独立分量分析及其在多道信号处理中的应用研究
|
摘要
近年来，信号处理的理论与方法获得了迅速发展。事实证明，信号处理是推动众学科发展的一个重要基石。独立分量分析技术(ICA)是信号处理领域发展较晚的一种理论与方法，已迅速成为该领域内重要的组成部分，且其发展逐渐趋向成熟化与系统化。
本文系统地介绍了独立分量分析的理论与方法，同时将其应用到有价值的实际诱发脑电信号和数字图像信号分离中。
本文内容包括：脑科学及脑电信号分析方法现状，独立分量分析理论，诱发电位提取技术研究进展，数字图像水印分类与处理算法以及独立分量分析在脑电(EEG/VEP)及数字图像水印处理中的试验，最后总结了本论文的工作及需要解决的一些问题与下一步的发展方向。
独立分量分析技术可以很好地分离出EEG数据中的肌电与眼动等干扰；实现了VEP的增强与提取；很好地分离数字图像信号中的水印与密钥，即使在很强的干扰情况下，仍具有很好的性能指标。这些结果充分展示出ICA技术良好的信号处理能力和应用价值。
Recently, theories and methods of signal processing are obtained to develope quickly. In fact, signal processing is a important foundation stone for all kinds of subjects' developing. Independent component analysis (ICA) technology developing is a later theory or method in fields of signal processing, but it was important that rapidly become a part of constitution of signal processing fields, and its developing tends gradually to maturity and systematization.
This paper introduced theory and method of independent component analysis by the numbers. At the same time , it was applied to process Evoked Electroencephalography (EEG) signal which it has very important value, and to process digital image signal.
This paper's contents include: brain science and analytical actuality of EEG signal, theory of ICA, research headway of evoked potential (EP) extraction technique, classes and processing algorithms of digital image watermarking, and ICA technology's application (in EEG/VEP and digital image watermarking processing), finally, summarized the work of this paper and some problems on the ICA technique which to be solved were presented and the direction of the development pointed out.
The ICA technique separated successfully electromyographic (EMG) and electro-Oculographic (EOG) artifacts from EEG data ; accomplished VEP strengthening and extraction; separated watermarking and key from mixing digital image signals, even though under strong noise, the results still have very good performance index. All above showed nicer signal processing ability and factual value of ICA technique.
本文系统地介绍了独立分量分析的理论与方法，同时将其应用到有价值的实际诱发脑电信号和数字图像信号分离中。
本文内容包括：脑科学及脑电信号分析方法现状，独立分量分析理论，诱发电位提取技术研究进展，数字图像水印分类与处理算法以及独立分量分析在脑电(EEG/VEP)及数字图像水印处理中的试验，最后总结了本论文的工作及需要解决的一些问题与下一步的发展方向。
独立分量分析技术可以很好地分离出EEG数据中的肌电与眼动等干扰；实现了VEP的增强与提取；很好地分离数字图像信号中的水印与密钥，即使在很强的干扰情况下，仍具有很好的性能指标。这些结果充分展示出ICA技术良好的信号处理能力和应用价值。
Recently, theories and methods of signal processing are obtained to develope quickly. In fact, signal processing is a important foundation stone for all kinds of subjects' developing. Independent component analysis (ICA) technology developing is a later theory or method in fields of signal processing, but it was important that rapidly become a part of constitution of signal processing fields, and its developing tends gradually to maturity and systematization.
This paper introduced theory and method of independent component analysis by the numbers. At the same time , it was applied to process Evoked Electroencephalography (EEG) signal which it has very important value, and to process digital image signal.
This paper's contents include: brain science and analytical actuality of EEG signal, theory of ICA, research headway of evoked potential (EP) extraction technique, classes and processing algorithms of digital image watermarking, and ICA technology's application (in EEG/VEP and digital image watermarking processing), finally, summarized the work of this paper and some problems on the ICA technique which to be solved were presented and the direction of the development pointed out.
The ICA technique separated successfully electromyographic (EMG) and electro-Oculographic (EOG) artifacts from EEG data ; accomplished VEP strengthening and extraction; separated watermarking and key from mixing digital image signals, even though under strong noise, the results still have very good performance index. All above showed nicer signal processing ability and factual value of ICA technique.
