短波语音通话下的飞机类型识别研究
|
摘要
短波语音通话下的飞机驾驶舱背景声进行飞机类型的识别，是非合作目标识别领域的一个新课题。现有的飞机识别技术，主要基于雷达和图像进行飞机识别，而两者的不足之处在于作用距离近。但目前短波语音通话下的飞机识别主要途径是由资深专业人员进行倾听识别，关于这方面的研究，国内外缺乏公开的文献报道。本文结合国家自然科学基金“基于短波语音通话的飞机类型识别研究”，以飞机的短波语音通话为目标信号，进行了飞机类型辨识的相关系列研究工作。
首先，为区分目标信号的来源是地面目标还是飞机目标，本文利用Welch算法，对地面信号/飞机信号进行了目标信号的辨识，结果表明地面目标与飞机目标存在着很大的差异性，实现了目标信号辨识。对飞机驾驶舱背景声进行特性分析，是飞机类型识别工作中的首要任务。为分析飞机驾驶舱背景声信号相关物理特性，提出集合均值和集合方差的分析方法，应用在目标信号的频域以及倒谱上，对飞机驾驶舱背景声信号进行相关物理特性分析，结果表明不同飞机类型的驾驶舱背景声的谱峰特性差异明显，为语音抑制、特征提取以及类型识别做了积极的铺垫。
其次，在原始信号中包含有大段清晰的飞行员话音，此处语音相对于飞机背景声成为了强干扰噪声，因此需要去除。利用了基于谱熵的语音端点检测，对所检测到的原始信号流中的语音段采用两种方式对目标信号进行语音抑制。利用经验模态分解与小波分析（EMD_WT）进行语音抑制，但在其语音抑制的过程中引入了大量的过程噪声，为后续的相关信号处理带来了很大的隐患。于是提出利用全局经验模态分解与小波包分解结合（EEMD_WP）实现语音抑制。通过时域与频域的对比，EEMD_WP较EMD_WT方式取得了较好的性能，对语音进行了有效的抑制，增强了背景声。
第三，特征提取是飞机类型识别工作中的一个重要内容。从听觉感知的角度对飞机驾驶舱背景声进行了特征提取的研究。依据MFCC提取流程，提出用基于听觉感知的Bark小波包提取特征，结合所分析的目标信号的物理特性，利用对角切片的方式选取小波包结点，得到了体现目标信号物理特性的特征，并结合具有时变特性的MFCC、PLP的Delta特征，为减少特征参数的高阶冗余，经过PCA降维和F比评价得到简洁的多角度的混合特征参数，实验表明，它比单种特征有更好的分类性能，实现对目标信号进行建模的目的，为飞机类型识别作必要准备。
最后，针对飞机的高斯混合模型(GMM)参数冗余的问题，利用基于贝叶斯阴阳(BYY)和谐学习理论的普通梯度算法进行模型选择，实现模型参数估计及阶数的自动选择。而普通梯度学习算法往往会陷入和谐函数的局部极值而导致不理想的结果，提出基于黎曼流形的自然梯度的贝叶斯阴阳和谐学习的模型选择算法。结果表明，基于自然梯度的模型选择算法不仅能够实现自动模型选择，对模型复杂度进行了不同程度的有效压缩，并且自然梯度学习的收敛性能优于普通梯度学习算法，使高斯混合模型对飞机目标信号的分布空间的拟合程度更进一步，提高了极少信息量下的识别性能。
Aircraft type recognition based on shortwave speech communication, is a new topic innon-cooperation field, which is different from radar and image means. The former has moredetection operating range. However, the way to identify aircraft type mainly is by experiencedprofessionals and no open relevant literatures are reported about it. Combined with theNational Natural Science Foundation ‘Research on aircraft type recognition based onshortwave speech communication’, based on the research of the target signal, the relatedresearch work on aircraft type recognition is conducted.
Firstly, to identify the target signal, Welch algorithm is used to analyze the backgroundsignal/aircraft signal in this paper. The result shows that there exists a lot of differencebetween the ground signal and aircraft signal, which realizes the location of target signal inthe aircraft type recognition. The physical characteristics of aircraft cockpit background soundis the primary task of studied based on shortwave speech communication. Ensemble mean andensemble variance are defined and this paper applys them to frequency domain and cepstrum,which analyzes physical characteristics of background sound. The result reveals that differenttype aircraft has the different characteristics in spectral peak, which is the preparation workfor speech suppression and feature extraction.
Secondly, the target signal is the background sound of aircraft cockpit and not the pilots’speech. Here the speech becomes serious interference which must be removed. Speechdetection algorithm in noise conditions based on spectral entropy is used to get speechsegment. For suppressing speech, For suppressing speech, two algorithms, empirical modedecomposition and wavelet transform(EMD_WT) and the algorithm of ensemble empiricalmode decomposition and wavelet packet (EEMD_WP) was proposed. In the process ofspeech suppression, EMD_WT produces extra noise pollution and brought hidden trouble.EEMD_WP keeps the aircraft cockpit background sound and weakens speech greatly.Comparision tests prove that EEMD_WP has better performance both in time domain and infrequency domain.
Thirdly, the feature extraction is an important work in aircraft type recognition. In theview of the auditory perception, feature is extracted by bark wavelet packet. According to the procedure of Mel cepstrum coefficient and physical characteristics, nodes of wavelet packetdecomposition are selected by diagonal slice mode. The above features are combined withtime sequential Delta feature of MFCC and PLP. Pricipal component analysis is used toreduce the redundancy of high order features. The performance of all features components areestimated by Fisher criterion which produces brief and multi-angle features. The experimentsshow that the The mixed features have better classification performance than signal featureand implement the modeling of the aircraft cabin background sound, which makes preparationfor the further research on aircraft type recognition.
Finally, to overcome the problem in parameter redundancy of Gaussian mixturemodel(GMM), general gradient learning algorithm is used to implement BayesianYing-Yang(BYY) harmony learning and model selection. In the process, number of Gaussiancomponent and parameter estimation are finished automatically. However, general gradientlearning algorithm easily traps into a local maximum value of the harmony function so thatthe modeling or clustering result is not reasonable. In order to overcome this disadvantage,natural gradient learning algorithm based on Riemann manifold was proposed to implementBYY harmony learning. The method can realize automated model selection and number ofGMM component can be pruned in various degrees effeciently. Moreover, the speed ofconverges is more quickly and accurately than general gradient learning algorithm. Furthermore, it makes space distribution fittting better for target signal and improve recognitionperformance in very few information.
首先，为区分目标信号的来源是地面目标还是飞机目标，本文利用Welch算法，对地面信号/飞机信号进行了目标信号的辨识，结果表明地面目标与飞机目标存在着很大的差异性，实现了目标信号辨识。对飞机驾驶舱背景声进行特性分析，是飞机类型识别工作中的首要任务。为分析飞机驾驶舱背景声信号相关物理特性，提出集合均值和集合方差的分析方法，应用在目标信号的频域以及倒谱上，对飞机驾驶舱背景声信号进行相关物理特性分析，结果表明不同飞机类型的驾驶舱背景声的谱峰特性差异明显，为语音抑制、特征提取以及类型识别做了积极的铺垫。
其次，在原始信号中包含有大段清晰的飞行员话音，此处语音相对于飞机背景声成为了强干扰噪声，因此需要去除。利用了基于谱熵的语音端点检测，对所检测到的原始信号流中的语音段采用两种方式对目标信号进行语音抑制。利用经验模态分解与小波分析（EMD_WT）进行语音抑制，但在其语音抑制的过程中引入了大量的过程噪声，为后续的相关信号处理带来了很大的隐患。于是提出利用全局经验模态分解与小波包分解结合（EEMD_WP）实现语音抑制。通过时域与频域的对比，EEMD_WP较EMD_WT方式取得了较好的性能，对语音进行了有效的抑制，增强了背景声。
第三，特征提取是飞机类型识别工作中的一个重要内容。从听觉感知的角度对飞机驾驶舱背景声进行了特征提取的研究。依据MFCC提取流程，提出用基于听觉感知的Bark小波包提取特征，结合所分析的目标信号的物理特性，利用对角切片的方式选取小波包结点，得到了体现目标信号物理特性的特征，并结合具有时变特性的MFCC、PLP的Delta特征，为减少特征参数的高阶冗余，经过PCA降维和F比评价得到简洁的多角度的混合特征参数，实验表明，它比单种特征有更好的分类性能，实现对目标信号进行建模的目的，为飞机类型识别作必要准备。
最后，针对飞机的高斯混合模型(GMM)参数冗余的问题，利用基于贝叶斯阴阳(BYY)和谐学习理论的普通梯度算法进行模型选择，实现模型参数估计及阶数的自动选择。而普通梯度学习算法往往会陷入和谐函数的局部极值而导致不理想的结果，提出基于黎曼流形的自然梯度的贝叶斯阴阳和谐学习的模型选择算法。结果表明，基于自然梯度的模型选择算法不仅能够实现自动模型选择，对模型复杂度进行了不同程度的有效压缩，并且自然梯度学习的收敛性能优于普通梯度学习算法，使高斯混合模型对飞机目标信号的分布空间的拟合程度更进一步，提高了极少信息量下的识别性能。
Aircraft type recognition based on shortwave speech communication, is a new topic innon-cooperation field, which is different from radar and image means. The former has moredetection operating range. However, the way to identify aircraft type mainly is by experiencedprofessionals and no open relevant literatures are reported about it. Combined with theNational Natural Science Foundation ‘Research on aircraft type recognition based onshortwave speech communication’, based on the research of the target signal, the relatedresearch work on aircraft type recognition is conducted.
