摘要
无人值守地面传感器网络(UGS)最早是在战争中发展起来的,对地面目标进行探测、分类、跟踪和定位的智能侦察系统。光纤传感器网络有着灵敏度高,不受电磁干扰,与光纤通信网络兼容,易于大规模组网等诸多优势,所以成为无人值守地面传感器系统的有力补充。
地震动传感器是地面传感器网络的重要组成部分,光纤地震动传感器网络成为战场环境监测的重要手段。在战场环境下,当地面目标运动时,所激发的地面振动信号沿地球表面向四面传播,到达传感器端,通过对信号的分析、处理,就可以实现对地面目标的识别、定位和跟踪等。本文讨论了战场环境下,光纤地震动传感器网络对地面目标的探测和识别相关问题。
本文首先研究了地面运动目标的地震动信号的发生和传播机理。对地面人员和车辆信号的目标特性进行了分析,分析了地面人员目标的探测方法,建立了地面车辆目标的信号发生的模型。然后,着重讨论地震波的传播、衰减、频散等各方面的影响,通过波动方程定性的分析了传播过程对信号本身特性及其识别带来的影响。然后,分析实际外场环境下可能的背景噪声状况,提出了基于自适应阈值的恒虚警率目标探测方法,在保证探测率的同时,极大程度的降低了系统虚警率。
在此基础上,对地面目标信号进行了特征的提取和优化。分别从信号的时频、频域和时频域分析了信号的基本特性。重点讨论了地面人员特征量的提取,运用自适应阈值小波去噪方法对地面人员信号进行了去噪处理,极大程度的提高了信号的信噪比。地面车辆信号在时频域具有明显的特征,与频域相结合,提出了多特征联合的特征提取方法。根据外场实际采集的目标样本,建立了大规模的目标特征库。
为了应对外界复杂情况,提高识别率,采用了基于统计学习理论和支持向量机的目标识别方法,通过与其它模式识别方法的对比实验,说明支持向量机的分类方法具有更好分类能力和泛化性,在目标识别领域有广阔的应用前景。
在光纤传感器网络中,多传感器协同处理是提高系统整体性能的基本方法,本文采用基于DS证据理论的判决融合方法,提高了系统的分类正确率。在实际外场应用中,多混叠目标的探测与识别问题是亟待研究和解决的问题。本文首次采用独立分量分析的方法对多混叠目标的地震动信号进行了分析,首先将混叠信号进行分离,然后再分别进行识别。建立了多混叠信号的基本模型,运用FastICA算法进行了仿真实验,说明独立成分分析是解决多目标问题的有效算法。
In this paper, the problems about ground target detection, classificationin fiber seismic senor network are discussed. The unattended ground sensor(UGS) system was developed in military for ground target detection, classificationand tracking in battle field surveillance. The fiber seismic senors have a lotof advantages such as high sensitivity, electromagnetic resistance and networkorganizing easiness. The fiber sensor network has been an important complementfor ground sensor network. In the battle field, the movement of ground targetsuch as personnel, tracked vehicle or wheeled vehicle motivates the seismic wavesthat propagating along with the ground surface. The seismic signal received bythe seismic senor can be analyzed for target classification.
First, the generation and propagation of seismic signal caused by movingtarget is discussed. An detection method of personnel footstep is discussed ac-cording to its impulse nature. Then a model of vehicle vibration when movingis built that can be used to analyze the signal characteristic. The propaga-tion model of seismic wave is more complicated to be analyzed. In propagationprocedure, the amplitude of the signal is attenuated, and the frequency has dis-persion. An improved constant false alarm rate (CFAR) detection method basedon adaptive threshold is used to decrease the false alarm in high backgroundnoise. Second, the feature extraction and selection of seismic signal is discussed.The traditional feature extraction methods in time domain, frequency domainand time-frequency domain are discussed, and an united multi-feature method isproposed. The principle component analysis (PCA) is used to reduce the dimen-sion of the feature vector. According to real-life ground target signal acquired inthe field experiment, a large feature data set of di?erent targets is built. Third,the ground target classification method based on statistical learning theory andsupport vector machine is discussed. The traditional classifiers such as Bayes clas-sifier, neural network have some disadvantages when the sample number is small.The comparison experiment shows that the support vector machine (SVM) clas- sifier outperforms the other method, and have bright future in this application.At last, the problems about multi-sensor and multi-target detection and recog-nition are discussed. The decision fusion strategy based on DS evidence theoryis used to improve the overall performance of sensor network. The multi-targetmixed signal processing is a di?cult problem to be tacked in sensor network. Weuse the independent component analysis (ICA) to separate the mixed signal, andthen to classify them separately. The emulation experiment shows that the ICAis an e?ective method.
