirps using monto carlo[J]. IEEE Trans On Signal Processing, 2002, 5(2):224-230.
[24]钱城.基于DSP的雷达信号参数估计准实时处理方法研究[D].南京:南京航空航天大学,2008.
[25]刘鹭航,周云生.基于循环谱的相位编码信号调制参数估计[J].遥测遥感,2009,33(3):47-53.
[26]吴江标,万方,郁春来.基于小波变换法的相位编码信号脉内特征提取[J].航天电子对抗,2005,21(3):38-40.
[27]许丹.辐射源指纹机理及识别方法研究[D].长沙:国防科技大学,2008.
[28]任春晖.通信电台的个体特征分析[D] .成都:电子科技大学,2006.
[29] Serinken N O Ureten. Generalised dimension characterisation of radio transmitter turn-on transients[J]. Electonics Letters. Vol.36,pp:1064-1066. 2000.
[30]刘庆云,陆飞飞,朱伟强等.辐射源细微特征用于个体识别的可行性分析[J].航天电子对抗,2008, 24(2):40-42.
[31] Adam Kawalec, Robert Owczarek, Radar Emitter Recognition Using Intrapulse Data[J], Radar conference, 2004,11(2):435-438.
[32]张旻,王若冰,钟子发.通信电台个体识别中的载波稳定度特征提取技术研究[J].电子与信息学报,2008,30(10):45-51.
[33] Ming-Wei Liu and John F Doherty, Sepcific Emitter Identification using Nonlinear Device Estimation[J] . Microwave, radar and wireless communication,2008,1-5.
[34]张贤达.现代信号处理[M].北京,清华大学出版社,2002.(381-383).
[35]杨福生.小波变换的工程分析与应用[M] .北京,科学出版设,1999:(34-36).
[36]张国华,袁中凡,李彬彬.心音信号特征提取小波包算法研究[J].振动与冲击,2008,27(7):47-49.
[37]刘毅,张彩明.基于多尺度小波包分析的肺音特征提取与分类[J].计算机学报,2006,29(5):770-776.
[38]王大凯,彭进业.小波分析及其在信号处理中的应用[M].北京:电子工业出版社,2006.(167-169).
[39]张葛祥,荣海娜.基于小波包变换和特征选择的雷达辐射源信号识别[J].电路与系统学报,2006,11(6):45-49.
[40]王玲霞,袁佳,张效义.基于小波包变换的辐射源信号特征提取和识别[J] .通信技术,2009,3(42):215-217.
[41]袁曾任.人工神经元网络及其应用[M].北京,清华大学出版社,1999.(81-89).
[42]符东旭.跳汰机人工神经网络控制技术研究[D].太原,太原理工大学,2007.
[43] Nguyen D,B Window,Improving the learning speed of 2-layer Neural Networks by choosing Intial Values of Adaptive Weights[J], In Proceedings of the International Joint Conference on Neural Network, 1990, 21-26.
[44] Sergios theodoridis Konstantinos Koutroumbas,李晶皎(译). Pattern Recognition[M].北京:电子工业出版社,2004.(112-116).
[45]高隽.人工神经网络原理及仿真实例[M].北京:机械工业出版社,2007.(6-10).
[46] Boualem Boashash. Estimating and interpreting the intantaneous frequency of a signal-part 1: Fundamentals [J]. Processding of the IEEE. 1992, 80(4): 520-538
[47]陈平,李庆民,赵彤.瞬时频率估计算法研究进展综述[J] .电测与仪表,487(43):1-6.
[48] P Maragos,J F Kaiser. Energy separation in signal modulation to speech analysis[J]. IEEE Transactions on Signal Process, 1993, 41(10) :3024-3051.
[49]陈平.信号瞬时频率的估计方法及其应用[D].济南:山东大学,2007.
[50] Henri Maitre .现代数字图像处理[M] .北京:电子工业出版社,2006:(186-189).