Firstly, to identify the target signal, Welch algorithm is used to analyze the backgroundsignal/aircraft signal in this paper. The result shows that there exists a lot of differencebetween the ground signal and aircraft signal, which realizes the location of target signal inthe aircraft type recognition. The physical characteristics of aircraft cockpit background soundis the primary task of studied based on shortwave speech communication. Ensemble mean andensemble variance are defined and this paper applys them to frequency domain and cepstrum,which analyzes physical characteristics of background sound. The result reveals that differenttype aircraft has the different characteristics in spectral peak, which is the preparation workfor speech suppression and feature extraction.
Secondly, the target signal is the background sound of aircraft cockpit and not the pilots’speech. Here the speech becomes serious interference which must be removed. Speechdetection algorithm in noise conditions based on spectral entropy is used to get speechsegment. For suppressing speech, For suppressing speech, two algorithms, empirical modedecomposition and wavelet transform(EMD_WT) and the algorithm of ensemble empiricalmode decomposition and wavelet packet (EEMD_WP) was proposed. In the process ofspeech suppression, EMD_WT produces extra noise pollution and brought hidden trouble.EEMD_WP keeps the aircraft cockpit background sound and weakens speech greatly.Comparision tests prove that EEMD_WP has better performance both in time domain and infrequency domain.
Thirdly, the feature extraction is an important work in aircraft type recognition. In theview of the auditory perception, feature is extracted by bark wavelet packet. According to the procedure of Mel cepstrum coefficient and physical characteristics, nodes of wavelet packetdecomposition are selected by diagonal slice mode. The above features are combined withtime sequential Delta feature of MFCC and PLP. Pricipal component analysis is used toreduce the redundancy of high order features. The performance of all features components areestimated by Fisher criterion which produces brief and multi-angle features. The experimentsshow that the The mixed features have better classification performance than signal featureand implement the modeling of the aircraft cabin background sound, which makes preparationfor the further research on aircraft type recognition.
Finally, to overcome the problem in parameter redundancy of Gaussian mixturemodel(GMM), general gradient learning algorithm is used to implement BayesianYing-Yang(BYY) harmony learning and model selection. In the process, number of Gaussiancomponent and parameter estimation are finished automatically. However, general gradientlearning algorithm easily traps into a local maximum value of the harmony function so thatthe modeling or clustering result is not reasonable. In order to overcome this disadvantage,natural gradient learning algorithm based on Riemann manifold was proposed to implementBYY harmony learning. The method can realize automated model selection and number ofGMM component can be pruned in various degrees effeciently. Moreover, the speed ofconverges is more quickly and accurately than general gradient learning algorithm. Furthermore, it makes space distribution fittting better for target signal and improve recognitionperformance in very few information.
引文
[1] Huange Wang; Xiaoguang Gao; Thompson, C.P. Aircraft Target Recognition Based onRecursive Inference of Fuzzy Discrete DBNs. International Conference on AdvancedComputer Control,2009:183-187P
[2] Zunhua Guo, Shaohong Li.One-dimensional frequency-domain features for aircraftrecognition from radar range profiles, IEEE Transactions on Aerospace and ElectronicSystems,2010,46(4):1880-1892P
[3] Saidi, M.N. Daoudi, K.; Khenchaf, A.; Hoeltzener, B.; Aboutajdine, D. Automatictarget recognition of aircraft models based on ISAR images.Geoscience and RemoteSensing Symposium, IEEE International,IGARSS2009,4: IV-685, IV-688P
[4] Saidi, M.N.; Toumi, A.; Hoeltzener, B.; Khenchaf, A.; Aboutajdine, D. Aircraft TargetRecognition: A novel approach for features extraction from ISAR images. RadarConference-Surveillance for a Safer World,2009. RADAR. International2009:1-5P
[5] Secmen, M, Turhan-Sayan, G, Hizal, A. A radar target recognition method with MUSICalgorithm: Application to aircraft targets with measured scattered data.2008IEEE RadarConference, Rome, Italy,2008,1-6P
[6] Toumi, A. Khenchaf, A. Hoeltzener, B.A retrieval system from inverse synthetic apertureradar images: Application to radar target recognition. Information Sciences,2012,196(8):73-96P
[7]李新德,杨伟东,DEZERT Jean.一种飞机图像目标多特征信息融合识别方法.自动化学报,2012,38(8):1298-1307P
[8]王宝帅,杜兰,刘宏伟等.基于经验模态分解的空中飞机目标分类.电子与信息学报,2012,34(9):2116-2121P
[9]李勇.王德功.杨佐龙.模糊神经网络与证据理论的飞机目标敌我识别.吉林大学学报(信息科学版),2012,30(1):78-82P
[10]韩兆林.基于Rough/SVM的飞机目标识别模型.飞机设计,2011,31(5):68-72P
[11] Zhang Feng, Liu Shang-qian, Wang da-bao,et al. Aircraft recognition in infrared imageusing wavelet moment invariants. Image and Vision Computing,2009,27(4);313-318P
[12]林霖.基于几何特征的飞机部位识别跟踪算法.红外,2007,28(8):23-26页
[13]张守娟,周诠.基于加权小波分形的卫星遥感图像飞机类型识别.微电子学与计算机,2007,24(6):173-175页
[14]吴秋荣,杨万麟.基于NMFs-LDA的雷达目标距离像识别.现代电子技术,2007,19:63-65P
[15] Savy,L., Gaie, C. Model-Based Classification of Aircraft Range Profiles using DataAssociation Algorithms. Radar Conference,2007IEEE,2007,4:822-827P
[16] Fernandez, L.P.S,Ruiz, A.R,Pogrebnyak, O.B. Noise monitoring of aircrafts taking offbased on neural model. IEEE Conference on Emerging Technologies&FactoryAutomation,2009:1-8P
[17] Vigliotti Andrea,Sorrentino Assunta. Design and testing of an Active Constrained LayerDamping demonstrator for aircraft's cabin noise reduction.38th International Congress andExposition on Noise Control Engineering,2009,3:1763-1772P
[18] Xiao Hu Clark, G.Travis, M.et al.Aircraft cabin noise minimization via neural networkinverse model.Neural Networks,2005. IJCNN '05. Proceedings.2005IEEE InternationalJoint Conference on.2005,4(4):2341-2346P
[19] Grewal A.Zimcik DG,Leigh B. Feedforward Piezoelectric Structural Control: AnApplication to Aircraft Cabin Noise Reduction.Journal of Aircraft,2001,38(1):164-173P
[20]乔渭阳,Ulf Michel.基于改进的传声器阵列数据分析技术的飞机集体噪声实验研究.航空学报.2008,29(3):527-533页
[21] HSIAO F B,HAN S Y,HSIEH S C,et al.Sound source separation and identification foraircraft cockpit voice recorder.Journal of Aerospace computing, Information andCommunication,2004(12):466-483P
[22]高谦,刘马宝,岳开宪.基于算法级联的飞机座舱语音增强系统.航空学报,2009,30(7):1203-1207页
[23]雷鸣,李学仁,李果.飞机舱音背景声下的鲁棒语音端点检测.振动与冲击,2008,27(10):883-886页
[24] Xiangbin Li, Guo Li, Xueren Li. Improved Voice Activity Detection Based on IterativeSpectral Subtraction and Double Thresholds for CVR.Power Electronics and IntelligentTransportation System, PEITS '08. Workshop on,2008
[25]姚永强,易本顺,姚远.航空噪声背景下的语音端点检测和语音增强.电声技术.2006,26(1):36-39页
[26] James R.Cash.Group Chairmans Report of Investigation Sound Spectrum study of CockpitVoice Recorder,American Airline Fight587,DCA02MA001,Belle Harbor,NY,2001,11-12P
[27]乔渭阳,徐开富,武兆伟等.大型客机起飞着陆过程噪声辐射特性对比分析.航空学报,2008,29(3):534-541页
