经验模态分解法与小波分解法应用於
滤除心电讯号的基准线飘移问题之比较
吴顺德,陈思予,陈虹伯
国立台湾师范大学机电科技学系
10610台北市大安区和平东路一段162号
E-mail: sdwu@ntnu.edu.tw
摘要
讯号处理在科学以及工程领域上皆是很重要的课题.自然界的讯号,大多都是非稳态(时变),非线性过程,故往往得到的讯号包含了杂讯的部分.传统的傅立叶转换,其处理的讯号限制为:线性,稳态(非时变)过程,所以并不能处理大部分的讯号特性.
经验模态分解法(Empirical mode decomposition, EMD)对於非线性,非稳态讯号提供一种多尺度(multi-scale),适应性(adaptive)的解析方式,其多尺度的特性与小波转换(Wavelet Transform, WT)的解析概念是相似的.本研究针对心电图(Electrocardiogram, ECG)讯号中基准线飘移(baseline wander)的问题,分别利用经验模态分解法与小波分解法滤除后,进行生医讯号(biomedical signals)物理意义之探讨以及杂讯滤除效率性之比较.
关键词:时间-尺度分解,经验模态分解法,小波分解法,心电图讯号,杂讯滤除.
Abstract
Signal processing is very important for science and engineering researches. Real world signal are often noisy, non-stationary, and obtained from nonlinear systems. However, the majority of signal processing algorithms proposed in the literature such as Fourier transform are better suited for analyzing the linear stationary signals with weak noise.
Empirical mode decomposition (EMD) provides a powerful tool for adaptive multi-scale analysis of nonlinear and non-stationary signals, and the multi-scale characteristic is similar to wavelet transform. In this paper, the EMD and wavelet decomposition (Harr transform) were used to process baseline wander in Electrocardiogram (ECG) signals for de-noising. It is very important to reserve the physical meanings for processing biomedical signals, and the results of EMD and wavelet decomposition were compared after de-noising.

Keywords: time-scale decomposition, empirical mode decomposition (EMD), wavelet decomposition, Electrocardiogram (ECG) signals, de-noising.
1. 前言
由於社会快速变迁,无论是饮食习惯趋於快速化,生活习惯趋於多样化,人口结构趋於老年化等因素,衍生出许多的慢性疾病,特别是心血管方面的病变对国人的健康带来很大的影响.
心电图是记录心脏收缩时的电流运动,利用摆放身体特定部位电极不同的组合,产生十二个不同角度的电压记录,所以又叫十二诱导心电图(12-lead ECG).在量测心电讯号时,会受到人为干扰(artifact)的影响,而无法得到正确的心电图,因此如何滤除这些干扰所产生的杂讯,是一个重要的课题.
利用数位讯号处理(DSP)滤波器,会造成讯号相位延迟的现象;若是运用讯号分解的方式,就不会有这样的问题.近年来,许多学者针对一种名为经验模态分解法(EMD)的讯号处理方式进行研究,并对於此法与小波转换的分解法提出许多的特性比较[1].本研究便是将这两种方式,应用於心电讯号中基准线飘移问题上,根据结果进行比较与探讨.

下一页