[51]苏严华. Visual C++数字图像识别技术典型案例[M].北京:人民邮电出版社,2004,347~349.
[52]朱本仁.蒙特卡罗方法引论[M].济南:山东大学出版社,1987:(80-83).
[53]刘次华.随机过程[M].太原:华中科技大学出版设,2001.(42-49).
[54]金美娜.基于MCMC方法的宽带信号源数和DOA联合估计方法研究[D].郑州:信息工程学院,2008.
[55] Metropolis, Equation of State calculation by fast computing machines [J], Journal of Chemical Physics, 1953 , 21 (6):1087-1092.
[56] W.K. Hastings, Monte Carlo sampling methods using Markov chain and their application [J], Biometrika, Vol. 57, No. 1, pp. 97-109, 1970.
[57] Christophe Andrieu, Arnaud Doucet. Joint Bayesian Model Selection and Estimation of Noisy Sinusoids via Reversible Jump MCMC [J], IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 47, NO. 10, OCTOBER 1999.
[58]张贤达,保峥.非平稳信号分析与处理[M].北京:国防工业出版社,2001.(153-180).
[59]邹建彬.高分辨率频率测量方法研究[D].长沙:国防科技大学,2006.
[60]张英龙.基于DSP的雷达信号参数估计及其硬件实现[D].南京:南京航空航天大学,2007.
[61]刘崇春,裘正定.一种基于Morlet小波的振动信号包络提取方法[J].电路与系统学报,2000,5(1):65-71.
[62]刘爱霞,赵国庆.雷达信号包络特征检测与分析[J] .电子对抗技术,2002,17(3):12-16.
[63]张国柱.雷达辐射源识别技术研究[D].长沙:国防科技大学,2005.
[64]林彦,王秀坛,彭应宇.基于MCMC的线性调频信号最大似然参数估计[J].清华大学学报,2004,4(44):511-514.
[65]华旭东.线性调频信号参数快速估计方法研究[D].上海:上海交通大学,2007.
[2] Liu W K, Zhu D J, Zhang C H. The extraction of modulationcharacteristics of radar signal using wavelet transform[C]. Proceedings of the 4th International Conference on Signal Processing, 1998, 288-291.
[3] P. Prakasam and M.Madheswaran, Digital Modulation Identification Model Using Wavelet Transform and Statistical Parameters[J], Journal of Computer Systems, Networks and Communications, 2008, 1-8.
[4]田野,雷英杰,李军.基于小波变换的雷达信号特征提取方法[J].空军工程大学学报,2006,7(5):22-24.
[5]柳征,姜文利,周一宇.基于小波包变换的辐射源信号识别[J].信号处理,2005,21(5):460-464.
[6] Milne PR,Pace PE.Wigner distribution and analysis of FMCW and P-4 polyphase LPI waveforms[J]. 2002 IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP’02), Orlando, EL, 2002, 39(4):44-47.
[7] Ananthram Swami, Brian M.Sadler, Hierarchical Digital Modulation Classification Using Cumulants[J], IEEE TRANSACTIONS ON COMMUNICATIONS, VOL.48, NO.3, MARCH 2000 .
[8] Wei Dai, Wouzheng Wang,Jing Wang, Joint Power Estimation and Modulation Classification Using Second-and Higher Statistics [C], Wireless Communications and Networking Conference, 2002, 155-158.
[9] Ataollah Abrahamzadeh, Seyed Alireza Seyedin and Mehdi Dehghan. Digital Signal Type Identification Using an Efficient Identifier[C], EURASIP Journal on Advances in Signal Processing, EURASIP Journal on Advances in Signal Processing, 2007, 1-9
[10] JPY Lee. A multi-channel digital receiver: intra-pulse analysis and direction finding[J]. Processdings of IEEE Pacific Rim Conference on Communications, Computers and Signal Processing. 1999,589-592.
[11]蔡权伟.多分量信号的信号分量分离技术研究[D].成都:电子科技大学,2006.