引文
[1]马春婷,高萍,关士成,高文汉。地面战场侦察传感器系统的发展状况及技术分析。军械工程学院学报, 15(4):15-20, 2003
[2]聂伟荣。多传感器探测与网络控制技术。南京理工大学博士论文, 2001
[3] Kenneth Elgort, Scott Komaromy, John Madden, Andmv Taylor. Us ArmyEmployment Of Unattended Ground Sensors. Proceeding of 2005 Systemand Information Engineering Design Symposium, 2005
[4] Yan Zhang, Sanguo Li, Zhifan Yin, Hong-Liang Cui. Unattended groundsensor based on fiber Bragg grating technology. Proceedings of SPIE,5796:133-140, 2005
[5] Roger I.Crickmore, Philip J. Nash and John P.F. Wooler. Fibre Optic Se-curity Systems for Land and Sea Based Applications. Proceedings of SPIE,5611:79-86, 2004
[6]石荣,高培德,郑春雷,封松林。无线传感器网络技术的研究进展。中国科学院无线传感器网络技术学术研讨会,上海,2006
[7] Yan Zhang, Sanguo Li, Zhifan Yin, Robert Pastore, Kurt O’Donnell, MichaelPellicano, John Kosinski, and Hong-Liang Cui Fiber Bragg grating sensorsfor seismic wave detection. Proc. of SPIE, 5855:1008-1011, 2005
[8] George Succi, Gervasio Prado, Robert Gampert, Torstein Pedersen and Har-dave Dhaliwal. Problems in Seismic Detection and Tracking. Proceedingsof SPIE, 4040:165-173, 2000
[9] (美)Richard O. Duda, Peter E. Hart, David G. Stork。模式分类,第二版。机械工业出版社,北京,2003
[10] Gerard E. Sleefe, Mark D. Ladd, Timothy S. McDonald, and Gregory J.Elbring. Acoustic and Seismic modalities for unattended ground sensors.Proceedings of SPIE, 3713: , 1999
[11]黄琦。智能传感器侦察网络中的地面目标识别。中国科学技术大学博士论文, 2006
[12] Stephen A. Ketcham, Member, IEEE, Mark L. Moran, James Lacombe,Roy J. Greenfield, and Thomas S. Anderson. Seismic Source Model forMoving Vehicles. IEEE Transactions On Geoscience And Remote Sensing,43(2):248-256, 2005
[13] Thomas S. Anderson, Mark L. Moran, Stephen A. Ketcham, James La-combe, US Army Corps of Engineers. Tracked Vehicle Simulations and Seis-mic Wavefield Synthesis in Seismic Sensor Systems. Computing in Science& Engineering, 6(6):22-28, 2004
[14] Mark D. Ladd, M. Kathleen Alam, Gerard E. Sleefe, and Hung D. Nguyen.Seismic and acoustic signal identification algorithms. Proceedings of SPIE,4040:106-120, 2000
[15] Songhwai Oh, Phoebus Chen, etc. Instrumenting Wireless Sensor Networksfor Real-Time Surveillance. International Conference on Robotics and Au-tomation (ICRA), 2006
[16] Anne M. Mayoral, Ilene L. McCool, Richard A. Gramann. Detection andClassification of Time-Critical Targets Using Seismic Sensors. Proceedingsof SPIE, 4743:80-90, 2002
[17] Marco F. Duarte and Yu Hen Hu. Vehicle Classification in DistributedSensor Networks. Journal of Parallel and Distributed Computing, 64(7):826-833, 2004
[18] Anh Nguyen, Leng Sim, Ramaz Kvavilashvili. The U.S. Army ResearchLaboratory’s Acoustic & Seismic Database and User Interface Software.ADA390232, 1999
[19] Dan Li, Kerry D. Wong, Yu H. Hu and Akbar M. Sayeed. Detection, Clas-sification and Tracking of Targets in Distributed Sensor Networks. IEEESignal Processing Magazine 19:17-19, 2002
[20] Lan Jinhui, Zhou Zhaoying, Lan Tim. Neural Networks Recognition ofSeismic Signal for Vehicle Targets. Proceedings of the 3d World Congresson Intelligent Control and Automation, 1062-1064, 2002
[21] Douglas Lake. E?cient Maximum Likelihood Estimation for Multiple andCoupled Harmonics. Army Research Laboratory, ARL-TR-2014, 2002