[28] Fung,Andrew K.; Davis,Evan B. Prediction of airplane aft-cabin noise using statisticalenergy analysis. Acoustical Society of America Journal,2005,118(3):1847-1847P
[29] Suzanne D.Smith. Dynamic characteristics and human perception of vibration aboard amilitary propeller aircraft. International Journal of Industrial Ergonomics.2008,38:868-879P
[30] Mathur,Gopal; Fuller,Christopher. Recent advances in active control of aircraft cabin noise.The Journal of the Acoustical Society of America,2002,112(5):2346-2346P
[31] Volker Mellert, Ingo Baumann, Nils Freese, Reinhard Weber.Impact of sound and vibrationon health, travel comfort and performance of flight attendants and pilots.Aerospace Scienceand Technology.2008,12(1):18-25P
[32] Cheng Dao-Lai,Yi Chui-Jie,Yao, Hong-Yu.De-noise comprehensive research on airplanecockpit signals recorded by CVR. Journal of Computers,2011,6(9):1920-1925P
[33]丁文婷,徐涛,杨国庆.单个飞机噪声事件最小顶点覆盖模型的几场噪声监测点分布方法.噪声与振动控制,2012,3：166-170页
[34]仪垂杰,程道来,郭健翔等.基于不同方法的飞机舱音背景声频谱特征的获取与分析.振动与冲击,2007,30(4):9-13页
[35] Yi Chui-Jie, Cheng,Dao-Lai, Guo Jian-Xiang, Yao Hong-Yu, Yang Lin.Analyses oncharacteristic frequencies of background sound in cockpit of airplane based on differentmethods.Journal of Vibration and Shock.2007,26(8):109-111,121P
[36] Cheng, Dao Lai,Wang, Qing Cheng, Yi, Chui Jie, et al.Analysis and research for airplanecockpit Sound'ICA denoise based on blind source separation principle. Advanced MaterialsResearch,2011,204-210(2011):209-215P
[37] Cheng Daolai, Yi Chuijie, Guo Jianxiang, Yao Hongyu,Yang Lin, ZhongMinzhu.Nonstationary quiver spindle background sound analysis of airplane viawigner-ville and wavelet time-scale distribution.Journal of Mechanical Engineering.2007,43(5):150-154P
[38]刘志坤,唐小明.滑窗式谱减法在航空降噪中的应用.海军航空工程学院学报.2008,23(4):419-422页
[39] N.E.Huang. The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinearand Non-stationary Time Series Analysis, J.Proc.R.Soc.Lond.A,1998,45(4):903-995P
[40] Norden E.Huang, Samuel S.P.Shen. Hilbert-Huang Transform and Its Applications.Interdisciplinary Mathematical Sciences,2005,5
[41] WU Zhaohua,Huang E. Ensemble Empirical Mode Decomposition:A Noise-Assisted DataAnalysis Method.Advances in Adaptive Data Analysis,2009,1(1):1-41P
[42] Huang N E. A confidence limit for the Empirical Mode Decomposition and Hilbert spectralanalysis．J.proc.R.Soc.Lond2003(459):2317-2345P
[43] N. E. Huang, Z. Shen, R. S. Long.A new view of nonlinear water waves—the Hilbertspectrum, Ann. Rev. Fluid Mech.1999,31:417–457P
[44] Rilling, G.; Flandrin, P. On the Influence of Sampling on the Empirical ModeDecomposition, IEEE International Conference on Acoustics, Speech and SignalProcessing,2006,3:444-447P
[45] Kopsinis, Y. McLaughlin, S.. Investigation of the Empirical Mode Decomposition Basedon Genetic Algorithm Optimization Schemes. IEEE International Conference onAcoustics, Speech and Signal Processing,2007,3:1397-1400P
[46]盖强,殷福亮,邹岩崑.二维局域波分析在图像消噪中的应用研究.电子学报,2008,36(9):1763-1765页
[47]盖强,殷福亮.二维Hilbert-Huang变换的分解方法研究.电子与信息学报,2006,28(4):610-613页
[48]钟佑明,秦树人,汤宝平.Hilbert-Huang变换中的理论研究.振动与冲击.2002,21(4):13-17页
[49]钟佑明,秦树人.希尔伯特-黄变换的统一理论依据研究.振动与冲击.2006,25(3):40-44页
[50]钟佑明,秦树人,汤宝平.希尔伯特-黄变换中边际谱的研究.系统工程与电子技术.2004,26(9):1323-1326页
[51]钟佑明,秦树人.HHT的理论依据探讨——Hilbert变换的局部乘积定理.振动与冲击.2006,25(2)12-15页
[52] Qian T,Chen Q H and Li L Q. Analytic unit quadrature signals with nonlinearphase.Physica D:Nonlinear Phenomena.2005,203(1):80-87P
[53] Qian T. Analytic signals and harmonic measures. Journal of Mathematical Analysis andApplications,2006,314(2):526-536P
[54]孙晖,朱善安.基于过零点-极点估计的瞬时频率幅度算法.电子与信息学报.2006,28(5):905-908页
[55]郭明威,倪世宏,朱家海.振动信号中HHT/EMD端点延拓方法研究.振动与冲击.2012,31(8):62-66页
[56]孙艳争,黄炜,余波.基于EMD的非线性信号自适应分析.电子科技大学学报,2007,36(1):24-26页
[57]高强,李良敏等.EMD趋势分析方法及其应用研究.振动与冲击,2007,26(8):98-100P
[58]许宝杰,张建民,徐小力.抑制EMD端点效应方法的研究.北京理工大学学报.2006,26(3):196-200页
[59]李雪耀,张汝波,王武.基于支持向量回归机的HHT边界效应处理.智能系统学报.2007,2(3):39-44页
[60]舒忠平,杨智春.抑制经验模分解边缘效应的极值点对称延拓法.西北工业大学学报.2006,24(5):639-643页
[61]谢桂海,李浩,杨磊本征模式函数与经验模式分解方法分析,军械工程学院学报.2007,19(1):61-65页
[62]陈隽,徐幼麟.经验模分解在信号趋势项提取中的应用.振动、测试与诊断.2005,25(2):101-153页
[63]盖广洪.经验模态分解的一种改进算法.西安交通大学学报.2004,38(11):1199-1202页
[64] Flandrin P, Rilling G, Goncalvés P. Empirical mode decomposition as a filter bank. IEEESignal Processing Letters,2004,11(2):112-114P
[65] Semion Kizhner, Thomas P. Flatley,ed. On the Hilbert-Huang Transform data processingsystem development, IEEE Aerospace Conference Proceedings,2004
[66] Kizhner S, Blank K, ed. On Certain Theoretical Developments Underlying theHilbert-Huang Transform, IEEE Aerospace Conference,2006:1-14P
[67] Penga Z K, Peter W,ed. An improved Hilbert-Huang transform and its application invibration signal analysis. Journal of Sound and Vibration,2005：187-205P
[68] Chen Q H, Huang N E, Riemenschneider S, Xu Y S. A B-spline approach for empiricalmode decompositions. Advadces in Computational Mathematics,2004,10:1-25P
[69] Zhang H Y, Zhu L J, Zou Y K. Analysis of time-frequency conglomeration ofHilbert-Huang transform. Proceedings Of The8th Joint Conference On InformationSciences,2005,1(3):636-639P
[70] Cai D C, Zou Y, Zhang K. The improvement of the HHT for processing nonlinear signals.6th International Symposium On Test And Measurement,2005,1(9):7526-7529P
[71] Z. Wu and N. E. Huang.A study of the characteristics of white noise using the empiricalmode decomposition method.Proc. Roy. Soc. London. A.2004,460:1597–1611P
[72] WU Zhaohua,Huang E. Ensemble Empirical Mode Decomposition:A Noise-Assisted DataAnalysis Method.Advances in Adaptive Data Analysis,2009,1(1):1-41P
[73]李大虎,赖敏,何强等.基于聚类经验模态分解（EEMD）的汶川Ms8.0强震动记录时频特性分析.地震学报,2012,34(3):350-362页
[74] Guo, W,Tse, PW. A novel signal compression method based on optimal ensembleempirical mode decomposition for bearing vibration signals. Journal of sound and vibration.2012,332(2):423-441P
[75] Guo Wei,Tse Peter W. and Djordjevich Alexandar.Faulty bearing signal recovery fromlarge noise using a hybrid method based on spectral kurtosis and ensemble empirical modedecomposition. Measurement,2012,45(5):1308-1322P
[76]徐可君,秦海勤,江龙平.基于EMD和HHT的航空发动机转子-机匣振动信号分析,2011,30(7):237-240页
[77] Chen Xiangxian, Huang Hai and Zhang Jiafang. In-car speech enhancement based onensemble empirical mode decomposition. Computer Systems Science AndEngineering,2011,26(1):39-46P
[78] Liao, AH,Shen, CC,Li, PC. Potential Comparision Improvement in Ultrasound PulseInversion Imaging Using EMD and EEMD. Ieee Transactions On UltrasonicsFerroelectrics And Frequency Control,2010,57(2):317-326P
[79] Lei Xu, A. Krzyzak&E.Oja,(1993),"Rival Penalized Competitive Learning for ClusteringAnalysis, RBF net and Curve Detection", IEEE Trans. on Neural Networks,1993,14(4):636-649P
[80] Lei Xu.Bayesian-Kullback Coupled YING-YANG Machines: Unified Learnings and NewResults on Vector Quantization. Proceedings of International Conference on NeuralInformation Processing, Keynote Speaker, Beijing, China,1995,10:977-988P
[81] Lei Xu. Fundamentals, Challenges, and Advances of Statistical Learning for KnowledgeDiscovery and Problem Solving: A BYY Harmony Perspective. Proceedings ofInternational Conference on Neural Networks and Brain,2005,1:24-55P
