改进的LMD算法及其在EEG信号特征提取中的应用  

The Improved LMD Algorithm and Its Application in the EEG Feature Extraction

作　　者：朱晓军[1] 吕士钦[2] 王延菲[1] 余雪丽[1] 

机构地区：[1]太原理工大学计算机科学与技术学院,太原030024 [2]太原理工大学数学学院,太原030024

出　　处：《太原理工大学学报》2012年第3期339-343,共5页Journal of Taiyuan University of Technology

基　　金：国家自然科学基金(60873139);教育部高校博士点基金(200801120007)

摘　　要：针对LMD方法在对脑电信号处理时会产生端点效应,为了降低这种端点效应的影响,增强脑电信号的特征提取效果,提出了一种基于相似波形加权匹配的端点延拓算法。该算法主要利用在一段波形中相似子波会反复出现这一特点,提取与端点处波段相似的子波,求得其加权平均波,并利用得到的平均波对原始信号的左右端点进行延拓。仿真结果表明,与传统的LMD方法分解相比,该算法能够有效抑制传统LMD的端点效应,更好地提取脑电的特征信息。Aimed at the end effects produced in the process of local mean decomposition,a new method based on weighted matching similar waveform was proposed to reduce the impact of end effects and strengthen the effect of EEG feature extraction, Take advantage of the feature that similar wavelets repeat in one waveform, wavelets being sim ilar to tip waveform can be extracted to obtain the weighted averaged wavelet.T hen this wavelet is used to extend the original signal.The simulation results indic ate that this new method restrained the LMD end effects and extracted the EEG fe atures effectively.

关 键 词：局部均值分解 端点效应 相似波形加权匹配 特征提取 脑电信号 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]