[22] Mark C. Wellman, Nassy Srour, and David B. Hillis. Feature extraction andfusion of acoustic and seismic sensors for target identification. Proceedingsof SPIE, 3081:139-145, 2002
[23] Mark C. Wellman, Nassy Srour. Enhanced Target Identification UsingHigher Order Shape Statistics. Army Research Laboratory, ARL-TR-1732,1999
[24] James F. Scholl, Jonathan R. Agre, and Loren P. Clare. Wavelet PacketBased Target Classification Schemes. Proc 1999 Meeting of MSS SpecialtyGroup on Battlefield Acoustic and Seismic Sensing, 1999
[25] James F. Scholl, Jonathan R. Agre, Loren P. Clare and Marine C. Gill. Alow power Impulse Signal Classifier Using Harr Wavelet Transform. Proc.of SPIE, 3577:136-145, 1999
[26] Amir Z. Averbuch,Valery A. Zheludev and Inna Kozlov. Wavelet basedalgorithm for acoustic detection of moving ground and airborne targets.Proc. of SPIE, Vol. 4119, 2000
[27] Howard C. Choe. Wavelet-ground Vehicle recognition using acoustic signals.Proc. of SPIE, 2762:434-455, 1996
[28] M. R. Azimi-Sadjadi, Y. Jiang and S. Srinivasan. Acoustic Classificationof Battlefield Transient Events Using Wavelet Sub-band Features. Proc. ofSPIE, 6562(15):1-8, 2007
[29] Huadong Wu, Mel Siegel, and Pradeep Khosla. Vehicle Sound SignatureRecognition by Frequency Vector Principal Component Analysis. IEEETRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,48(5):1005-1009, 1999
[30] Graham L. Goodman. Detection and Classification for Unattended GroundSensors. Proceedings of Information Decision and Control 99, pp. 419-424,1999
[31] Lin Gu, Dong Jia, etc.. Lightweight Detection and Classification for WirelessSensor Networks in Realistic Environments. Proc. of the 3rd internationalconference on Embedded networked sensor system, 205-217, 2005.
[32] William J.J. Roberts, Howard W. Sabrin, Yariv Ephraim. Ground VehicleClassification using Hidden Markov Models. Meeting of the MSS SpecialtyGroup on Battlefield Acoustic and Seismic Sensing, Magnetic and ElectricField Sensors, 2001
[33] George Succi, Torstein K. Pedersen. Acoustic Target Tracking and TargetIdentification - Recent Result. Proc. of SPIE, 3713:10-21, 1999
[34] Mark C. Wellman and Nassy Srour. Feature extraction for a neural networkclassifier. Army Research Laboratories, ARL-TR-1166, 1997
[35] Jose E. Lopez, Hung Han Chen, Jennifer Saulnier. Target IdentificationUsing Wavelet-based Feature Extraction and Neural Network Classifiers.Report A983904, CYTEL System Inc., 1999
[36]胡寿伟,王维明。目标车辆震动场模拟系统的研究。2005全国仿真技术学术会议论文集,pp. 93-95,2005
[37]蓝金辉,马宝华,李科杰。典型机动目标地震动特性的试验研究。仪器仪表学报,21(5):169-170, 2000
[38]聂伟荣,朱继南,张元春。地震波传播理论在目标识别中的应用。探测与控制学报,21(4):50:58,1999
[39]朱继南,张元春,聂伟荣多层介质中的瑞雷波传播的数学模型。南京理工大学学报,24(6):510-514,2000
[40]杨成林等瑞雷波勘探地质出版社,北京,1993
[41] J. F. Scholl, L. P. Clare, and J. R. Agre. Seismic Attenuation Character-ization Using Tracked Vehicles. Proc. 1999 Meeting of the MSS SpecialtyGroup on Battlefield Acoustic and Seismic Sensing, Sept. 1999
[42] Jason Ding. Vehicle Detection by Sensor Networks. Master’s Thesis ofUNIVERSITY of CALIFORNIA at BERKELEY, 2003
[43] Gerard E. Sleefe, Mark D. Ladd, Daniel E. Gallegos, Carl W. Sicking, andIreena A. Erteza. A novel algorithm for real-time adaptive signal detectionand identification. Proc. of SPIE, 3374:495-504, 1998