[82] Lei Xu. Advances on BYY Harmony Learning: Information Theoretic Perspective,Generalized Projection Geometry, and Independent Factor Auto-determination. IEEE Transon Neural Networks,2004,15(4):885-902P
[83] Lei Xu. Trends on Regularization and Model Selection in Statistical Learning: APerspective from Bayesian Ying Yang Learning. Studies in ComputationalIntelligence,2007,63:365-406P
[84] Lei Xu.Bayesian Ying-Yang Machine, Clustering and Number of Clusters", PatternRecognition Letters,1997,18(11-13):1167-1178P
[85] Lei Xu. New Advances on Bayesian Ying-Yang Learning System With Kullback andNon-Kullback Separation Functionals.Proceedings of1997IEEE-(INNS) Conference onNeural Networks, Houston, TX,1997,3,1942-1947P
[86] Lei Xu.RBF Nets, Mixture Experts, and Bayesian Ying-Yang Learning, Neurocomputing,1998,19(1-3):223-257P
[87] Lei Xu.Bayesian Ying-Yang Dimension Reduction and Determination. Journal ofComputational Intelligence in Finance, Vol.6, No.5, a special issue on Complexity andDimensionality Reduction in Finance.1998,6-18P
[88] Lei Xu.Independent Component Analysis and Extensions with Noise and Time: A BayesianYing-Yang Learning Perspective.Neural Information Processing-Letters and Reviews,2003,1(1):1-52P
[89] SUN Ke and Lei Xu. Bayesian Ying-Yang learning on orthogonal binary factoranalysis. Neural Network World.2009:19(5):611-624P
[90] Lei Xu.Bayesian Ying-Yang system, best harmony learning, and five action circling, Aspecial issue on Emerging Themes on Information Theory and Bayesian Approach, Journalof Frontiers of Electrical and Electronic Engineering in China.2010,5(3):281-328P
[91] Shi, L, Tu S, Lei Xu, Gene clustering by structural prior based local factor analysis modelunder Bayesian Ying-Yang harmony learning. In: Proceedings of the BIBM2010International on Bioinformatics and Biomedicine, Hong Kong,2010,12,18-21,696-699P
[92] Penghui WANG, Lei SHI, Lan DU, Hongwei LIU, Lei Xu, Zheng BAO.Radar HRRPstatistical recognition with temporal factor analysis by automatic Bayesian Ying-Yangharmony learning.A special issue on Machine learning and intelligence science:IScIDE2010(B), Journal of Frontiers of Electrical and Electronic Engineering in China6(2),2011,300–317P
[93] Shi L, Wang P, Liu H, Lei Xu, and Bao Z(2011), Radar HRRP Statistical Recognition WithLocal Factor Analysis by Automatic Bayesian Ying-Yang Harmony Learning, IEEE Trans.Signal Process.,2011,59(2):610–617P
[94] Lei Xu.Another perspective of BYY harmony learning: representation in multiple layers,co-decomposition of data covariance matrices, and applications to network biology. Aspecial issue on Machine learning and intelligence science: IScIDE2010(A), Journal ofFrontiers of Electrical and Electronic Engineering in China6(1),2011,86-119P
[95] Lei Xu.On essential topics of BYY harmony learning: Current status, challenging issues,and gene analysis applications, A special issue on Machine learning and intelligencescience: IScIDE (C). Journal of Frontiers of Electrical and Electronic Engineering7(1),2012,147-196P
[96] Lei Xu.Semi-Blind Bilinear Matrix System, BYY Harmony Learning, and Gene AnalysisApplications.Proc. of6th International Conf. on New Trends in Information Science,Service Science and Data Mining, Taipei,2012,10:661-666P
[97] Lei Xu. Temporal BYY Encoding, Markovian State Spaces, and Space DimensionDetermination.IEEE Trans on Neural Networks,2004,15(5):1276-1295P
[98] Ma, J, Wang, T, and Lei Xu. A gradient BYY harmony learning rule on Gaussian mixturewith automated model selection.Neurocomputing,2004,56:481-487P
[99] Lei Xu. BYY learning, regularized implementation, and model selection on modularnetworks with one hidden layer of binary units.Neurocomputing,2003,51:277-301P
[100] Lei Xu.BYY Learning System and Theory for Parameter Estimation, Data SmoothingBased Regularization and Model Selection. Neural, Parallel and ScientificComputations.2000,8:55-82P
[101] Wing-Kai Lam,Ning Ouyang,Lei Xu. Application of Bayesian Ying-Yang criteria forselecting the number of hidden units with back propagation learning to electrocardiogramclassification. Proceedings of Fourteenth International Conference on PatternRecognition,1998,2:1686-1688P
[102] Lum, Nguyen, Shi,D. GA based FCMAC-BYY model for bank solvency analysis. IEEECongress on Evolutionary Computation,2007,1221-1225P
[103] Nguyen, Shi, D. Quek. FCMAC-BYY:Fuzzy CMAC Using Bayesian Ying–Yang Learning.IEEE Congress on Evolutionary Computation,2006:2375-2381P
[104] M.N. Nguyen, D. Shi, J. Fu..An online Bayesian Ying–Yang learning applied to fuzzyCMAC. Neurocomputing,2008,72(1-3):562-572P
[105] Lei Li, Jinwen Ma.BYY scale-incremental EM algorithm for Gaussian mixturelearning.Applied Mathematics and Computation,2008,205(2):832-840P
[106] Jinwen Ma, Xuefeng He.A fast fixed-point BYY harmony learning algorithm on Gaussianmixture with automated model selection.Pattern Recognition Letters,2008,29(6):701-711P
[107] Douar, M. A.,Mekhaldi, A.,Bouzidi, M. C.Welch Periodogram Analysis of theLeakage Current on Insulator Model under Wetted Contaminated Conditions forFlashover Prediction. IEEE Mediterranean Electrotechnical Conference-MELECON,2010,1636-1641P
[108] Alkan, Ahmet,Yimaz, Ahmet S.Frequency domain analysis of power system transientsusing Welch and Yule-Walker AR methods. Energy Conversion AndManagement,2007,48(7):2129-2135P
[109] Drugman, Thomas; Bozkurt, Bans; Dutoit, Thierry. Causal-anticausal decomposition ofspeech using complex cepstrum for glottal source estimation. SpeechCommunication.2011,53(6):855-866P
[110] Kim, Hong Kook; Rose, Richard C.Cepstrum-Domain Model Combination Based onDecomposition of Speech and Noise Using MMSE-LSA for ASR in Noisy Environments.IEEE Transactions On Audio Speech And Language Processing,2009,17(4):704-713P
[111] Shannon C E. A mathematica1theory of communication. The Bell System Technica1Journal, l948,27(7):379-423P
[112] Lu, Yong, Li, Yourong, Xiao, Han. A sliding singular spectrum entropy method and itsapplication to gear fault diagnosis. Lecture Notes In ComputerScience,2008,5226:669-678P
[113] Chatlani, Navin; Soraghan, John J.EMD-Based Filtering (EMDF) of Low-Frequency Noisefor Speech Enhancement. IEEE Transactions On Audio Speech And LanguageProcessing,2012,20(4):1158-1166P
[114] Roy, Sujan Kumar,Molla, Md, Khademul Islam, et al.Pitch Estimation of Noisy SpeechSignals using EMD-Fourier Based Hybrid Algorithm. IEEE International Symposium onCircuits and Systems.2010,5,2658-2661P
[115] Chen Guanghua,Liu Junhai and Ye Jun. An improved method of endpoints detection basedon energy-frequency-value.2006Conference on High Density Microsystem Design andPackaging and Component Failure Analysis,2006:9-11P
[116] ZHANG Xinyu,LI Xueyao,ZHANG Rubo and LIU Guanqun. Aircraft Type RecognitionBased on Shortwave Speech Communication With Auditory Perception.2009First IEEEInternational Conference on Information Science and Engineering.2009,12:477-480P
[117] Bin G. F., Gao J. J., Li X. J,et al. Early fault diagnosis of rotating machinery based onwavelet packets-Empirical mode decomposition feature extraction and neural network.Mechanical Systems And Signal Processing,2012,27:696-711P