[12]冀贞海,朱伟强,赵力.基于时频分布图像和主分量分析的脉内调制识别算法研究[J].电路与系统学报,2009,2(14):22-26.
[13]张葛祥,胡来招,金炜东.基于熵特征的雷达辐射源信号识别[J].电波科学学报,2005, 20(4):440-445.
[14]褚鹏飞,唐斌,祝俊.基于特征参数的雷达信号识别算法及DSP实现[J].电讯技术,2009,2(49):5-8.
[15]毕大平,董灰,姜秋喜.基于瞬时频率的脉内调制识别技术[J] .电子对抗技术,2005,3(2):6-9.
[16]普运伟,金炜东,胡来招.基于瞬时频率二次特征提取的辐射源信号分类[J].西南交通大学学报,2007,34(3):373-379.
[17] Xia X G.Discrete Chip-Fourier transform and its application to chip rate estimation[J].IEEE Trans. Signal Processing, 2000, 48(11):3122-3133.
[18] G.F.Boudreax-Bartels, Time-Varying signal processing using the Wigner distribution, In Advances of Geophysical Data Processing, Vol 2, M.Simaan, Ed.Greenwich,1985.
[19] Barbarossa S. Analysis of multi comonent LFM signals by a combined Wigner-Hough transform[J]. IEEE Trans Signal Processing, 1995, 43(6):1511-1515.
[20] Wang M S, Chan A K, Chui C K. Linear Frequency modulated signal detection using Radon-Ambituty transform[J]. IEEE Trans on Signal Processing, 1998, 46(3):571-586.
[21]李文臣,王雪松.线性调频参数估计方法的数学统一[J].信号处理,2009, 42(8):1292-1297.
[22] T.Jabatzoglou, Fast maximum likelihood joint estimation of frequency and frequecy rate[J], IEEE Trans On AES, 1986, 22(6):708-715.
[23] Supratim Saha and Steven M.Kay, Maximum likehood parameter estimation of super imposed chirps using monto carlo[J]. IEEE Trans On Signal Processing, 2002, 5(2):224-230.
[24]钱城.基于DSP的雷达信号参数估计准实时处理方法研究[D].南京:南京航空航天大学,2008.
[25]刘鹭航,周云生.基于循环谱的相位编码信号调制参数估计[J].遥测遥感,2009,33(3):47-53.
[26]吴江标,万方,郁春来.基于小波变换法的相位编码信号脉内特征提取[J].航天电子对抗,2005,21(3):38-40.
[27]许丹.辐射源指纹机理及识别方法研究[D].长沙:国防科技大学,2008.
[28]任春晖.通信电台的个体特征分析[D] .成都:电子科技大学,2006.
[29] Serinken N O Ureten. Generalised dimension characterisation of radio transmitter turn-on transients[J]. Electonics Letters. Vol.36,pp:1064-1066. 2000.
[30]刘庆云,陆飞飞,朱伟强等.辐射源细微特征用于个体识别的可行性分析[J].航天电子对抗,2008, 24(2):40-42.
[31] Adam Kawalec, Robert Owczarek, Radar Emitter Recognition Using Intrapulse Data[J], Radar conference, 2004,11(2):435-438.
[32]张旻,王若冰,钟子发.通信电台个体识别中的载波稳定度特征提取技术研究[J].电子与信息学报,2008,30(10):45-51.
[33] Ming-Wei Liu and John F Doherty, Sepcific Emitter Identification using Nonlinear Device Estimation[J] . Microwave, radar and wireless communication,2008,1-5.
[34]张贤达.现代信号处理[M].北京,清华大学出版社,2002.(381-383).
[35]杨福生.小波变换的工程分析与应用[M] .北京,科学出版设,1999:(34-36).
[36]张国华,袁中凡,李彬彬.心音信号特征提取小波包算法研究[J].振动与冲击,2008,27(7):47-49.
