摘要
针对小波独立分量分析法(W-ICA)在心电信号消噪中小波变换缺乏自适应性,且较难选取最优小波基的问题,提出了一种将经验模式分解与独立分量分析相结合的小波独立分量分析法。该方法结合经验模式分解与独立分量分析各自的优点,利用经验模式分解对心电信号进行自适应分解,然后应用独立分量分析法对选取的本征模态函数进行分离,将分离后的分量进行两层重构,从而得消噪后的心电信号。通过利用MIT-BIH心率失常数据库中的数据进行仿真实验,结果表明该方法可以较好地消除心电信号中的噪声,消噪后信号与原信号的相关系数可达0.96。
To improve the adaptability and difficulties in selecting wavelet base when using the Wavelet independent component analysis(W-ICA)to cancel the noise in electrocardiogram(ECG),a new method,E-ICA,is proposed which combines the advantages of empirical mode decomposition(EMD)with independent component analysis.The ECG is decomposed adaptively by EMD and then the proper intrinsic mode function is selected by ICA.The ICA components are reconstructed by two-layer reconstruction,that is the inverse ICA and inverse EMD,to remove the noise in ECG.The simulation based on MIT-BIH database shows that E-ICA can eliminate the noise efficiently,and the correlation coefficient of clean ECG and original ECG is 0.96.
作者
行鸿彦
侯进勇
Xing Hong-yan;Hou Jin-yong(School of Electronicsand Information nEngineering,Nanjing University of Information Science and Technology,Nanjing210044,Jiangsu Province,China)
出处
《中国组织工程研究与临床康复》
CAS
CSCD
北大核心
2009年第4期651-654,共4页
Journal of Clinical Rehabilitative Tissue Engineering Research
基金
江苏省"青蓝工程"中青年学术带头人基金资助