[118] Litvin Yevgeni, Cohen Israel.Single-Channel Source Separation of Audio Signals UsingBark Scale Wavelet Packet Decomposition. Journal Of Signal Processing Systems ForSignal Image And Video Technolog,2011,65(3):339-350P
[119] Qian Huaming,Ma Jichen.Research on fiber optic gyro signal de-noising based on waveletpacket soft-threshold. Journal Of Systems Engineering And Electronics,2009,20(3):607-612P
[120] Conrey, B,Pisoni, DB.Auditory-visual speech perception and synchrony detection forspeech and nonspeech signals. Journal Of The Acoustical Society OfAmerica,2006,119(6):4065-4073P
[121] Tao, Zhi,Zhao, He-Ming,Zhang, Xiao-Jun.Speech Enhancement Based on the Multi-Scalesand Multi-Thresholds of the Auditory Perception Wavelet Transform. Archives OfAcoustics,2011,36(3):519-532P
[122] K.D.Martin.Sound-Source Recognition: A Theory and Computational Model.PhDdissertation, Department of Electrical Engineering and Computer Science,Massachusetts Institute of Technology,1999
[123] D.W. Robinson and R.S.Dadson. A re-determination of equal-loudness relations forpure tones. British Journal of Applied Physics,1956,7:166-181P
[124] H.Fletcher. Auditory patterns．Reviews of Modern Physics,1940,12:47-65P
[125]彭园,王科俊,李雪耀.感知线性预测在水下目标分类中的应用研究.声学学报,2006,31(2):146-149页
[126] Gamliel Oron,Shallom Ilan D.Perceptual Time Varying Linear Prediction Model forSpeech Applications.2009IEEE International Conference On Acoustics, Speech, AndSignal Processing,2009,1-8:4601-4604P
[127] Greewood D D. A Cochlear frequency-position function for several species-29years later.J A coust Soc Am,1990,87(6):2592-2605P
[128] Farooq O,Datta S. Mel filter-like admissible wavelet packet stucture for speechrecognition.IEEE Signal Processing Letters,2001,8(7):196-198P
[129] Shao Renping, Guo Wanlin and Li Zongbin.A new method for identifying the geartransmission system based on higher order cumulant.Proceedings of the5thInternational Conference on Vibration Engineering,2002,9:207-211P
[130] Karhunen J, Joutsensalo J.Generalizations of principal component analysis,optimization problems, and neural networks. Neural Networks,1995,549-562P
[131] Aston, John A. D. Chiou, Jeng-Min Evans, Jonathan P.Linguistic pitch analysis usingfunctional principal component mixed effect models Journal Of The Royal StatisticalSociety Series C-Applied Statistics.2010,59:297-317P
[132]章万锋.基于PCA与LDA的说话人识别研究.浙江大学硕士学位论文,2004:58-60页
[133] Gao, Cui-Fang,Qiu, Zi-Xue,Wu, Xiao-Jun.A Novel Fuzzy Fisher Classifier for SignalPeptide Prediction. Protein And Peptide Letters,2011,18(8):831-838P
[134] Ramirez-Pacheco, Julio, Torres-Roman, Deni, Argaez-Xool, Jesus.Wavelet q-FisherInformation for Scaling Signal Analysis.Entropy.2012,14(8):1478-1500P
[135] Ramirez-Pacheco, Julio,Torres-Roman, Deni,Rizo-Dominguez, Luis,et al.WaveletFisher's Information Measure of1/f(alpha) Signals. Entropy,2011,13(9):1648-1663P
[136] D. A. Reynolds. A Gaussian Mixture Modeling Approach to Text-IndependentSpeaker Identification. Ph. D. Thesis, Georgia Institute of Technology, Atlanta, GA,1992,37(2):160-171P
[137] D. A. Reynolds, R. Rose. Robust Text-Independent Speaker Identification UsingGaussian Mixture Speaker Models. IEEE Trans. on Speech and Audio Processing,1995,3(1):72-83P
[138] Metkar Prashant,Cohen Aaronm, Parhi Keshab.Improved Approach for CalculatingModel Parameters in Speaker Recognition using Gaussian Mixture Models.Conference Record of the Asilomar Conference on Signals Systems andComputers,2010,567-570P
[139] Dempster A, Laird N, Rubin D. Maximum likelihood estimation from incomplete datavia EM algorithm. J. Royal Statistical Society Series B,1977,12(6):1-38P
[140] Blimes J. A gentle tutorial of the EM algorithm and its application to parameterestimation for Gaussian mixture and hidden markov model. TR-97-021, April,1998,3(8):120-125P
[141] Vondra Martin,Vich Robert.Recognition of Emotions in German Speech UsingGaussian Mixture Models. Lecture Notes in Artificial Intelligence,2009,5398:256-263P
[142]孙晓霞,刘晓霞,谢倩茹.模糊C-均值(FCM)聚类算法的实验.计算机应用与软件,2008,32(4):48-50页
[143]丁爱明.基于MFCC和GMM的说话人识别系统研究.河海大学硕士学位论文,2006:40-50页
[144] MA Jin-wen.Automated Model Selection(AMS) on Finite Mixture:a New Perspectivefor Data.Chinese Journal of Engineering Mathematics,2007,24(4):571-584P
[145]汪洋.BYY谐和学习的自然梯度算法及其应用.北京大学硕士学位论文,2004:4-18页
[146]张伟.BYY和谐学习与广义RBF网络在实际数据序列预测中应用,北京大学硕士学位论文,2008:11-16页
[147] Amari S,Cichocki A,Yang H H. A New Learning Algorithm for Blind SourceSeparation.In Advances in Neural Information Processing Systems. NIPS-1995.Boston, MA: MIT Press,1996,8:757-763P
[148] S. Amari, A. Cichocki, and H. H. Yang. Novel online adaptive learning algorithms forblind deconvolution using the natural gradient approach. In Proc. IFAC Symp. Syst.Identification,1997:1055–1062P
[149] Amari S. Natural Gradient Works Efficiently in Learning. Neural Computation,1998,10(2):251-276P
[150] Amari S., Douglas S.C. Why Natural Gradient?. Acoustics, Speech andSignalProcessing (ICASSP), IEEE International Conference,1998:1213-1216P
[151] S. Amari and H. Nagaoka, Methods of Information Geometry. AmericanMathematical Society, Providence, RI,2000
[152] Jinwen Ma, Bin Gao, Yang Wang And Qiansheng Cheng.Conjugate and NaturalGradient Rules For Byy Harmony Learning on Gaussian Mixture with AutomatedModel Selection. International Journal of Pattern Recognition and ArtificialIntelligence.2005,19(5):701–713P
[2] Zunhua Guo, Shaohong Li.One-dimensional frequency-domain features for aircraftrecognition from radar range profiles, IEEE Transactions on Aerospace and ElectronicSystems,2010,46(4):1880-1892P
[3] Saidi, M.N. Daoudi, K.; Khenchaf, A.; Hoeltzener, B.; Aboutajdine, D. Automatictarget recognition of aircraft models based on ISAR images.Geoscience and RemoteSensing Symposium, IEEE International,IGARSS2009,4: IV-685, IV-688P
[4] Saidi, M.N.; Toumi, A.; Hoeltzener, B.; Khenchaf, A.; Aboutajdine, D. Aircraft TargetRecognition: A novel approach for features extraction from ISAR images. RadarConference-Surveillance for a Safer World,2009. RADAR. International2009:1-5P
[5] Secmen, M, Turhan-Sayan, G, Hizal, A. A radar target recognition method with MUSICalgorithm: Application to aircraft targets with measured scattered data.2008IEEE RadarConference, Rome, Italy,2008,1-6P
[6] Toumi, A. Khenchaf, A. Hoeltzener, B.A retrieval system from inverse synthetic apertureradar images: Application to radar target recognition. Information Sciences,2012,196(8):73-96P
[7]李新德,杨伟东,DEZERT Jean.一种飞机图像目标多特征信息融合识别方法.自动化学报,2012,38(8):1298-1307P
[8]王宝帅,杜兰,刘宏伟等.基于经验模态分解的空中飞机目标分类.电子与信息学报,2012,34(9):2116-2121P
[9]李勇.王德功.杨佐龙.模糊神经网络与证据理论的飞机目标敌我识别.吉林大学学报(信息科学版),2012,30(1):78-82P
[10]韩兆林.基于Rough/SVM的飞机目标识别模型.飞机设计,2011,31(5):68-72P
[11] Zhang Feng, Liu Shang-qian, Wang da-bao,et al. Aircraft recognition in infrared imageusing wavelet moment invariants. Image and Vision Computing,2009,27(4);313-318P
[12]林霖.基于几何特征的飞机部位识别跟踪算法.红外,2007,28(8):23-26页
[13]张守娟,周诠.基于加权小波分形的卫星遥感图像飞机类型识别.微电子学与计算机,2007,24(6):173-175页
[14]吴秋荣,杨万麟.基于NMFs-LDA的雷达目标距离像识别.现代电子技术,2007,19:63-65P
[15] Savy,L., Gaie, C. Model-Based Classification of Aircraft Range Profiles using DataAssociation Algorithms. Radar Conference,2007IEEE,2007,4:822-827P
