基于快速独立分量分析的脑电波信号降噪

发布时间：2019-03-05 16:25

【摘要】：针对原始脑电波信号存在非平稳性且非常容易受到各种信号的干扰等特点,对基于小波变换和快速独立分量分析的脑电波信号的滤波降噪方法进行了研究,说明了小波变换和快速独立分量分析的降噪原理,并通过对利用MindSet耳机采集得到的原始脑电波数据的处理,证明了快速独立分量分析算法可以将原始脑电波信号中包含的心电伪迹和工频干扰等多种干扰信号成功地分离出来,同时比较了两种方法的性能,验证了基于快速独立分量分析的降噪方法具有明显的优越性。
[Abstract]:In view of the non-stationarity of the original EEG signal and its very easy to be interfered by various signals, the filtering and de-noising method of EEG signal based on wavelet transform and fast independent component analysis is studied, and the method of filtering and denoising based on wavelet transform and fast independent component analysis is studied. The de-noising principle of wavelet transform and fast independent component analysis is explained, and the processing of the original EEG data obtained by using MindSet headphones is presented. It is proved that the fast independent component analysis algorithm can successfully separate the ECG artifacts and power-frequency interference signals contained in the original EEG signals, and compare the performance of the two methods at the same time. The advantages of the denoising method based on fast independent component analysis (FICA) are verified.
【作者单位】：南京理工大学自动化学院;
【基金】：国家自然科学基金(51175266) 江苏省高校自然科学基金(12KJB510008) 江苏省普通高校研究生科研创新计划(CXZZ13-0207)
【分类号】：TN911.4

【参考文献】