[44] Jinhui Lan, Tian Lan, and Saeid Nahavandi. A Novel Application of aMicroaccelerometer for Target Classification. IEEE SENSORS JOURNAL,VOL. 4, NO. 4, AUGUST 2004
[45] Stefan Pittner, Sagar V. Kamarthi . Feature Extraction From Wavelet Co-e?cients for Pattern Recognition Tasks. IEEE Trans. Pattern Analysis andMachine Intelligence, Vol. 21, No. 1:83-88, January 1999
[46] N. Intrator, Q. Q. Huynh, Y. S. Wong, B. H. K. Lee. Wavelet Feature Ex-traction for Discrimination Tasks. Proceedings of The Canadian Workshopon Information Theory, pp. 83-86, 1997
[47] Gary G. Yen, and Kuo-Chung Lin. Wavelet Packet Feature Extraction forVibration Monitoring. IEEE TRANSACTIONS ON INDUSTRIAL ELEC-TRONICS, VOL. 47, NO. 3:650-667, JUNE 2000
[48]王炳锡,屈丹,彭煊。实用语音识别基础。国防工业出版社,北京,2005
[49] Y. Shoji, T. Takasuka, and H. Yasukawa. Personal Identification UsingFootstep Detection. Proc. of The IEEE 2004 International Symposium onIntelligent Signal Processing and Communication Systems (ISPACS2004),pp.43-47, 2004
[50] ROSS E. BLAND. ACOUSTIC AND SEISMIC SIGNAL PROCESSINGFOR FOOTSTEP DETECTION. Master’s thesis of Engineering in Electri-cal Engineering and Computer Science at the MASSACHUSETTS INSTI-TUTE OF TECHNOLOGY, 2006
[51] A. Pakhomov, A. Sicignano, M. Sandy, and T. Goldburt. Seismic FootstepSignal Characterization. Proc. of SPIE, Vol. 5071:297-305, 2003
[52] Alex Pakhomov and Tim Goldburt. Seismic Signals and Noise Assessmentfor Foot Step Detection Range Estimation in Di?erent Environments. Proc.of SPIE, Vol. 5417:87-98, 2004
[53] Giorgos P. Mazarakis, John N. Avaritsiotis. A Prototype Sensor Node forFootstep Detection. Proceeedings of the Second European Workshop onWireless Sensor Networks, pp. 415- 418, 2005
[54] SAEED V. VASEGHI. Advanced Digital Signal Processing and Noise Re-duction. John Wiley & Sons LTD., 2000
[55]张贤达。现代信号处理(第二版)。清华大学出版社,北京,2002
[56]彭玉华。小波变换与工程应用。科学出版社,北京,1999
[57] [美] Ingrid Daubechies著;李建平,杨万年译。小波十讲。国防工业出版社,北京,2004
[58] Jose E. Lopez, Juei Cheng Lo, Jennifer Saulnier. Vehicle Signal Enhance-ment Using Packet Wavelet Transform and Nonlinear Noise Processing Tech-niques. ADA409392,CYTEL System Inc., 1999
[59] David L. Donoho. De-nosing by soft-thresholding. IEEE Trans. InformationTheory, pp. 613-627, 1995
[60] Shie Qian. Introduction to Time-Frequency and Wavelet Transforms. Pear-son Education Asia Limited and China Machine Press, 2005
[61] Zujun Hou. Adaptive singular value decomposition in wavelet domain forimage denoising. Pattern Recognition, Vol. 36, pp. 1747-1763, 2003
[62] Mario E. Munich. BAYESIAN SUBSPACE METHODS FOR ACOUSTICSIGNATURE RECOGNITION OF VEHICLES. Proc. of the 12th EuropeanSignal Processing Conf. (EUSIPCO 2004), Sep. 6-10, 2004, Vienna, Austria
[63]边肇琪。模式识别(第二版)。清华大学出版社,2000
[64] http://www.ece.wisc.edu/ sensit
[65] Anil K. Jain, Fellow, IEEE, Robert P.W. Duin, and Jianchang Mao, SeniorMember, IEEE. Statistical Pattern Recognition: A Review. IEEE TRANS-ACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,VOL. 22, NO. 1, JANUARY 2000
[66] Burhan F. NECIOGL, Carol T. CHRISTOU, E. Bryan GEORGE, Garry M.JACYNA. Vehicle Acoustic Classification in Netted Sensor Systems UsingGaussian Mixture Models. Proceedings of SPIE, vol. 5809, pp. 409-419, 2005
[67] Simon Haykin著,叶世伟史忠植译。神经网络原理。机械工业出版社,北京,2004
