摘要
该文提出了一种脑电信号中自适应去除眼电和心电伪迹的算法。首先针对线性混合信号盲分离时,源信号概率密度与激活函数难以确定的困难,导出一种基于峭度的自适应盲源分离开关算法,然后将其应用于脑电信号的眼电与心电伪迹去除。实验证明,该算法具有良好的分离效果。
This paper presents an adaptive algorithm to remove VEOG and EKG artifacts in EEG.Firstly,an adaptive switching algorithm of blind source separation based on kurtosis is presents because it is difficult to know the probability density function and activation function for the linear mixed signal separation.Then the algorithm is used to remove the VEOG and EKG artifacts.The experiment shows that the algorithm has good separation efficiency.
出处
《杭州电子科技大学学报（自然科学版）》
2012年第6期129-132,共4页
Journal of Hangzhou Dianzi University：Natural Sciences
基金
国家自然科学基金资助项目(61172134)
关键词
盲源分离
峭度
脑电信号
伪迹去除
blind source separation
kurtosis
EEG
artifact removal