引文
[1] 张贤达 《现代信号处理》 清华人学出版社 1994年4月
[2] 杨福生 高上凯 《生物医学信号处理》 高等教育出版社 1989年5月
[3] 沈民奋 孙丽莎 《现代随机信号与系统分析》 科学出版社 1998年
[4] 勃雷兹尔 《神经系统的电活动》 科学出版社 1984年
[5] 侯宗濂等《中国医学百科全书·生理学卷》 上海科学技术出版社 1985年
[6] 诺贝克等《神经解剖学基础》 科学出版社 1983年
[7] 张香桐 脑研究的新近展及发展趋势 生理科学进展 No．1 1979年
[8] 张香桐 神经科学在中国的发展 生理科学进展Vol．14 No．2 1983年
[9] 姚志彬 陈以慈 脑研究前沿 广东科技出版社 1995．5
[10] 吕国蔚 医学神经生物学 高等教育出版社 2000．2
[11] 杨福生 高上凯 《生物医学信号处理》 高等教育出版社 1989年5月
[12] 杨雄里 脑科学的现代进展 上海科技教育出版社 1998．12
[13] 陈兴时，张明岛．脑诱发电位学．上海：上海科技教育出版社，1997．
[14] 朱常芳，胡广书．诱发电位快速提取算法的新进展．国外医学生物医学工程分册，2000，23(4)，211—216．
[15] 洪波，唐庆玉，杨福生等．ICA在视觉诱发电位的少次提取与波形分析中的应用中国生物医学工程学报 2000．9，19(3)，334—341．
[16] 杨福生 洪波 唐庆玉．独立分量分析及其在生物医学工程中的应用．国外医学生物医学工程分册 2000，23(3)，129—134．
[17] 陈兴时 张明岛 脑电发展史实 脑电图学与精神疾病杂志 Vol．5 No．1 1989年
[18] 李鲁平，吴延军，程敬之．诱发脑电动态提取方法研究[J]．国外医学生物医学工程册，1995，18：195-201．
[19] 关力 诱发电何检测技术的进展[J]．国外医学生物医学工程分册，1 995，18：125-129
[20] 刘建成 蔡湛宇 脑电信号(EEG)分析方法的现状与发展 中国医学物理学杂志 第卷第4期1998年10月 CHINESE JMED PHYS Vol．1 5No．4Oct．1 998：252～255
[21] 吴祈耀 现代信号处理在生物医学工程在的应用 北京理工大学学报
[22] 吴祈耀 吴祈宗 脑电信号的现代谱分析技术 北京理工大学学报 V．15 No．2 1995
[23] 李勇 脑电信号现代分析方法的发展 国外医学生物医学工程分册 V．19 No．4 1996
[24] Comon P. Independent component analysis. A new concept?[J]. Signal Processing,1994,36,287-314.
[25] Hyvarinen A. Fast and robust fixed-pointed algorithms for independent component analysis[J].IEEE Trans Neural Networks, 1999, 10, 626-634.
[26] Bell AJ, Sejnowski TJ. An information-maximization approach to blind separation and blind deconvolution[J]. Neural Computation, 1995,7,1129-1159.
[27] Aapo Hyvarinen and Erkki Oja. Independent Component Analysis: Algorithms and Application. Neural Networks, 2000(13), 411-430.
[28] The Fast ICA MATLAB package. Available at http://www.cis.hut.fi/projects/ica/fastica/.
[29] E Hartung and M. Kutter, Multimedia Watermarking Techniques. Proceedings of the IEEE, vol.87, no.7, pp.1079-1107, 1999.
[30] K. Stefan and F. A. P. Petitcolas, Eds., Information Hiding Techniques for Steganography and
Sigital Watermarking. Artech House, Boston, 2000.
[31] Aapo Hyvarinen and Erkki Oja. Independent Component Analysis: A Tutorial. April 1999.
[32] J.-F. Cardoso. Blind Signal Separation: Statistical Principles, Proceedings of the IEEE, vol.9, no. 10, pp.2009-2026, 1998.
[33] D.Yu, F.Sattar, and K.-K.Ma. Watermark Detection and Extraction Using a Independent Component Analysis Method. EURASIP Journal on Applied Signal Proccssing 2002: 1, 92-104.
[34] A. Hyvarinen and E.Oja. A Fast-fixed Point Algorithm for Independent Component Analysis. Neural Computation, 9(7): 1483-1492, 1997.
[35] F. A. P. Petitcolas, R. J. Anderson. Evaluation of Copyright Marking Systems. in Proc, of IEEE Multimedia Systems'99, vol. 1, pp.574-579, Florence, Italy, June 1999.
[36] A.Cichocki and A.K.Barros, Robust batch algorithm for sequential blind extraction of noisy biomedical signals. In 5th International Symposium on Signal Processing and its Applications (ISSPA'99),vol. 1,pp.363-366,August 1999.
[37] R.L.Kashyap and K.-B.Eom, Robust image models and their applications. In Advances in Electronics and Electron Physics,P.W.Hawkes,Ed.,vo1.70,pp.79-157,Academic Press, 1988.
[38] W.A.Fuller, Introduction to Statistical Times Series. Wiley, New York, 2nd edition, 1996.
[39] J.ECardoso and Souloumiac, Blind beamforming for non-Gaussian signals, IEE Proceedings-F, Vol. 140: 362-370, No.6,December 1993
[40] Jean-Francois Cardoso and Beate Hvam Laheld, Equivariant adaptive Source Separation, IEEE Trans. On Signal Processing. Vol.44: 3017-3029, No.12,December, 1996
[41] Cardoso JE Higher order contrasts for independent component analysis[J]. Neural Computation,1999,11 (1):157-193
[42] Girolamic,M.(1997a).An alternative Perspective on adaptive independent component analysis algorithms.Technical report,issn 1461-6122,Department of computing and Information System.Paisley University ,Scotland
[43] Herault,J.and Jutten,C(1986).Space or time adaptive signal processing by network models In in Denker J.S.(ed),editor,Neural networks for Computing:AIP conference proceedings 151,New York.American Institute for Physis
[44] Metin Akay. Wavelet applications in medicine. IEEE Spectrum. MAY 1997
[45] Makeig S, Bell A J, Jung T-P, and Sejnowski TJ, Independent component analysis of Electroencephalographic data. Advances in NIPS 8, 145-151,1996.
