双声道心音信号的欠定盲分离方法的研究  

Research of Underdetermined Blind Source Separation Method on Dual-channel Heart Sound Signal

作　　者：王路飞[1] 成谢锋[1] 孙科学[1] 

机构地区：[1]南京邮电大学电子科学与工程学院,江苏南京210046

出　　处：《南京邮电大学学报（自然科学版）》2012年第4期9-13,共5页Journal of Nanjing University of Posts and Telecommunications：Natural Science Edition

基　　金：国家自然科学基金(61073115;61271334;61073115);南京邮电大学校科研基金(NY207139;JG03311JX12;2012XSG04;NY210036)资助项目

摘　　要：基于两路人体心声信号的专用检测平台,提出了一种针对双声道心音信号的欠定盲分离方法。首先对数据点进行频域聚类计算,利用观测信号中稀疏信号所表现出的特征对源信号个数进行分析,以实现对混叠矩阵的估计;然后根据观测信号的散列图分离出其中一路或多路源信号,从而使观测信号变为适定或者超定;最后用现有的适定或者超定盲分离方法分离剩余源信号。分别对一组人工混合信号和实际检测的双声道心声信号进行欠定盲分离实验,验证了本方法的有效性和可行性。This paper presents an underdetermined blind source separation method for the dual-channel heart sound signal based on a special test platform of two proprietary human voice signals.Firstly we performed the cluster calculation in the frequency domain for the data points,and analyzed the number of source signals to achieve the estimated mixing matrix by using the sparse signal characteristics shown in the observed signals.Then we have separated one or multiple source signals according to the scatter plot of observed signals.The observed signals are thereby turned into well-posed or overdetermined signals.Finally we have separated the remaining source signals by using traditional well-posed or over determined blind source separation method.The validity and the feasibility of this method have been confirmed through the separation experiment on the artificial mixed-signals and the actual detected two-channel voice signals respectively.

关 键 词：稀疏分量分析 心音信号 双声道欠定盲分离 

分 类 号：TN911.6[电子电信—通信与信息系统]