【摘要】 独立分量分析 (ICA)是近年来涌现的用于盲信号分离的新技术 ,本文利用独立分量分析对心音信号进行了预处理 :消除工频干扰。心音信号由自制的心音传感器获得。在分析了独立分量分析的基本原理的基础上 ,建立了基于互信息极小的目标函数 ,研究了目标函数优化的迭代算法 ,给出了利用此算法的ICA实现步骤。实验结果表明 ,利用独立分量分析有效地对心音信号进行预处理 ,能成功地从心音中分离出工频干扰信号。更多还原

【Abstract】 Independent component analysis(ICA) is a novel method developed in recent years for Blind Source Separation. In this paper, the principles of ICA were investigated. A contrast function was built based on mutual information minimization; the iterative algorithm was presented; the step of ICA was given. The experiments have shown that ICA can successfully separate the PCG signal and power line interference from their mixed signals. ICA has better performance in comparison with other method.更多还原