[46] Ghahremani D, Makeig S, Jung T-P, Bell AJ, Sejnowski TJ, Independent component analysis of simulated EEG using a three-shell spherical head model, Tech Rep. INC-9601, Institute for Neural Computation, University of California, San Diego, May 1996.
[47] Makeig S, Jung T-P, Ghahremani D and Sejnowski TJ, Independent component analysis of Simulated ERP Data, Tech Rep. INC-9606, Institute for Neural Computation, University of California, San Diego, 1996.
[48] Jung T-P, Makeig S, Bell AJ, and Sejnowski TJ, "Independent Component Analysis of Electroencephalographic and Event-related Data." In: (Poon P. and Brugge J, ed) Auditory Processing and Neural Modeling, 189-197, 1998.
[49] Makeig S, Jung T-P, Bell AJ, and Sejnowski TJ, Blind Separation of Auditory Event-related Brain Responses into Independent Components Proc. of National Academy of Sciences, 94:10979-84, 1997.
[50] Lee TW, et al. Independent component analysis using an extended infomax algorithm for
mixed Subgaussian and Supergaussian,Neural Compution, 1999,11(2):409～433
[51] Amri S,et al. Adaptive blind signal processing, Neural Network approaches,Proc IEEE 1998,86(10):2026～2049
[52] Cichocki A,et al. Robust learning algorithm for blind separation of signals,Electronics Letters, 1994,30(17): 1386～1387
[53] Cardoso JF. Blind signal separation:statistical principles, Proc IEEE, 1998,86(10):2009～2025
[54] J.-E Cardoso. Infomax and maximum likelihood for source separation. IEEE Letters on Signal Processing, 4:112-114, 1997.
[55] S.-I. Amari, A. Cichocki, and H.H. Yang. A new leaming algorithm for blind source separation. In Advances in Neural Information Processing Systems 8, pages 757-763. MIT Press, Cambridge, MA, 1996.
[56] S.-I. Amari. Neural learning in structured parameter spaces-natural riemannian gradient. In Advances in Neural Information Processing 9, pages 127-133.MITPress,Cambridge, MA, 1997.
[57] S.-I. Amari and A. Cichocki. Adaptive blind signal processing-neural network approaches. Proceedings of the IEEE,9,1998
[58] J.J. Atick. Entropy minimization: A design principle for sensory perception? International Journal of Neural System,3: 81-90, 1992. Supp. 1992.
[59] H.B.Barlow. Possible principle underlying the transformation of sensory message. In W.A. Rosenblith,editor, Sensory Communication,pages 217-234.MIT Press, 1961.
[60] H.B.Barlow. Single units and sensation: A neuron doctrine for perceptual psychology? Perception, 1: 371-394,1972.
[61] H.B.Barlow. Unsupervised learning. Neural Compution, 1: 295-311, 1989
[62] H.B.Barlow. What is the computational goal of the necortex? In C. Koch and J.L.Davis, editors,Large-scale neuronal theories of the brain. MIT Press,Cambridge, MA, 1994
[63] H.B.Barlow, T.P.Kaushal, and G.J.Mitchison. Finding minimum entropy codes. Neural Computaion, 1: 412-423,1989.
[64] M.S.Bartlett. A note on the multiplying factors for various chi-square approximations. J.Roy.Stat.Soc., 16ser B.. 296-298,1989
[65] A.J.Bell and T.J.Sejnowski. Learning higher-order structure of a natural sound. Network,7:261-266,1996.
[66] J.F.Cardoso. Source separation using higher order moments. In Proc. ICASSP'89, pages 2109-2112, 1989
[67] J.F.Cardoso. Eigen-structure of the fourth-order cumulant tensor with application to the blind separation problem. In Proc. ICASSP'90, pages 2655-2658, Albuquerque, NM, USA, 1990.
[68] J.F.Cardoso. Super-symmetric decomposition of the fourth-order cumulant tensor. Blind identification of more sources than sensors. In Proc. ICASSP'91, pages 3109-3112, 1991
[69] J.F.Cardoso. Iterative techniques for blind sources separation using only fourth-order cumulants. In Proc. EUSIPCO, pages 739-742, Brussels, Belgium, 1992.
[70] J.-F. Cardoso. Infomax and maximum likelihood for source separation. IEEE Letters on Signal Processing, 4:112-114, 1997.
[71] J.-F. Cardoso. Entropic contrasts for source separation. In. Haykin, editor, Adaptive Unsupervised Learning. 1999.
[72] J.-F. Cardoso and P.Comon. Independent component analysis, a survey of some algebraic
methods. In Proc. lCASSP'96, vol. 2, P. 93-96, 1996.
