基于HilbertHuang边际谱的肌电信号特征提取*

作者：叶明，孙承奎，马文杰 日期：2008-11-03/span> 浏览：3679 查看PDF文档

基于HilbertHuang边际谱的肌电信号特征提取*

叶明，孙承奎，马文杰
（杭州电子科技大学 智能控制机器人研究所，浙江 杭州 310018）

摘要：提出了一种基于固有模态函数(IMF)瞬时频率的自适应边际谱分段方法。该方法基于各层IMF的频率有效度，选择合适的IMF，通过自适应边际谱分段方法来确定边际谱分段，然后求取边际谱各段能量，以获得边际谱形状特征。经实验得到了一种优化的归一化处理方法，最后通过仿真得到了HHT边际谱形状特征。
关键词：固有模态函数；表面肌电信号；HilbertHuang变换；特征提取
中图分类号：TP183；TP274文献标识码：A文章编号：1001-4551(2008)09-0080-05

SEMG feature extraction based on the HHT
YE Ming, SUN Chengkui, MA Wenjie
(Institute of Intelligence Control and Robots, Hangzhou Dianzi University, Hangzhou 310018, China)
Abstract: In the HHT margin spectrum shape feature extraction algorithm, there were four steps. Firstly, the useful intrinsic made function(IMF) was selected and the selfadapting subsection for HHT margin spectrum was determined. The algorithm was proposed . Secondly, the energy of each margin spectrum subsection was considered as the sharp feature of the margin spectrum for each surface electronmyography(SEMG). Thirdly, two feature vectors corresponding to the two channels SEMG were merged into one handmotion feature.  Finally, each dimension of the feature vector was normalized based on the method. Ultimately, the margin spectrum shape feature vector was obtained, according to the above approach.
Key words: intrinsic made function(IMF); surface electronmyography(SEMG); HilbertHuang transform(HHT); feature extraction
参考文献(Reference):
［1］DELUCA C. Phsiology and mathematics of myoelectric signals［J］. IEEE Trans. Biomed Eng.,1979,26(6):313-325.
［2］罗志增,王人成.仿生假手触滑觉研究［J］.传感技术学报,2003,16(3):233-237.
［3］柴新禹,吴朝霞.时频分析方法及其在医学信号处理中的应用［J］.生物医学工程学杂志，2001,18（1）:138-144.
［4］HUANG N E. The empirical mode composition and the Hilbert spectrum for nonlinear and nonstationary time series analysis［J］. Proc R Soc Lond A, 454:903-995.
［5］罗志增.任晓亮.肌电信号的拾取和预处理［J］.传感技术学报,2004(2):221-223.
［6］罗志增.表面肌电信号的AR参数模型分析方法[J].传感技术学报,2003,16(4):385-386.