[16] Fernandez, L.P.S,Ruiz, A.R,Pogrebnyak, O.B. Noise monitoring of aircrafts taking offbased on neural model. IEEE Conference on Emerging Technologies&FactoryAutomation,2009:1-8P
[17] Vigliotti Andrea,Sorrentino Assunta. Design and testing of an Active Constrained LayerDamping demonstrator for aircraft's cabin noise reduction.38th International Congress andExposition on Noise Control Engineering,2009,3:1763-1772P
[18] Xiao Hu Clark, G.Travis, M.et al.Aircraft cabin noise minimization via neural networkinverse model.Neural Networks,2005. IJCNN '05. Proceedings.2005IEEE InternationalJoint Conference on.2005,4(4):2341-2346P
[19] Grewal A.Zimcik DG,Leigh B. Feedforward Piezoelectric Structural Control: AnApplication to Aircraft Cabin Noise Reduction.Journal of Aircraft,2001,38(1):164-173P
[20]乔渭阳,Ulf Michel.基于改进的传声器阵列数据分析技术的飞机集体噪声实验研究.航空学报.2008,29(3):527-533页
[21] HSIAO F B,HAN S Y,HSIEH S C,et al.Sound source separation and identification foraircraft cockpit voice recorder.Journal of Aerospace computing, Information andCommunication,2004(12):466-483P
[22]高谦,刘马宝,岳开宪.基于算法级联的飞机座舱语音增强系统.航空学报,2009,30(7):1203-1207页
[23]雷鸣,李学仁,李果.飞机舱音背景声下的鲁棒语音端点检测.振动与冲击,2008,27(10):883-886页
[24] Xiangbin Li, Guo Li, Xueren Li. Improved Voice Activity Detection Based on IterativeSpectral Subtraction and Double Thresholds for CVR.Power Electronics and IntelligentTransportation System, PEITS '08. Workshop on,2008
[25]姚永强,易本顺,姚远.航空噪声背景下的语音端点检测和语音增强.电声技术.2006,26(1):36-39页
[26] James R.Cash.Group Chairmans Report of Investigation Sound Spectrum study of CockpitVoice Recorder,American Airline Fight587,DCA02MA001,Belle Harbor,NY,2001,11-12P
[27]乔渭阳,徐开富,武兆伟等.大型客机起飞着陆过程噪声辐射特性对比分析.航空学报,2008,29(3):534-541页
[28] Fung,Andrew K.; Davis,Evan B. Prediction of airplane aft-cabin noise using statisticalenergy analysis. Acoustical Society of America Journal,2005,118(3):1847-1847P
[29] Suzanne D.Smith. Dynamic characteristics and human perception of vibration aboard amilitary propeller aircraft. International Journal of Industrial Ergonomics.2008,38:868-879P
[30] Mathur,Gopal; Fuller,Christopher. Recent advances in active control of aircraft cabin noise.The Journal of the Acoustical Society of America,2002,112(5):2346-2346P
[31] Volker Mellert, Ingo Baumann, Nils Freese, Reinhard Weber.Impact of sound and vibrationon health, travel comfort and performance of flight attendants and pilots.Aerospace Scienceand Technology.2008,12(1):18-25P
[32] Cheng Dao-Lai,Yi Chui-Jie,Yao, Hong-Yu.De-noise comprehensive research on airplanecockpit signals recorded by CVR. Journal of Computers,2011,6(9):1920-1925P
[33]丁文婷,徐涛,杨国庆.单个飞机噪声事件最小顶点覆盖模型的几场噪声监测点分布方法.噪声与振动控制,2012,3：166-170页
[34]仪垂杰,程道来,郭健翔等.基于不同方法的飞机舱音背景声频谱特征的获取与分析.振动与冲击,2007,30(4):9-13页
[35] Yi Chui-Jie, Cheng,Dao-Lai, Guo Jian-Xiang, Yao Hong-Yu, Yang Lin.Analyses oncharacteristic frequencies of background sound in cockpit of airplane based on differentmethods.Journal of Vibration and Shock.2007,26(8):109-111,121P
[36] Cheng, Dao Lai,Wang, Qing Cheng, Yi, Chui Jie, et al.Analysis and research for airplanecockpit Sound'ICA denoise based on blind source separation principle. Advanced MaterialsResearch,2011,204-210(2011):209-215P
[37] Cheng Daolai, Yi Chuijie, Guo Jianxiang, Yao Hongyu,Yang Lin, ZhongMinzhu.Nonstationary quiver spindle background sound analysis of airplane viawigner-ville and wavelet time-scale distribution.Journal of Mechanical Engineering.2007,43(5):150-154P
[38]刘志坤,唐小明.滑窗式谱减法在航空降噪中的应用.海军航空工程学院学报.2008,23(4):419-422页
[39] N.E.Huang. The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinearand Non-stationary Time Series Analysis, J.Proc.R.Soc.Lond.A,1998,45(4):903-995P
[40] Norden E.Huang, Samuel S.P.Shen. Hilbert-Huang Transform and Its Applications.Interdisciplinary Mathematical Sciences,2005,5
[41] WU Zhaohua,Huang E. Ensemble Empirical Mode Decomposition:A Noise-Assisted DataAnalysis Method.Advances in Adaptive Data Analysis,2009,1(1):1-41P
[42] Huang N E. A confidence limit for the Empirical Mode Decomposition and Hilbert spectralanalysis．J.proc.R.Soc.Lond2003(459):2317-2345P
[43] N. E. Huang, Z. Shen, R. S. Long.A new view of nonlinear water waves—the Hilbertspectrum, Ann. Rev. Fluid Mech.1999,31:417–457P
[44] Rilling, G.; Flandrin, P. On the Influence of Sampling on the Empirical ModeDecomposition, IEEE International Conference on Acoustics, Speech and SignalProcessing,2006,3:444-447P
[45] Kopsinis, Y. McLaughlin, S.. Investigation of the Empirical Mode Decomposition Basedon Genetic Algorithm Optimization Schemes. IEEE International Conference onAcoustics, Speech and Signal Processing,2007,3:1397-1400P
[46]盖强,殷福亮,邹岩崑.二维局域波分析在图像消噪中的应用研究.电子学报,2008,36(9):1763-1765页
[47]盖强,殷福亮.二维Hilbert-Huang变换的分解方法研究.电子与信息学报,2006,28(4):610-613页
[48]钟佑明,秦树人,汤宝平.Hilbert-Huang变换中的理论研究.振动与冲击.2002,21(4):13-17页
[49]钟佑明,秦树人.希尔伯特-黄变换的统一理论依据研究.振动与冲击.2006,25(3):40-44页
[50]钟佑明,秦树人,汤宝平.希尔伯特-黄变换中边际谱的研究.系统工程与电子技术.2004,26(9):1323-1326页
[51]钟佑明,秦树人.HHT的理论依据探讨——Hilbert变换的局部乘积定理.振动与冲击.2006,25(2)12-15页
[52] Qian T,Chen Q H and Li L Q. Analytic unit quadrature signals with nonlinearphase.Physica D:Nonlinear Phenomena.2005,203(1):80-87P
[53] Qian T. Analytic signals and harmonic measures. Journal of Mathematical Analysis andApplications,2006,314(2):526-536P
[54]孙晖,朱善安.基于过零点-极点估计的瞬时频率幅度算法.电子与信息学报.2006,28(5):905-908页
[55]郭明威,倪世宏,朱家海.振动信号中HHT/EMD端点延拓方法研究.振动与冲击.2012,31(8):62-66页
[56]孙艳争,黄炜,余波.基于EMD的非线性信号自适应分析.电子科技大学学报,2007,36(1):24-26页
[57]高强,李良敏等.EMD趋势分析方法及其应用研究.振动与冲击,2007,26(8):98-100P
[58]许宝杰,张建民,徐小力.抑制EMD端点效应方法的研究.北京理工大学学报.2006,26(3):196-200页
[59]李雪耀,张汝波,王武.基于支持向量回归机的HHT边界效应处理.智能系统学报.2007,2(3):39-44页
[60]舒忠平,杨智春.抑制经验模分解边缘效应的极值点对称延拓法.西北工业大学学报.2006,24(5):639-643页
[61]谢桂海,李浩,杨磊本征模式函数与经验模式分解方法分析,军械工程学院学报.2007,19(1):61-65页
[62]陈隽,徐幼麟.经验模分解在信号趋势项提取中的应用.振动、测试与诊断.2005,25(2):101-153页
[63]盖广洪.经验模态分解的一种改进算法.西安交通大学学报.2004,38(11):1199-1202页
[64] Flandrin P, Rilling G, Goncalvés P. Empirical mode decomposition as a filter bank. IEEESignal Processing Letters,2004,11(2):112-114P
[65] Semion Kizhner, Thomas P. Flatley,ed. On the Hilbert-Huang Transform data processingsystem development, IEEE Aerospace Conference Proceedings,2004
[66] Kizhner S, Blank K, ed. On Certain Theoretical Developments Underlying theHilbert-Huang Transform, IEEE Aerospace Conference,2006:1-14P
[67] Penga Z K, Peter W,ed. An improved Hilbert-Huang transform and its application invibration signal analysis. Journal of Sound and Vibration,2005：187-205P
[68] Chen Q H, Huang N E, Riemenschneider S, Xu Y S. A B-spline approach for empiricalmode decompositions. Advadces in Computational Mathematics,2004,10:1-25P
[69] Zhang H Y, Zhu L J, Zou Y K. Analysis of time-frequency conglomeration ofHilbert-Huang transform. Proceedings Of The8th Joint Conference On InformationSciences,2005,1(3):636-639P
[70] Cai D C, Zou Y, Zhang K. The improvement of the HHT for processing nonlinear signals.6th International Symposium On Test And Measurement,2005,1(9):7526-7529P
[71] Z. Wu and N. E. Huang.A study of the characteristics of white noise using the empiricalmode decomposition method.Proc. Roy. Soc. London. A.2004,460:1597–1611P