[73] J.-F. Cardoso and A.Souloumiac. Blind beamforming for non Gaussian signals. IEEE Trans. on Signal Processing, 44(12): 3017-3030,1996.
[74] Cichocki A., Unbehauen R. Neural Network for Signal Processing and Optimization. Wiley,1994.
[75] Cichocki A., Unbehauen R. Robust neural network with on-line learning for blind identification and BBS. IEEE Trans. on Circuits and System, 43(11): 894-906, 1996.
[76] Ciehoeki A., Unbehauen R. and E.Rummert. A new on-line adaptive algorithm for BBS signals. In Proc. Int. System on Artificial Neural Networks ISANN-94, P.406-411, Tainan, Taiwan, 1994.
[77] P.Comon. Blind identification in presence of noise. In Signal Processing Ⅵ: Theories and Application (Proc.EUSIPCO),P.835-838. Elsevier, 1992.
[78] G. Deco and D.Obradovic. Linear redundancy reduction learning. Neural Network, 8(5): 751-755, 1995.
[79] N.Delfosse and P.Loubaton. Adaptive blind separation of independent sources: a deflation approach. Signal Processing, 45: 59-83, 1995.
[80] D.Donoho. On minimum entropy deconvolution. In Applied Time Series Analysis Ⅱ, P.565-608. Academic Press, 1981.
[81] J.H. Friedman. Exploratory projection pursuit. J.of the American Statistical Association, 82(397):249-266, 1987.
[82] C. Fyfe and R.Baddeley. Non-liner data structure extraction using simple Hebbian networks. Biological Cybernetics, 72: 533-541, 1995.
[83] F.R. Hampel and E.M.Ronchetti, Robust Statistics. Wiley, 1986.
[84] H.H.Harman. Modern Factor Analysis. University of Chicago Press, 2nd edition, 1967.
[85] S.Haykin. editor. Blind Deconvolution. Prentice-Hall, 1994.
[86] S.Haykin. Adaptive Filter Theory. Prentice-Hall International, 3rd edition, 1996.
[87] R.Hecht-Nielsen. Replicator neural networks for universal optimal source coding. Science, 269: 1860-1863, 1995.
[88] P.J.Huber. Robust Statistics. Wiley, 1981.
[99] P.J.Huber. Projection pursuit. The Annals of Statistics, 13(2): 435-475, 1985.
[90] J.Hurri, A. Hyvrinen, and E. ja. Wavelets and natural image statistics. In Proc. Scandinavian Conf. On Image Analysis'97, Lappenranta, Finland, 1997.
[91] A. Hyvrinen. A family of fixed-point algorithm for independent component analysis. In Proc. IEEE ICASSP'97, P.3917-3920, Munich Germany, 1997.
[92] A. Hyvrinen. One-unit contrast functions for ICA: A statistical analysis. In Neural Network for Signal Processing Ⅶ, P.388-397, Amelia Island, Florida, 1997.
[92] A. Hyvrinen. New approximations of differential entropy for independent component analysis and projection pursuit. In Advances in Neural Information Processing Systems, volume 10, pages 273-279. MIT Press, 1998.
[93] A. Hyvrinen. The fixed-point algorithm and maximum likelihood estimation for independent component analysis. Neural Processing Letters, 10(1): 1-5, 1999.
[94] A. Hyvrinen and R.Cristescu. A fast algorithm for estimating overcomplete ICA bases for image windows. In Proc. Int. Joint Conf. on Neural Networks, Washington ,D.C., 1999.
[95] A. Hyvrinen and E. Oja. Simple neuron models for independent component analysis. Int.
Journal of Neural Systems, 7(6):671-687, 1996.
[96] A. Hyvrinen and E. Oja. Independent component analysis by general nonlinear Hebbian-like learning rules. Signal Processing, 64(3):301-313, 1998.
[97] I.T. Jolliffe. Principal Component Analysis. Springer-Verlag, 1986.
[98] M.C. Jones and Sibon. What is prejection pursuit? J.of the Royal Statistics Society, ser. A, 150: 1-36, 1987.
[99] C. Jutten and J. Herault. Blind separation of sources, part Ⅰ: An adaptive algorithm based on neuromimetic architecture. Signal Processing, 24:1-10, 1991.
[100] J.Karhunen, A. Hyvrinen and R.Vigario. Applications of neural blind separation to signal and image processing. In Proc.IEEE Int. Conf. on ICASSP'97, P.131-134, Munich,Germany, 1997
[101] J.Karhunen and J.Joutsensalo. Respresentatin and separation of signals using nonlinear PCA type learning. Neural Networks, 7(1): 113-127, 1994.
[102] J.Karhunen and J.Joutsensalo. Generalizations of principal component analysis, optimization problems, and neural networks. Neural Networks, 8(4): 549-562, 1995.
[103] J.Karhunen, E.Oja, L. Wang and J.Joutsensalo. A class of neural networks for IEEE Trans. on Neural Networks, 8(3): 486-504, 1997.