[37]刘毅,张彩明.基于多尺度小波包分析的肺音特征提取与分类[J].计算机学报,2006,29(5):770-776.
[38]王大凯,彭进业.小波分析及其在信号处理中的应用[M].北京:电子工业出版社,2006.(167-169).
[39]张葛祥,荣海娜.基于小波包变换和特征选择的雷达辐射源信号识别[J].电路与系统学报,2006,11(6):45-49.
[40]王玲霞,袁佳,张效义.基于小波包变换的辐射源信号特征提取和识别[J] .通信技术,2009,3(42):215-217.
[41]袁曾任.人工神经元网络及其应用[M].北京,清华大学出版社,1999.(81-89).
[42]符东旭.跳汰机人工神经网络控制技术研究[D].太原,太原理工大学,2007.
[43] Nguyen D,B Window,Improving the learning speed of 2-layer Neural Networks by choosing Intial Values of Adaptive Weights[J], In Proceedings of the International Joint Conference on Neural Network, 1990, 21-26.
[44] Sergios theodoridis Konstantinos Koutroumbas,李晶皎(译). Pattern Recognition[M].北京:电子工业出版社,2004.(112-116).
[45]高隽.人工神经网络原理及仿真实例[M].北京:机械工业出版社,2007.(6-10).
[46] Boualem Boashash. Estimating and interpreting the intantaneous frequency of a signal-part 1: Fundamentals [J]. Processding of the IEEE. 1992, 80(4): 520-538
[47]陈平,李庆民,赵彤.瞬时频率估计算法研究进展综述[J] .电测与仪表,487(43):1-6.
[48] P Maragos,J F Kaiser. Energy separation in signal modulation to speech analysis[J]. IEEE Transactions on Signal Process, 1993, 41(10) :3024-3051.
[49]陈平.信号瞬时频率的估计方法及其应用[D].济南:山东大学,2007.
[50] Henri Maitre .现代数字图像处理[M] .北京:电子工业出版社,2006:(186-189).
[51]苏严华. Visual C++数字图像识别技术典型案例[M].北京:人民邮电出版社,2004,347~349.
[52]朱本仁.蒙特卡罗方法引论[M].济南:山东大学出版社,1987:(80-83).
[53]刘次华.随机过程[M].太原:华中科技大学出版设,2001.(42-49).
[54]金美娜.基于MCMC方法的宽带信号源数和DOA联合估计方法研究[D].郑州:信息工程学院,2008.
[55] Metropolis, Equation of State calculation by fast computing machines [J], Journal of Chemical Physics, 1953 , 21 (6):1087-1092.
[56] W.K. Hastings, Monte Carlo sampling methods using Markov chain and their application [J], Biometrika, Vol. 57, No. 1, pp. 97-109, 1970.
[57] Christophe Andrieu, Arnaud Doucet. Joint Bayesian Model Selection and Estimation of Noisy Sinusoids via Reversible Jump MCMC [J], IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 47, NO. 10, OCTOBER 1999.
[58]张贤达,保峥.非平稳信号分析与处理[M].北京:国防工业出版社,2001.(153-180).
[59]邹建彬.高分辨率频率测量方法研究[D].长沙:国防科技大学,2006.
[60]张英龙.基于DSP的雷达信号参数估计及其硬件实现[D].南京:南京航空航天大学,2007.
[61]刘崇春,裘正定.一种基于Morlet小波的振动信号包络提取方法[J].电路与系统学报,2000,5(1):65-71.
[62]刘爱霞,赵国庆.雷达信号包络特征检测与分析[J] .电子对抗技术,2002,17(3):12-16.
[63]张国柱.雷达辐射源识别技术研究[D].长沙:国防科技大学,2005.
[64]林彦,王秀坛,彭应宇.基于MCMC的线性调频信号最大似然参数估计[J].清华大学学报,2004,4(44):511-514.
[65]华旭东.线性调频信号参数快速估计方法研究[D].上海:上海交通大学,2007.