EMD及其在脑电分析中的应用研究  

EMD and Research Employed to Electroencephalogram Analysis

作　　者：周百新[1] 梁忠诚[1] 赵阳[1] 王蔚[1] 赵新红[1] 

机构地区：[1]南京师范大学电气与自动化工程学院,江苏南京210042

出　　处：《南京师范大学学报（工程技术版）》2007年第1期8-13,共6页Journal of Nanjing Normal University(Engineering and Technology Edition)

基　　金：南京师范大学"211"学科建设基金(1843202529)资助项目

摘　　要：基于经验模态分解(EMD)的数据分析方法,是一种针对非线性、非平稳信号处理的新方法.使用EMD法可以将任意复杂的数据信号分解为多个有限的、数据量较小的“本征模函数”(IMF).这些本征模函数很适合求其Hilbert变换.信号的局部能量和瞬时频率都可以从其本征模函数中推导出来.这个完整的能量-频率-时间关系称为Hilbert谱,它是一种分析非线性、非平稳信号的理想方法.介绍了EMD法的原理和实现过程,给出了多个实例的本征模函数和Hilbert谱.并展示了它在非稳态信号处理中的特性.同时,还探索将这种基于EMD的分析方法应用于脑电信号的分析中,并给出了脑电信号的部分本征模函数(IMF)分量及Hilbert振幅和频谱图.试图用一种新的方法分析复杂的非平稳脑电信号.Based on the Empirical Mode Decomposition （EMD）, a new data analysis method for processing nonlinear and non-stationary signal has been developed. The key part of method is the EMD method with which any complicated data signal can be decomposed into a finite and often small number of Intrinsic Mode Functions （IMF） that admit well - behaved Hilbert transforms. The local energy and instantaneous frequency of the signal can be derived from the IMFs. This full energy-frequency-time relation of the data signal is called the Hilbert spectrum. It is an ideal for nonlinear and non-stationary data analysis. The principle and sifting process of the EMD method are introduced, and analysis results about some actual examples with EMD method are also shown. The possibility that the EMD method is applied in analysis of an Electroencephalogram （EEG） is discussed and a few Intrinsic Mode Function （IMF） components and Hilbert spectrum of the EEG are given. The EMD method presented is attempted to analysis the complex non - stationary EEG data signal.

关 键 词：经验模态分析(EMD) 脑电信号 非平稳信号 本征模函数(IMF) 希尔伯特谱 

分 类 号：TP274.5[自动化与计算机技术—检测技术与自动化装置]