摘要
盲信号处理技术是最近一段时间兴起的一个新的研究领域，它在实际中有着非常广泛的应用，盲信号源分离是盲信号处理技术的一个研究方向。盲分离是指在不知道源信号的特征，而且也不知道信号在传输通道中的混合过程，只假设源信号是相互统计独立的情况下，从观测信号中分离出源信号的技术。本文研究瞬时混合情况和卷积混合情况下的自适应盲分离算法。
盲分离算法需要用到高阶统计量，互累积量迫零法是一种直接利用高阶统计量的自适应盲分离算法。本文介绍了高阶累积量，通过互累积量的性质阐述了算法的基本思想，分析了算法的稳定性能，对算法进行了仿真，指出了算法的不足。
Infomax算法是另外一种自适应盲分离算法。Infomax算法通过引入非线性函数来引入高阶统计量，并利用信息论的基本知识推导出了盲分离算法。Infomax算法适用于瞬时混合情况下的盲分离和单通道盲解卷积问题，本文将Infomax算法推广到多通道卷积混合，对算法进行仿真，取得了不错的效果。
自然梯度下降法是一种新的最优化方法，自然梯度下降法是在黎曼空间下提出的，自然梯度相比于标准梯度有很多优点。自适应盲分离算法中的系数空间是黎曼空间，本文将自然梯度下降法应用于Infomax算法可以改进算法的性能，有利于算法的数值稳定性和提高算法的收敛速度。针对基于自然梯度算法的特殊结构，还详细分析了算法的性能，包括稳定条件，步长因子和非线性函数对算法的影响。
将盲分离算法应用于实际是最根本的目的。本文通过用盲分离算法对水池实验数据进行处理，初步探索了盲分离在水声环境下的应用，并得到了在水声环境下应用盲分离算法的一些结论，指出了以后需要改进的地方。
Blind signal processing is a new orientation in the field of signal processing, it also has broad range of application in daily life. Blind source separation is to separate sources from mixed signals, without knowing about the characters of the sources and how they are mixed, only assuming that the sources are independent with each other. Blind source separation is a part of blind signal processing. This paper only does research on adaptive blind source separation algorithm in two mixture models梚nstantaneous mixture and convolutive mixture.
Blind source separation needs high order statistics. The algorithm based on cross cumulants uses high order statistics directly. The basic idea of the algorithm is shown by the characteristics of cumulants; the performance of the algorithm is discussed; the shortcomings of it are pointed out.
Infomax algorithm is another adaptive blind source separation. Infomax uses high order statistics by non-linear function and the idea from Information Theory. The original Infomax is only applied in the instantaneous mixture and in the single channel blind deconvolution. This thesis applies the idea to blind sources separation in multiple channel convolutive mixture. Through simulation, we can show that the algorithm has good performance.
Natural gradient is a new optimization way that is proposed in a special space?Riemannian Space. The parameter space in blind source separation is Riemannian Space. Natural gradient has many better characters compared with normal gradient. This thesis applies natural gradient to Infomax algorithm. And with the special character of the natural gradient, the performance of the algorithm is discussed.
Applying blind source separation to the real situation is the basic goal. This paper uses the blind source separation algorithms to process the data received form an experiment in water tank.
引文
[1] Christian Jutten and Jeanny Herault, "Blind separation of sources, Part I: An adaptive algorithm based on neuromimetic arthitecture," Signal Processing, Vol.24, 1991, 1-10．
[2] Christian Jutten and Jeanny Herault, "Blind separation of sources, Part II: Problems statement," Signal Processing, Vol.24, 1991, 11-20．
[3] E. Sorouchyari, "Blind separation of sources, Part III: Stability analysis," Signal Processing, Vol. 24, 1991, 21-29．
[4] Odile Macchi, Eric Moreau, "Self-Adaptive Source Separation, Part I: Convergence Analysis of a Direct Linear Network Controled by the Herault-Jutten Algorithm," IEEE TRANSACTIONS ON SIGNAL PROCESSING, Vol. 45, NO. 4, APRIL 1997, pp. 918-925．
[5] Eric Moreau and Odile Macchi, "Comparison of the Direct, Feedback, and Mixed Linear Network," IEEE TRANSACTIONS ON SIGNAL PROCESSING, Vol. 46, NO. 1, JANUARY, 1998, pp. 39-50．
[6] Pierre Comon, "Independent component analysis, Anew concept?," Signal Processing, Vol,36, 1994, 287-314．
[7] Daniel Yellin and Ehud Weinstein, "Criteria for Multichannel Signal Separation," IEEE TRANSACTIONS ON SIGNAL PROCESSING, Vol. 42, NO. 8, AUGUST, 1994, pp. 2158-2167．
[8] Simon Haykin, "Adaptive Filter Theory(Fourth Edition)," 电子工业出版社， 2002, 208-212．
[9] Jean-Francois Cardoso, "Blind Signal Separation: Statistical Principles," PROCEEDINGS OF THE IEEE, Vol. 86, NO. 10, OCTOBER 1998, pp. 2009-2025．
[10] Hoang-Lan Nguyen Thi, Christian Jutten, "Blind source separation for convolutive mixtures," Signal Processing, Vol.45, 1995, 209-214．
[11] Sergio Cruces, Luis Castedo, "Stability analysis of adaptive algorithms for blind source separation of convolutive mixtures," Signal Processing, Vol. 8, 1999,265-275．
[12] A. Bell and T. Sejnowski, "An Information-Maximization Approach to Blind Separation and Blind Deconvolution," Neural computation, 7(6) , 1995, 1129-1159．
[13] A. Cichocki, R. Unbehauen, L. Moszczynski, and E. Rummert, "A new on-line adaptive learning algorithm for blind separation of source signals," Proc.
