摘要
生物医学信号检测和处理过程中,50Hz(或60Hz)工频干扰的消除一直非常重要的技术问题。本文提出了一种基于独立分量分析的工频干扰消除新方法。我们用该方法对模拟数据和实测脑电(EEG)信号中的工频干扰进行消除,并与自适应滤波工频干扰消除方法进行了比较。实验表明,本文所提出的方法在工频干扰消除上具有一定的优势。
In the field of signal processing, interference from power lines (50 or 60Hz) continues to be a problem because amplitude of the interference is comparable to the amplitude of the signal. In this paper, according to the theory of statistical signal processing and blind source separation, we propose a new idea of removing power line interference. We consider the fact that the power line interference and the biomedical signal are independent each other, so the ICA method of blind source separation can be applied to remove 50Hz(or 60Hz) interference. The comparison between ICA-based method and LMS adaptive filter is given. The experiment results using simulation data and real-life EEG data show that the ICA-based method is more effective.
出处
《信号处理》
CSCD
2003年第1期81-84,共4页
Journal of Signal Processing
基金
国家自然科学基金资助项目(60071023
60271024)
安徽省自然科学基金资助项目(0043214)
关键词
盲源分离方法
工频干扰
信号检测
信号采集
生物医学
blind source separation
independent component analysis
power line interference
adaptive filter