[72] WU Zhaohua,Huang E. Ensemble Empirical Mode Decomposition:A Noise-Assisted DataAnalysis Method.Advances in Adaptive Data Analysis,2009,1(1):1-41P
[73]李大虎,赖敏,何强等.基于聚类经验模态分解（EEMD）的汶川Ms8.0强震动记录时频特性分析.地震学报,2012,34(3):350-362页
[74] Guo, W,Tse, PW. A novel signal compression method based on optimal ensembleempirical mode decomposition for bearing vibration signals. Journal of sound and vibration.2012,332(2):423-441P
[75] Guo Wei,Tse Peter W. and Djordjevich Alexandar.Faulty bearing signal recovery fromlarge noise using a hybrid method based on spectral kurtosis and ensemble empirical modedecomposition. Measurement,2012,45(5):1308-1322P
[76]徐可君,秦海勤,江龙平.基于EMD和HHT的航空发动机转子-机匣振动信号分析,2011,30(7):237-240页
[77] Chen Xiangxian, Huang Hai and Zhang Jiafang. In-car speech enhancement based onensemble empirical mode decomposition. Computer Systems Science AndEngineering,2011,26(1):39-46P
[78] Liao, AH,Shen, CC,Li, PC. Potential Comparision Improvement in Ultrasound PulseInversion Imaging Using EMD and EEMD. Ieee Transactions On UltrasonicsFerroelectrics And Frequency Control,2010,57(2):317-326P
[79] Lei Xu, A. Krzyzak&E.Oja,(1993),"Rival Penalized Competitive Learning for ClusteringAnalysis, RBF net and Curve Detection", IEEE Trans. on Neural Networks,1993,14(4):636-649P
[80] Lei Xu.Bayesian-Kullback Coupled YING-YANG Machines: Unified Learnings and NewResults on Vector Quantization. Proceedings of International Conference on NeuralInformation Processing, Keynote Speaker, Beijing, China,1995,10:977-988P
[81] Lei Xu. Fundamentals, Challenges, and Advances of Statistical Learning for KnowledgeDiscovery and Problem Solving: A BYY Harmony Perspective. Proceedings ofInternational Conference on Neural Networks and Brain,2005,1:24-55P
[82] Lei Xu. Advances on BYY Harmony Learning: Information Theoretic Perspective,Generalized Projection Geometry, and Independent Factor Auto-determination. IEEE Transon Neural Networks,2004,15(4):885-902P
[83] Lei Xu. Trends on Regularization and Model Selection in Statistical Learning: APerspective from Bayesian Ying Yang Learning. Studies in ComputationalIntelligence,2007,63:365-406P
[84] Lei Xu.Bayesian Ying-Yang Machine, Clustering and Number of Clusters", PatternRecognition Letters,1997,18(11-13):1167-1178P
[85] Lei Xu. New Advances on Bayesian Ying-Yang Learning System With Kullback andNon-Kullback Separation Functionals.Proceedings of1997IEEE-(INNS) Conference onNeural Networks, Houston, TX,1997,3,1942-1947P
[86] Lei Xu.RBF Nets, Mixture Experts, and Bayesian Ying-Yang Learning, Neurocomputing,1998,19(1-3):223-257P
[87] Lei Xu.Bayesian Ying-Yang Dimension Reduction and Determination. Journal ofComputational Intelligence in Finance, Vol.6, No.5, a special issue on Complexity andDimensionality Reduction in Finance.1998,6-18P
[88] Lei Xu.Independent Component Analysis and Extensions with Noise and Time: A BayesianYing-Yang Learning Perspective.Neural Information Processing-Letters and Reviews,2003,1(1):1-52P
[89] SUN Ke and Lei Xu. Bayesian Ying-Yang learning on orthogonal binary factoranalysis. Neural Network World.2009:19(5):611-624P
[90] Lei Xu.Bayesian Ying-Yang system, best harmony learning, and five action circling, Aspecial issue on Emerging Themes on Information Theory and Bayesian Approach, Journalof Frontiers of Electrical and Electronic Engineering in China.2010,5(3):281-328P
[91] Shi, L, Tu S, Lei Xu, Gene clustering by structural prior based local factor analysis modelunder Bayesian Ying-Yang harmony learning. In: Proceedings of the BIBM2010International on Bioinformatics and Biomedicine, Hong Kong,2010,12,18-21,696-699P
[92] Penghui WANG, Lei SHI, Lan DU, Hongwei LIU, Lei Xu, Zheng BAO.Radar HRRPstatistical recognition with temporal factor analysis by automatic Bayesian Ying-Yangharmony learning.A special issue on Machine learning and intelligence science:IScIDE2010(B), Journal of Frontiers of Electrical and Electronic Engineering in China6(2),2011,300–317P
[93] Shi L, Wang P, Liu H, Lei Xu, and Bao Z(2011), Radar HRRP Statistical Recognition WithLocal Factor Analysis by Automatic Bayesian Ying-Yang Harmony Learning, IEEE Trans.Signal Process.,2011,59(2):610–617P
[94] Lei Xu.Another perspective of BYY harmony learning: representation in multiple layers,co-decomposition of data covariance matrices, and applications to network biology. Aspecial issue on Machine learning and intelligence science: IScIDE2010(A), Journal ofFrontiers of Electrical and Electronic Engineering in China6(1),2011,86-119P
[95] Lei Xu.On essential topics of BYY harmony learning: Current status, challenging issues,and gene analysis applications, A special issue on Machine learning and intelligencescience: IScIDE (C). Journal of Frontiers of Electrical and Electronic Engineering7(1),2012,147-196P
[96] Lei Xu.Semi-Blind Bilinear Matrix System, BYY Harmony Learning, and Gene AnalysisApplications.Proc. of6th International Conf. on New Trends in Information Science,Service Science and Data Mining, Taipei,2012,10:661-666P
[97] Lei Xu. Temporal BYY Encoding, Markovian State Spaces, and Space DimensionDetermination.IEEE Trans on Neural Networks,2004,15(5):1276-1295P
[98] Ma, J, Wang, T, and Lei Xu. A gradient BYY harmony learning rule on Gaussian mixturewith automated model selection.Neurocomputing,2004,56:481-487P
[99] Lei Xu. BYY learning, regularized implementation, and model selection on modularnetworks with one hidden layer of binary units.Neurocomputing,2003,51:277-301P
[100] Lei Xu.BYY Learning System and Theory for Parameter Estimation, Data SmoothingBased Regularization and Model Selection. Neural, Parallel and ScientificComputations.2000,8:55-82P
[101] Wing-Kai Lam,Ning Ouyang,Lei Xu. Application of Bayesian Ying-Yang criteria forselecting the number of hidden units with back propagation learning to electrocardiogramclassification. Proceedings of Fourteenth International Conference on PatternRecognition,1998,2:1686-1688P
[102] Lum, Nguyen, Shi,D. GA based FCMAC-BYY model for bank solvency analysis. IEEECongress on Evolutionary Computation,2007,1221-1225P
[103] Nguyen, Shi, D. Quek. FCMAC-BYY:Fuzzy CMAC Using Bayesian Ying–Yang Learning.IEEE Congress on Evolutionary Computation,2006:2375-2381P
[104] M.N. Nguyen, D. Shi, J. Fu..An online Bayesian Ying–Yang learning applied to fuzzyCMAC. Neurocomputing,2008,72(1-3):562-572P
[105] Lei Li, Jinwen Ma.BYY scale-incremental EM algorithm for Gaussian mixturelearning.Applied Mathematics and Computation,2008,205(2):832-840P
[106] Jinwen Ma, Xuefeng He.A fast fixed-point BYY harmony learning algorithm on Gaussianmixture with automated model selection.Pattern Recognition Letters,2008,29(6):701-711P
[107] Douar, M. A.,Mekhaldi, A.,Bouzidi, M. C.Welch Periodogram Analysis of theLeakage Current on Insulator Model under Wetted Contaminated Conditions forFlashover Prediction. IEEE Mediterranean Electrotechnical Conference-MELECON,2010,1636-1641P
[108] Alkan, Ahmet,Yimaz, Ahmet S.Frequency domain analysis of power system transientsusing Welch and Yule-Walker AR methods. Energy Conversion AndManagement,2007,48(7):2129-2135P