[104] Karhunen, J., and Pajunen, E Blind Source Separation Using Least-Squares Type Adaptive Algorithms. In Proc. IEEE 1997 Int. Conf on Acoustics, Speech. and Signal Processing(ICASSP'97), April 21-24, 1997, Munich, Germany, pp. 3361-3364.
[105] M.Kendall. Multivariate Analysis. Charles Griffin&Co., 1975.
[106] K. Kiviluoto and E. Oja. Independent component analysis for parallel financial time series. In Proc. ICONIP'98, volume 2, pages 895-898, Tokyo, Japan, 1998.
[107] McKeown, M., Makeig, S., Brown, G., Jung, T-E, Kindermann, S., Bell, Iragui, V. and Sejnowski, T.J., Blind separation of functional magnetic resonance imaging (fMRI) data, Human Brain Mapping,. 6:160,18, 1998a.
[108] E. Moulines, J.-E Cardoso, and E. Gassiat. Maximum likelihood for blind separation and deconvolution of noisy signals using mixture models. In Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP'97), pages 3617-3620, Munich, Germany, 1997 o
[109] E. Oja. The nonlinear PCA learning rule in independent component analysis. Neurocomputing, 17(1):25-46, 1997.
[110] B. A. Olshausen and D. J. Field. Sparse coding with an overcomplete basis set: A strategy employed by V1? Vision Research, 37:3311-3325, 1997.
[111] Pham D. T., Garat Ph. et Jutten Ch. (1992) Separation of mixture of independent sources through a maximum likelihood approach. Dans Signal Processing Ⅵ, Proceeding EUSIPCO'92, 771-774, Eds. J. Vandewalle, R. Boite, M. Moonen and A. Oosterlinck.
[112] T. Ristaniemi, K. Raju, J. Karhunen, E. Oja, "Jammer Cancellation in DS-CDMA Array Systems: Pre and Post Switching of ICA and RAKE", In Proc. of the IEEE International Symposium on Neural Networks for Signal Processing (NNSP), Martingy, Switzerland. September 2-4, 2002, pp 495-504
[113] R.N. Vigario," Extraction of ocular artifacts from EEG using independent component analysis", Electroencephy and clinical Neurophysiology, No. 103, pp.395404,1997.
[2] 杨福生 高上凯 《生物医学信号处理》 高等教育出版社 1989年5月
[3] 沈民奋 孙丽莎 《现代随机信号与系统分析》 科学出版社 1998年
[4] 勃雷兹尔 《神经系统的电活动》 科学出版社 1984年
[5] 侯宗濂等《中国医学百科全书·生理学卷》 上海科学技术出版社 1985年
[6] 诺贝克等《神经解剖学基础》 科学出版社 1983年
[7] 张香桐 脑研究的新近展及发展趋势 生理科学进展 No．1 1979年
[8] 张香桐 神经科学在中国的发展 生理科学进展Vol．14 No．2 1983年
[9] 姚志彬 陈以慈 脑研究前沿 广东科技出版社 1995．5
[10] 吕国蔚 医学神经生物学 高等教育出版社 2000．2
[11] 杨福生 高上凯 《生物医学信号处理》 高等教育出版社 1989年5月
[12] 杨雄里 脑科学的现代进展 上海科技教育出版社 1998．12
[13] 陈兴时，张明岛．脑诱发电位学．上海：上海科技教育出版社，1997．
[14] 朱常芳，胡广书．诱发电位快速提取算法的新进展．国外医学生物医学工程分册，2000，23(4)，211—216．
[15] 洪波，唐庆玉，杨福生等．ICA在视觉诱发电位的少次提取与波形分析中的应用中国生物医学工程学报 2000．9，19(3)，334—341．
[16] 杨福生 洪波 唐庆玉．独立分量分析及其在生物医学工程中的应用．国外医学生物医学工程分册 2000，23(3)，129—134．
[17] 陈兴时 张明岛 脑电发展史实 脑电图学与精神疾病杂志 Vol．5 No．1 1989年
[18] 李鲁平，吴延军，程敬之．诱发脑电动态提取方法研究[J]．国外医学生物医学工程册，1995，18：195-201．
[19] 关力 诱发电何检测技术的进展[J]．国外医学生物医学工程分册，1 995，18：125-129
[20] 刘建成 蔡湛宇 脑电信号(EEG)分析方法的现状与发展 中国医学物理学杂志 第卷第4期1998年10月 CHINESE JMED PHYS Vol．1 5No．4Oct．1 998：252～255
[21] 吴祈耀 现代信号处理在生物医学工程在的应用 北京理工大学学报
[22] 吴祈耀 吴祈宗 脑电信号的现代谱分析技术 北京理工大学学报 V．15 No．2 1995
[23] 李勇 脑电信号现代分析方法的发展 国外医学生物医学工程分册 V．19 No．4 1996
[24] Comon P. Independent component analysis. A new concept?[J]. Signal Processing,1994,36,287-314.
[25] Hyvarinen A. Fast and robust fixed-pointed algorithms for independent component analysis[J].IEEE Trans Neural Networks, 1999, 10, 626-634.
