基于快速独立分量分析的脑电波信号降噪

EEG Signal Denoising Based on Fast Independent Component Analysis

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 terms of non-stationary for original EEG signals and very vulnerable to a variety of signal interference, this paper studied the method for EEG signal denoising based on wavelet transform and fast independent component analysis. It has explained the denoising principle of wavelet transform and fast independent component analysis, and it has proved that independent component analysis can effectively separate ECG artifacts and power line interference from original EEG siganls by processing original EEG data collected by MindSet headset and compared the performance of the two methods simultaneously. Then, it has proved the superiority of the denoising method based on fast independent component analysis.