[109] Drugman, Thomas; Bozkurt, Bans; Dutoit, Thierry. Causal-anticausal decomposition ofspeech using complex cepstrum for glottal source estimation. SpeechCommunication.2011,53(6):855-866P
[110] Kim, Hong Kook; Rose, Richard C.Cepstrum-Domain Model Combination Based onDecomposition of Speech and Noise Using MMSE-LSA for ASR in Noisy Environments.IEEE Transactions On Audio Speech And Language Processing,2009,17(4):704-713P
[111] Shannon C E. A mathematica1theory of communication. The Bell System Technica1Journal, l948,27(7):379-423P
[112] Lu, Yong, Li, Yourong, Xiao, Han. A sliding singular spectrum entropy method and itsapplication to gear fault diagnosis. Lecture Notes In ComputerScience,2008,5226:669-678P
[113] Chatlani, Navin; Soraghan, John J.EMD-Based Filtering (EMDF) of Low-Frequency Noisefor Speech Enhancement. IEEE Transactions On Audio Speech And LanguageProcessing,2012,20(4):1158-1166P
[114] Roy, Sujan Kumar,Molla, Md, Khademul Islam, et al.Pitch Estimation of Noisy SpeechSignals using EMD-Fourier Based Hybrid Algorithm. IEEE International Symposium onCircuits and Systems.2010,5,2658-2661P
[115] Chen Guanghua,Liu Junhai and Ye Jun. An improved method of endpoints detection basedon energy-frequency-value.2006Conference on High Density Microsystem Design andPackaging and Component Failure Analysis,2006:9-11P
[116] ZHANG Xinyu,LI Xueyao,ZHANG Rubo and LIU Guanqun. Aircraft Type RecognitionBased on Shortwave Speech Communication With Auditory Perception.2009First IEEEInternational Conference on Information Science and Engineering.2009,12:477-480P
[117] Bin G. F., Gao J. J., Li X. J,et al. Early fault diagnosis of rotating machinery based onwavelet packets-Empirical mode decomposition feature extraction and neural network.Mechanical Systems And Signal Processing,2012,27:696-711P
[118] Litvin Yevgeni, Cohen Israel.Single-Channel Source Separation of Audio Signals UsingBark Scale Wavelet Packet Decomposition. Journal Of Signal Processing Systems ForSignal Image And Video Technolog,2011,65(3):339-350P
[119] Qian Huaming,Ma Jichen.Research on fiber optic gyro signal de-noising based on waveletpacket soft-threshold. Journal Of Systems Engineering And Electronics,2009,20(3):607-612P
[120] Conrey, B,Pisoni, DB.Auditory-visual speech perception and synchrony detection forspeech and nonspeech signals. Journal Of The Acoustical Society OfAmerica,2006,119(6):4065-4073P
[121] Tao, Zhi,Zhao, He-Ming,Zhang, Xiao-Jun.Speech Enhancement Based on the Multi-Scalesand Multi-Thresholds of the Auditory Perception Wavelet Transform. Archives OfAcoustics,2011,36(3):519-532P
[122] K.D.Martin.Sound-Source Recognition: A Theory and Computational Model.PhDdissertation, Department of Electrical Engineering and Computer Science,Massachusetts Institute of Technology,1999
[123] D.W. Robinson and R.S.Dadson. A re-determination of equal-loudness relations forpure tones. British Journal of Applied Physics,1956,7:166-181P
[124] H.Fletcher. Auditory patterns．Reviews of Modern Physics,1940,12:47-65P
[125]彭园,王科俊,李雪耀.感知线性预测在水下目标分类中的应用研究.声学学报,2006,31(2):146-149页
[126] Gamliel Oron,Shallom Ilan D.Perceptual Time Varying Linear Prediction Model forSpeech Applications.2009IEEE International Conference On Acoustics, Speech, AndSignal Processing,2009,1-8:4601-4604P
[127] Greewood D D. A Cochlear frequency-position function for several species-29years later.J A coust Soc Am,1990,87(6):2592-2605P
[128] Farooq O,Datta S. Mel filter-like admissible wavelet packet stucture for speechrecognition.IEEE Signal Processing Letters,2001,8(7):196-198P
[129] Shao Renping, Guo Wanlin and Li Zongbin.A new method for identifying the geartransmission system based on higher order cumulant.Proceedings of the5thInternational Conference on Vibration Engineering,2002,9:207-211P
[130] Karhunen J, Joutsensalo J.Generalizations of principal component analysis,optimization problems, and neural networks. Neural Networks,1995,549-562P
[131] Aston, John A. D. Chiou, Jeng-Min Evans, Jonathan P.Linguistic pitch analysis usingfunctional principal component mixed effect models Journal Of The Royal StatisticalSociety Series C-Applied Statistics.2010,59:297-317P
[132]章万锋.基于PCA与LDA的说话人识别研究.浙江大学硕士学位论文,2004:58-60页
[133] Gao, Cui-Fang,Qiu, Zi-Xue,Wu, Xiao-Jun.A Novel Fuzzy Fisher Classifier for SignalPeptide Prediction. Protein And Peptide Letters,2011,18(8):831-838P
[134] Ramirez-Pacheco, Julio, Torres-Roman, Deni, Argaez-Xool, Jesus.Wavelet q-FisherInformation for Scaling Signal Analysis.Entropy.2012,14(8):1478-1500P
[135] Ramirez-Pacheco, Julio,Torres-Roman, Deni,Rizo-Dominguez, Luis,et al.WaveletFisher's Information Measure of1/f(alpha) Signals. Entropy,2011,13(9):1648-1663P
[136] D. A. Reynolds. A Gaussian Mixture Modeling Approach to Text-IndependentSpeaker Identification. Ph. D. Thesis, Georgia Institute of Technology, Atlanta, GA,1992,37(2):160-171P
[137] D. A. Reynolds, R. Rose. Robust Text-Independent Speaker Identification UsingGaussian Mixture Speaker Models. IEEE Trans. on Speech and Audio Processing,1995,3(1):72-83P
[138] Metkar Prashant,Cohen Aaronm, Parhi Keshab.Improved Approach for CalculatingModel Parameters in Speaker Recognition using Gaussian Mixture Models.Conference Record of the Asilomar Conference on Signals Systems andComputers,2010,567-570P
[139] Dempster A, Laird N, Rubin D. Maximum likelihood estimation from incomplete datavia EM algorithm. J. Royal Statistical Society Series B,1977,12(6):1-38P
[140] Blimes J. A gentle tutorial of the EM algorithm and its application to parameterestimation for Gaussian mixture and hidden markov model. TR-97-021, April,1998,3(8):120-125P
[141] Vondra Martin,Vich Robert.Recognition of Emotions in German Speech UsingGaussian Mixture Models. Lecture Notes in Artificial Intelligence,2009,5398:256-263P
[142]孙晓霞,刘晓霞,谢倩茹.模糊C-均值(FCM)聚类算法的实验.计算机应用与软件,2008,32(4):48-50页
[143]丁爱明.基于MFCC和GMM的说话人识别系统研究.河海大学硕士学位论文,2006:40-50页
[144] MA Jin-wen.Automated Model Selection(AMS) on Finite Mixture:a New Perspectivefor Data.Chinese Journal of Engineering Mathematics,2007,24(4):571-584P
[145]汪洋.BYY谐和学习的自然梯度算法及其应用.北京大学硕士学位论文,2004:4-18页
[146]张伟.BYY和谐学习与广义RBF网络在实际数据序列预测中应用,北京大学硕士学位论文,2008:11-16页
[147] Amari S,Cichocki A,Yang H H. A New Learning Algorithm for Blind SourceSeparation.In Advances in Neural Information Processing Systems. NIPS-1995.Boston, MA: MIT Press,1996,8:757-763P
[148] S. Amari, A. Cichocki, and H. H. Yang. Novel online adaptive learning algorithms forblind deconvolution using the natural gradient approach. In Proc. IFAC Symp. Syst.Identification,1997:1055–1062P
[149] Amari S. Natural Gradient Works Efficiently in Learning. Neural Computation,1998,10(2):251-276P
[150] Amari S., Douglas S.C. Why Natural Gradient?. Acoustics, Speech andSignalProcessing (ICASSP), IEEE International Conference,1998:1213-1216P
[151] S. Amari and H. Nagaoka, Methods of Information Geometry. AmericanMathematical Society, Providence, RI,2000
[152] Jinwen Ma, Bin Gao, Yang Wang And Qiansheng Cheng.Conjugate and NaturalGradient Rules For Byy Harmony Learning on Gaussian Mixture with AutomatedModel Selection. International Journal of Pattern Recognition and ArtificialIntelligence.2005,19(5):701–713P