[68] Jerzy Jackowski, Roman Wantoch-Rekowski. Classification of Wheeled Mil-itary Vehicle Using Neural Networks. Proceeding of ISCEng’05, 2005
[69] Mahmood R. Azimi-Sadjadi, De Yao, Qiang Huang, and Gerald J. Dobecki.Underwater Target Classification Using Wavelet Packets and Neural Net-works. IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 11,NO. 3, pp. 784-794, MAY 2000
[70] James A. Robertson, and John C. Mossing. Artificial Neural Networks forAcoustic Target Recognition. Proceeding of SPIE, VOL. 2492, pp. 939-949,1994
[71] Qi Huang, Tao Xing, and Hai Tao Liu. Vehicle Classification in WirelessSensor Networks Based on Rough Neural Network. ISNN 2006, LNCS 3973,pp. 58–65, 2006
[72] Hongwei Wu and Jerry M. Mendel. Multi-Category Classification of GroundVehicles Based on the Acoustic Data of Multiple Terrains Using Fuzzy LogicRule-Based Classifiers. Proceedings of SPIE, Vol. 5796, pp. 28-39, 2005
[73] Amir Averbuch, Eyal Hulata, Valery Zheludev, Inna Kozlov. A waveletpacket algorithm for classification and detection of moving vehicles. Multi-dimensional Systems and Signal Processing, Volume 12 , Issue 1, pp. 9-31,2001
[74] Andrew Webb. Statistical Pattern Recognition. John Wiley & Sons Ltd,The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England,2002
[75] T.K. Moon. The expectation-maximization algorithm. IEEE Signal Pro-cessing Magazine, Vol. 13(6):47-60, 1995
[76] YU HEN HU and JENQ-NENG HWANG. Handbook of Neural NetworkSignal Processing. CRC Press LLC, 2001.
[77] Marc Robinson, Mahmood R. Azimi-Sadjadi and Jaime Salazar. Multi-Aspect Target Discrimination Using Hidden Markov Models and NeuralNetworks. IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 16,NO. 2, pp. 447-459, MARCH 2005
[78]张学工。关于统计学习和支持向量机。自动化学报,26(1):32-42,2000
[79] Christopher J.C. Burges. A Tutorial on Support Vector Machines for PatternRecognition.自Data Mining and Knowledge Discovery, 2(2):121-167, 1998
[80] Damian Eads, Daniel Hill, Sean Davis, Simon Perkins, Junshui Ma, ReidPorter, and James Theiler. Genetic Algorithms and Support Vector Ma-chines for Time Series Classification. Proceedings of the SPIE, Volume 4787,pp. 74-85, 2002
[81] Vladimir N. Vapnik. An Overview of Statistical Learning Theory. IEEETRANSACTIONS ON NEURAL NETWORKS, VOL. 10, NO. 5, SEPTEM-BER 1999
[82] [英] Nello Crisianini, John Shawe-Taylor著李国正,王猛,曾华军译。支持向量机导论。电子工业出版社,北京,2000
[83] Vladimir N. Vapnik著,张学工译。统计学习理论的本质。清华大学出版社,北京,2000
[84]杜树新,吴铁军模式识别中的支持向量机方法。浙江大学学报,37(5):521-527,2003
[85] Andrea Boni, Fernando Pianegiani, and Dario Petri. Low-Power and Low-Cost Implementation of SVMs for Smart Sensors. IEEE TRANSACTIONSON INSTRUMENTATION AND MEASUREMENT, VOL. 56, NO. 1, pp.39-44, FEBRUARY 2007
[86] F. Pianegiani, A. Boni, M. Hu, D. Petri. Energy–Aware Signals Classifi-cation in Ad–hocWireless Sensor Networks. IMTC 2005 - Instrumentationand Measurement Technology Conference, pp. 1912-1916, 2005
[87] Davide Anguita, Andrea Boni, and Sandro Ridella. A Digital Architecturefor Support Vector Machines: Theory, Algorithm, and FPGA Implementa-tion. IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 14, NO.5, pp. 993-1009, SEPTEMBER 2003
