• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2015, Vol. 37 ›› Issue (04): 790-795.

• 论文 • 上一篇    下一篇

基于小波包变换的脑电波信号降噪及特征提取

刘珑1,李胜1,王轶卿2   

  1. (1.南京理工大学自动化学院,江苏 南京 210094;2.南京工业大学自动化与电气工程学院,江苏 南京210009)
  • 收稿日期:2014-01-13 修回日期:2014-05-21 出版日期:2015-04-25 发布日期:2015-04-25
  • 基金资助:

    国家自然科学基金资助项目(51175266);江苏省高校自然科学基金资助项目(12KJB510008);江苏省普通高校研究生科研创新计划资助项目(CXZZ130207)

EEG signal denoising and feature extraction
based on wavelet packet transform 

LIU Long1,LI Sheng1,WANG Yiqing2   

  1. (1.School of Automation,Nanjing University of Science and Technology,Nanjing 210094;
    2.School of Automation and Electrical Engineering,Nanjing University of Technology,Nanjing 210009,China)
  • Received:2014-01-13 Revised:2014-05-21 Online:2015-04-25 Published:2015-04-25

摘要:

针对原始脑电波信号存在非平稳性且非常容易受到各种信号干扰等特点,对基于小波变换和小波包变换的脑电波信号的滤波降噪方法,和基于小波包变换的脑电波信号特征提取方法进行了研究。首先利用MindSet采集到原始脑电波数据,然后分别应用小波变换和小波包变换对其进行降噪处理,比较了两种方法的性能,验证了基于小波包变换的降噪方法的优越性和特征提取方法的有效性。

关键词: 脑电波信号, 降噪, 特征提取, 小波变换, 小波包变换

Abstract:

Since original EEG signals are non-stationary and very vulnerable to a variety of signal interference,we propose a method for EEG signal denoising based on wavelet transform and wavelet packet transform, and a method for EEG signal feature extraction based on wavelet packet transform.First,we use MindSet headset to collect original EEG data and denoise the data by wavelet transform and wavelet packet transform respectively. We compare the two methods,and the results prove the superiority of the proposed denoising method and the validity of the feature extraction method based on wavelet packet transform.

Key words: EEG signal;denoising;feature extraction;wavelet transform;wavelet packet transform