[26] Bell AJ, Sejnowski TJ. An information-maximization approach to blind separation and blind deconvolution[J]. Neural Computation, 1995,7,1129-1159.
[27] Aapo Hyvarinen and Erkki Oja. Independent Component Analysis: Algorithms and Application. Neural Networks, 2000(13), 411-430.
[28] The Fast ICA MATLAB package. Available at http://www.cis.hut.fi/projects/ica/fastica/.
[29] E Hartung and M. Kutter, Multimedia Watermarking Techniques. Proceedings of the IEEE, vol.87, no.7, pp.1079-1107, 1999.
[30] K. Stefan and F. A. P. Petitcolas, Eds., Information Hiding Techniques for Steganography and
Sigital Watermarking. Artech House, Boston, 2000.
[31] Aapo Hyvarinen and Erkki Oja. Independent Component Analysis: A Tutorial. April 1999.
[32] J.-F. Cardoso. Blind Signal Separation: Statistical Principles, Proceedings of the IEEE, vol.9, no. 10, pp.2009-2026, 1998.
[33] D.Yu, F.Sattar, and K.-K.Ma. Watermark Detection and Extraction Using a Independent Component Analysis Method. EURASIP Journal on Applied Signal Proccssing 2002: 1, 92-104.
[34] A. Hyvarinen and E.Oja. A Fast-fixed Point Algorithm for Independent Component Analysis. Neural Computation, 9(7): 1483-1492, 1997.
[35] F. A. P. Petitcolas, R. J. Anderson. Evaluation of Copyright Marking Systems. in Proc, of IEEE Multimedia Systems'99, vol. 1, pp.574-579, Florence, Italy, June 1999.
[36] A.Cichocki and A.K.Barros, Robust batch algorithm for sequential blind extraction of noisy biomedical signals. In 5th International Symposium on Signal Processing and its Applications (ISSPA'99),vol. 1,pp.363-366,August 1999.
[37] R.L.Kashyap and K.-B.Eom, Robust image models and their applications. In Advances in Electronics and Electron Physics,P.W.Hawkes,Ed.,vo1.70,pp.79-157,Academic Press, 1988.
[38] W.A.Fuller, Introduction to Statistical Times Series. Wiley, New York, 2nd edition, 1996.
[39] J.ECardoso and Souloumiac, Blind beamforming for non-Gaussian signals, IEE Proceedings-F, Vol. 140: 362-370, No.6,December 1993
[40] Jean-Francois Cardoso and Beate Hvam Laheld, Equivariant adaptive Source Separation, IEEE Trans. On Signal Processing. Vol.44: 3017-3029, No.12,December, 1996
[41] Cardoso JE Higher order contrasts for independent component analysis[J]. Neural Computation,1999,11 (1):157-193
[42] Girolamic,M.(1997a).An alternative Perspective on adaptive independent component analysis algorithms.Technical report,issn 1461-6122,Department of computing and Information System.Paisley University ,Scotland
[43] Herault,J.and Jutten,C(1986).Space or time adaptive signal processing by network models In in Denker J.S.(ed),editor,Neural networks for Computing:AIP conference proceedings 151,New York.American Institute for Physis
[44] Metin Akay. Wavelet applications in medicine. IEEE Spectrum. MAY 1997
[45] Makeig S, Bell A J, Jung T-P, and Sejnowski TJ, Independent component analysis of Electroencephalographic data. Advances in NIPS 8, 145-151,1996.
[46] Ghahremani D, Makeig S, Jung T-P, Bell AJ, Sejnowski TJ, Independent component analysis of simulated EEG using a three-shell spherical head model, Tech Rep. INC-9601, Institute for Neural Computation, University of California, San Diego, May 1996.
[47] Makeig S, Jung T-P, Ghahremani D and Sejnowski TJ, Independent component analysis of Simulated ERP Data, Tech Rep. INC-9606, Institute for Neural Computation, University of California, San Diego, 1996.
[48] Jung T-P, Makeig S, Bell AJ, and Sejnowski TJ, "Independent Component Analysis of Electroencephalographic and Event-related Data." In: (Poon P. and Brugge J, ed) Auditory Processing and Neural Modeling, 189-197, 1998.
[49] Makeig S, Jung T-P, Bell AJ, and Sejnowski TJ, Blind Separation of Auditory Event-related Brain Responses into Independent Components Proc. of National Academy of Sciences, 94:10979-84, 1997.
[50] Lee TW, et al. Independent component analysis using an extended infomax algorithm for
mixed Subgaussian and Supergaussian,Neural Compution, 1999,11(2):409～433
[51] Amri S,et al. Adaptive blind signal processing, Neural Network approaches,Proc IEEE 1998,86(10):2026～2049
[52] Cichocki A,et al. Robust learning algorithm for blind separation of signals,Electronics Letters, 1994,30(17): 1386～1387
[53] Cardoso JF. Blind signal separation:statistical principles, Proc IEEE, 1998,86(10):2009～2025
[54] J.-E Cardoso. Infomax and maximum likelihood for source separation. IEEE Letters on Signal Processing, 4:112-114, 1997.
