摘要
为了有效实现单路心音混合信号的盲分离,本文提出了一种基于EMD分解和独立成分函数的单路含噪心音信号盲源分离的方法。首先讨论了单路混合信号的分离模型,含噪信号预处理的方法,以及如何利用EMD变换进行窄带分层和获取独立成分函数的技术;然后通过独立成分函数作为基函数对单路含噪心音信号进行分解,使单路心音信号由一维向量转变为多维向量,从而实现心音信号的盲源分离;最后通过实际的心音分离实验,验证了本方法的实用性,其分离结果的相似度达到0.9792。
Based on empirical mode decomposition(EMD) layered technique and independent primary functions, a signal processing technique to accomplish blind source separation given only a single-channel mixture heart sound signal is introduced in the paper. Firstly the single-channel mixed heart sound signal separation model, the noisy signal pre-processing methods before BSS, and how to carry on the layered using EMD to gain the technique of the independent primary functions are discussed. Then through a combination of independent primary function into the one-way mixed heart sound signal, multi-dimensional vector is translated from one-dimensional vector, therefore its blind source separa- tion is achieved. Finally through the signal separation experiment for single-channel mixed noisy heart sounds, the validity and usability of the method is verified, the similarity of separation result is 0.9792.
出处
《电子测量与仪器学报》
CSCD
2012年第6期498-502,共5页
Journal of Electronic Measurement and Instrumentation
关键词
盲源分离
经验模值分解
独立基函数
单路含噪信号
Blind source separation
empirical mode decomposition
independent primary function
sin-gle-channel noise signal
作者简介
孙科学:1981年出生,硕士,现为南京邮电大学电子科学与工程学院讲师。主要研究方向为嵌入式技术与通信信号处理。E-mail:sunkx@njupt.edu.cn成谢锋:1955年出生,现为南京邮电大学电子科学与工程学院教授。主要研究方向为智能信息处理、智能仪表。E—mail:chengxf@njupt.edu.cn