摘要
ICA算法是求解盲源分离问题的有效算法。建立了ICA算法的数学模型,对模型的求解条件及多解性进行了分析。给出一种基于负熵极大的FastICA算法,讨论该算法在地震信号去噪中的应用。仿真实验验证了该算法的有效性。
ICA algorithm is a effective algorithm for blind source separation(BSS) problem.This paper established the mathematical model of ICA algorithm,and analyzed its solving condition and multiplicity solutions.Gave a Fast ICA algorithm based on maximum negentropy,discussed its application in seismic signal noise elimination.A simulation experiment verifies the effectiveness of the algorithm.
引文
[1]COMON P.Independent component analysis,a new concept[J].Sig-nal Processing,1994,36(3):287-314.
[2]BELL AJ,SEJNOWSKI TJ.An information-maximization approach toblind separation and blind deconvolution[J].Neural Computation,1995,7(6):1004-1034.
[3]HYVARINEN A,OJA E.A fast fixed-point algorithm for independentcomponent analysis[J].Neural Computation,1997,9(7):1483-1492.
[4]LEE TW,GIROLAMI M,SEJNOWSKI TJ,et al.Independent compo-nent analysis using an extended infomax algorithm for mixed sub-Gaussian and super-Gaussian sourees[J].Neural Computation,1999,11(2):417-441.
[5]HYVARINEN A,OJA E.Independent component analysis:algorithmsand applications[J].Neural Networks,2000,13(4-5):411-430.
[6]百培瑞,牛海军,师卫.基于盲分离的肺音信号中心音干扰的去除[J].太原理工大学学报,2002,33(3):329-331.
[7]夏文静,傅行军.基于ICA在强背景噪声振动信号中的去噪研究[J].汽轮机技术,2006,48(2):121-123.
[8]钟飞,谭中军,史铁林,等.基于ICA和小波变换的轴承故障特征提取[J].微计算机信息,2007,23(10):154-155.
[9]曾生根,夏德深.独立分量分析在多光谱遥感图像分类中的应用[J].计算机工程与应用,2004,40(21):108-110.
[10]胡天跃.地震资料叠前去噪技术的现状与未来[J].地球物理学进展,2002,2(17):218-223.
[11]AL-YAHYA K M.Application of the partial Karhunen-Loeve trans-form to suppress random noise in seismic sections[J].GeophysicalProspecting,1993,39(1):77-93.
[12]DAUBECHIESL S W.Factoring wavelet transform into lifting steps[J].Journal of Fourier Analysis Application,1994,4(3):247-269.
[13]陈香朋,曹思远.第二代小波变换及其在地震信号去噪中的应用[J].石油物探,2004(6):51-54.
[14]杨福生,洪波.独立分量分析的原理与应用[M].北京:清华大学出版社,2006:7-46.
[15]康永强,田梦君.一种快速ICA算法及其在脑电信号处理中的应用[J].科学技术与工程,2006,6(3):529-533.
[16]SHI Zhen-wei,TANG Huan-wen,TANG Yi-yuan.A new fixed-pointalgorithm for independent component analysis[J].Neurocomputing,2004,56:467-473.
[17]刘喜武,刘洪,李幼铭.快速独立分量变换与去噪初探[J].中国科学院研究生院学报,2003,20(4):488-492.
[18]焦卫东,杨世锡,吴昭同.基于独立分量分析的噪声消除技术研究[J].浙江大学学报:工学版,2004,38(7):872-876.