[55] S.-I. Amari, A. Cichocki, and H.H. Yang. A new leaming algorithm for blind source separation. In Advances in Neural Information Processing Systems 8, pages 757-763. MIT Press, Cambridge, MA, 1996.
[56] S.-I. Amari. Neural learning in structured parameter spaces-natural riemannian gradient. In Advances in Neural Information Processing 9, pages 127-133.MITPress,Cambridge, MA, 1997.
[57] S.-I. Amari and A. Cichocki. Adaptive blind signal processing-neural network approaches. Proceedings of the IEEE,9,1998
[58] J.J. Atick. Entropy minimization: A design principle for sensory perception? International Journal of Neural System,3: 81-90, 1992. Supp. 1992.
[59] H.B.Barlow. Possible principle underlying the transformation of sensory message. In W.A. Rosenblith,editor, Sensory Communication,pages 217-234.MIT Press, 1961.
[60] H.B.Barlow. Single units and sensation: A neuron doctrine for perceptual psychology? Perception, 1: 371-394,1972.
[61] H.B.Barlow. Unsupervised learning. Neural Compution, 1: 295-311, 1989
[62] H.B.Barlow. What is the computational goal of the necortex? In C. Koch and J.L.Davis, editors,Large-scale neuronal theories of the brain. MIT Press,Cambridge, MA, 1994
[63] H.B.Barlow, T.P.Kaushal, and G.J.Mitchison. Finding minimum entropy codes. Neural Computaion, 1: 412-423,1989.
[64] M.S.Bartlett. A note on the multiplying factors for various chi-square approximations. J.Roy.Stat.Soc., 16ser B.. 296-298,1989
[65] A.J.Bell and T.J.Sejnowski. Learning higher-order structure of a natural sound. Network,7:261-266,1996.
[66] J.F.Cardoso. Source separation using higher order moments. In Proc. ICASSP'89, pages 2109-2112, 1989
[67] J.F.Cardoso. Eigen-structure of the fourth-order cumulant tensor with application to the blind separation problem. In Proc. ICASSP'90, pages 2655-2658, Albuquerque, NM, USA, 1990.
[68] J.F.Cardoso. Super-symmetric decomposition of the fourth-order cumulant tensor. Blind identification of more sources than sensors. In Proc. ICASSP'91, pages 3109-3112, 1991
[69] J.F.Cardoso. Iterative techniques for blind sources separation using only fourth-order cumulants. In Proc. EUSIPCO, pages 739-742, Brussels, Belgium, 1992.
[70] J.-F. Cardoso. Infomax and maximum likelihood for source separation. IEEE Letters on Signal Processing, 4:112-114, 1997.
[71] J.-F. Cardoso. Entropic contrasts for source separation. In. Haykin, editor, Adaptive Unsupervised Learning. 1999.
[72] J.-F. Cardoso and P.Comon. Independent component analysis, a survey of some algebraic
methods. In Proc. lCASSP'96, vol. 2, P. 93-96, 1996.
[73] J.-F. Cardoso and A.Souloumiac. Blind beamforming for non Gaussian signals. IEEE Trans. on Signal Processing, 44(12): 3017-3030,1996.
[74] Cichocki A., Unbehauen R. Neural Network for Signal Processing and Optimization. Wiley,1994.
[75] Cichocki A., Unbehauen R. Robust neural network with on-line learning for blind identification and BBS. IEEE Trans. on Circuits and System, 43(11): 894-906, 1996.
[76] Ciehoeki A., Unbehauen R. and E.Rummert. A new on-line adaptive algorithm for BBS signals. In Proc. Int. System on Artificial Neural Networks ISANN-94, P.406-411, Tainan, Taiwan, 1994.
[77] P.Comon. Blind identification in presence of noise. In Signal Processing Ⅵ: Theories and Application (Proc.EUSIPCO),P.835-838. Elsevier, 1992.
[78] G. Deco and D.Obradovic. Linear redundancy reduction learning. Neural Network, 8(5): 751-755, 1995.
[79] N.Delfosse and P.Loubaton. Adaptive blind separation of independent sources: a deflation approach. Signal Processing, 45: 59-83, 1995.
[80] D.Donoho. On minimum entropy deconvolution. In Applied Time Series Analysis Ⅱ, P.565-608. Academic Press, 1981.
[81] J.H. Friedman. Exploratory projection pursuit. J.of the American Statistical Association, 82(397):249-266, 1987.
[82] C. Fyfe and R.Baddeley. Non-liner data structure extraction using simple Hebbian networks. Biological Cybernetics, 72: 533-541, 1995.
[83] F.R. Hampel and E.M.Ronchetti, Robust Statistics. Wiley, 1986.
[84] H.H.Harman. Modern Factor Analysis. University of Chicago Press, 2nd edition, 1967.
[85] S.Haykin. editor. Blind Deconvolution. Prentice-Hall, 1994.
[86] S.Haykin. Adaptive Filter Theory. Prentice-Hall International, 3rd edition, 1996.
[87] R.Hecht-Nielsen. Replicator neural networks for universal optimal source coding. Science, 269: 1860-1863, 1995.