[88] S. Sathiya Keerthi and Chih-Jen Lin. Asymptotic Behaviors of SupportVector Machines with Gaussian Kernel. Neural Computation, 15(7): 1667-1689, 2003
[89] Chih-Wei Hsu, Chih-Chung Chang, and Chih-Jen Lin. A Practical Guideto Support Vector Classification. http://www.csie.ntu.edu.tw/ cjlin/libsvm,2003
[90] Chih-Chung Chang and Chih-Jen Lin. LIBSVM: a library for support vectormachines. Software available at http://www.csie.ntu.edu.tw/ cjlin/libsvm,2001
[91] RICHARD R. BROOKS, PARAMESWARAN RAMANATHAN, AND AK-BAR M. SAYEED. Distributed Target Classification and Tracking in SensorNetworks. PROCEEDINGS OF THE IEEE, VOL. 91, NO. 8, AUGUST2003
[92] Jayesh H. Kotecha, Member, IEEE, Vinod Ramachandran, and Akbar M.Sayeed, Senior Member, IEEE Distributed Multitarget Classification inWireless Sensor Networks. IEEE JOURNAL ON SELECTED AREAS INCOMMUNICATIONS, VOL. 23, NO. 4, pp. 703-713, APRIL 2005
[93] LIU Chun-Ting, HUO Hong, FANG Tao, LI De-Ren, SHEN Xiao. Classifi-cation Fusion in Wireless Sensor Networks. ACTA AUTOMATICA SINICA,Vol. 32, No. 36, pp. 947-955, 2006
[94] STAFSUDD J Z, ASGARI S, ALI A M, CHEN C E, HUDSON R E,LORENZELLI F, YAO K, TACIROGLU E. Analysis, Implementation,and Application of Acoustic and Seismic Arrays. ACTA AUTOMATICASINICA, Vol. 32, No. 6, pp. 929-937, 2006
[95] Xiaoling Wang. High Accuracy Distributed Target Detection and Classifica-tion in Sensor Networks Based on Mobile Agent Framework. Ph.D. Thesisof the University of Tennessee in Knoxville, 2004
[96]刘琚,何振亚。盲源分离和盲反卷积。电子学报,30(4):570-576,2002
[97] M. R. Azimi-Sadjadi and S. Srinivasan. Blind Separation of Multiple VehicleSignatures In Frequency Domain. Proceedings of SPIE, Vol. 5796, pp. 93-104, 2005
[98] A. Hyva¨rinen, E. Oja. Independent component analysis: algorithms andapplications. Neural Networks 13, pp. 411–430, 2001
[99] Marco F. Duarte. A Fixed Point Implementation of ICA and its Applicationin Acoustic Denoising and Multiple Vehicle Classification. ELEC631 Mini-Project, http://www.ece.rice.edu/ duarte/, 2005
[100] M. Duarte and Y. H. Hu. Distance-based decision fusion in a distributedwireless sensor network. Telecommunication Systems 26, pp. 339–350, 2004.
[101] David L. Hall and James Llinas. Handbook of multisensor data fusion CRCPress LLC,2000
[102]蓝金辉,马宝华,蓝天,周兆英。D-S证据理论数据融合方法在目标识别中的应用。清华大学学报(自然科学版),41(2):53-55,2001
[103] A. Hyvrinen, J. Karhunen and E. Oja. Independent Component Analysis.John Wiley and Sons, New York, 2001
[104]杨福生,洪波著。独立分量分析的原理和应用。清华大学出版社,北京,2006
[105]陈功,张雄伟,邓玉良。基于ICA-GMM的混叠振动目标识别。噪声与振动控制,2006年第5期,pp. 5-8,2006
[106] Tommy Yu. Target Identification Processor For Wireless Sensor Network.Ph.D. Thesis in Electrical Engineering of UNIVERSITY OF CALIFORNIALos Angeles, 1999
[107] Gert Cauwenberghs, Andreas Andreou, Jim West, Milutin Stanacevic, Ab-dullah Celik, Pedro Julian, Thiago Teixeira. A Miniature, Low-Power, In-telligent Sensor Node for Persistent Acoustic Surveillance. ....
[108] A. Hyva¨rinen. Fast and Robust Fixed-Point Algorithms for Indepen-dent Component Analysis. IEEE TRANSACTIONS ON NEURAL NET-WORKS, VOL. 10, NO. 3, pp. 626-634, MAY 1999
[109] M. R. Azimi-Sadjadi and K. Khorasani. Detection, Tracking and Classifi-cation of Multiple Targets using Advanced Beamforming and ClassificationMethod. SBIR report, 2002