ISANN-94, Taiwan, Dec. 1994, ch. 8, pp. 416-471．
[14] A. Cichocki, S. Amari, M. Adachi, and W. Kasprzak, "Self-adaptive neural networks for blind separation of sources," Proc. 1996 IEEE Trans. Int. Symp. Circuits and Systems, vol. 2, May 1996, pp. 157-160．
[15] A. Cichocki, B. Orsier, A. D. Back, and S. Amari, "On-line adaptive algorithms in non stationary enviroments using amodified conjugate gradient approach," Proc. IEEE Workshop Neural Networks for Signal Processing, Sept. 1997, pp. 316-325．
[16] A. Cichocki, R. Unbehauen, L. Mosezczynski, and E. Rummert, "A new on-line adaptive learning algorithm for blind separation of source signals," Proc. ISANN-94, Taiwan, Eec., ch. 8, 1994, pp. 406-411
[17] A. Cichocki, S. Amari, M. Adachi, and W. Kasprzak, "Self-adaptive neural networks for blind separation of sources," in Proc. 1996 IEEE Int. Symp. Circuit and Systems, Vol. 2, May 1996, pp. 157-160．
[18] Shun-Ichi Amari and Andrzej Cichocki, "Adaptive Blind Signal Processing-Neural Network Approaches," PROCEEDINGS OF THE IEEE, Vol. 86, NO. 10, OCTOBER 1998, pp. 2026-2048．
[19] E. Vittoz and X. Arreguit, "CMOS integration of Herault-Jutten cells for separation of sources," Portland, Oregon, 8, May 1989．
[20] Lang Tong, Ruey-wen Liu, Victor C. Soon and Yih-Fang Huang, "Indeterminacy and Identifiability of Blind Identification," IEEE TRANSACTIONS OF CIRCUITS AND SYSTEMS, Vol. 38, NO. 5, MAY 1991, pp. 499-509．
[21] S Haykin Ed., "Unsupervising Adaptive Filtering, Volume I: Blind Sources Separation," John Wiley & Sons Inc., 2000．
[22] S Haykin Ed., "Unsupervising Adaptive Filtering, Volume II: Blind Deconvolution," John Wiley & Sons Inc., 2000．
[23] A. Cichocki and S. Amari, "Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications," John Wiley & Sons Inc., 2002．
[24] John G. Proakis, "Digital Communications(Third Edition), " 电子工业出版社， 2001．
[25] Sato, Y. "A Method of Self-Recovering Equalization for Multilevel Amplitude-Modulation Systems," IEEE Trans. Commun., vol. COM-23, June 1975, pp.679-682．
[26] Kenneth R. Castleman, "Digital Signal Processing," 电子工业出版社， 2002．
[27] 王惠刚，“自适应盲信号处理理论及应用研究”，西北工业大学，2002， 93-100。
[28] S. Amari, "Natural gradient works efficiently in learning," Neural Computation, vol. 10, pp. 251-276, Jan. 1998.
[29] A. Cichocki, J. Karhunen, W. Kasprrzak and R. Vigario, "Neural Networks for Blind Separation with Unknown Number of Sources," Neurocomputing, Vol. 24, 1999, pp.53-93.
[30] 傅祖芸，“信息论——基础理论与应用”，电子工业出版社，2001。
[31] 顾金海，叶学千，“水声学基础”，西北工业大学出版社。
[32] 田琬逸，张效民，“信号检测与估值”，西北工业大学出版社。
[33] 刘燕青，李志舜，王惠刚和钱刚，“基于累积量的盲分离算法的稳定性分析”，系统工程与电子技术(被录用，待发表)。
[34] 程云鹏，“矩阵论”，西北工业大学出版社，2000年1月。
[35] 胡广书，“数字信号处理”，清华大学出版社，1997年8月。