[88] P.J.Huber. Robust Statistics. Wiley, 1981.
[99] P.J.Huber. Projection pursuit. The Annals of Statistics, 13(2): 435-475, 1985.
[90] J.Hurri, A. Hyvrinen, and E. ja. Wavelets and natural image statistics. In Proc. Scandinavian Conf. On Image Analysis'97, Lappenranta, Finland, 1997.
[91] A. Hyvrinen. A family of fixed-point algorithm for independent component analysis. In Proc. IEEE ICASSP'97, P.3917-3920, Munich Germany, 1997.
[92] A. Hyvrinen. One-unit contrast functions for ICA: A statistical analysis. In Neural Network for Signal Processing Ⅶ, P.388-397, Amelia Island, Florida, 1997.
[92] A. Hyvrinen. New approximations of differential entropy for independent component analysis and projection pursuit. In Advances in Neural Information Processing Systems, volume 10, pages 273-279. MIT Press, 1998.
[93] A. Hyvrinen. The fixed-point algorithm and maximum likelihood estimation for independent component analysis. Neural Processing Letters, 10(1): 1-5, 1999.
[94] A. Hyvrinen and R.Cristescu. A fast algorithm for estimating overcomplete ICA bases for image windows. In Proc. Int. Joint Conf. on Neural Networks, Washington ,D.C., 1999.
[95] A. Hyvrinen and E. Oja. Simple neuron models for independent component analysis. Int.
Journal of Neural Systems, 7(6):671-687, 1996.
[96] A. Hyvrinen and E. Oja. Independent component analysis by general nonlinear Hebbian-like learning rules. Signal Processing, 64(3):301-313, 1998.
[97] I.T. Jolliffe. Principal Component Analysis. Springer-Verlag, 1986.
[98] M.C. Jones and Sibon. What is prejection pursuit? J.of the Royal Statistics Society, ser. A, 150: 1-36, 1987.
[99] C. Jutten and J. Herault. Blind separation of sources, part Ⅰ: An adaptive algorithm based on neuromimetic architecture. Signal Processing, 24:1-10, 1991.
[100] J.Karhunen, A. Hyvrinen and R.Vigario. Applications of neural blind separation to signal and image processing. In Proc.IEEE Int. Conf. on ICASSP'97, P.131-134, Munich,Germany, 1997
[101] J.Karhunen and J.Joutsensalo. Respresentatin and separation of signals using nonlinear PCA type learning. Neural Networks, 7(1): 113-127, 1994.
[102] J.Karhunen and J.Joutsensalo. Generalizations of principal component analysis, optimization problems, and neural networks. Neural Networks, 8(4): 549-562, 1995.
[103] J.Karhunen, E.Oja, L. Wang and J.Joutsensalo. A class of neural networks for IEEE Trans. on Neural Networks, 8(3): 486-504, 1997.
[104] Karhunen, J., and Pajunen, E Blind Source Separation Using Least-Squares Type Adaptive Algorithms. In Proc. IEEE 1997 Int. Conf on Acoustics, Speech. and Signal Processing(ICASSP'97), April 21-24, 1997, Munich, Germany, pp. 3361-3364.
[105] M.Kendall. Multivariate Analysis. Charles Griffin&Co., 1975.
[106] K. Kiviluoto and E. Oja. Independent component analysis for parallel financial time series. In Proc. ICONIP'98, volume 2, pages 895-898, Tokyo, Japan, 1998.
[107] McKeown, M., Makeig, S., Brown, G., Jung, T-E, Kindermann, S., Bell, Iragui, V. and Sejnowski, T.J., Blind separation of functional magnetic resonance imaging (fMRI) data, Human Brain Mapping,. 6:160,18, 1998a.
[108] E. Moulines, J.-E Cardoso, and E. Gassiat. Maximum likelihood for blind separation and deconvolution of noisy signals using mixture models. In Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP'97), pages 3617-3620, Munich, Germany, 1997 o
[109] E. Oja. The nonlinear PCA learning rule in independent component analysis. Neurocomputing, 17(1):25-46, 1997.
[110] B. A. Olshausen and D. J. Field. Sparse coding with an overcomplete basis set: A strategy employed by V1? Vision Research, 37:3311-3325, 1997.
[111] Pham D. T., Garat Ph. et Jutten Ch. (1992) Separation of mixture of independent sources through a maximum likelihood approach. Dans Signal Processing Ⅵ, Proceeding EUSIPCO'92, 771-774, Eds. J. Vandewalle, R. Boite, M. Moonen and A. Oosterlinck.
[112] T. Ristaniemi, K. Raju, J. Karhunen, E. Oja, "Jammer Cancellation in DS-CDMA Array Systems: Pre and Post Switching of ICA and RAKE", In Proc. of the IEEE International Symposium on Neural Networks for Signal Processing (NNSP), Martingy, Switzerland. September 2-4, 2002, pp 495-504
[113] R.N. Vigario," Extraction of ocular artifacts from EEG using independent component analysis", Electroencephy and clinical Neurophysiology, No. 103, pp.395404,1997.