基于小波域隐马尔科夫模型的肌电信号滤波  

Wavelet-domain EMG de-noising using hidden Markov model

作　　者：罗志增[1] 王占玉[1] 

机构地区：[1]杭州电子科技大学机器人研究所,杭州310018

出　　处：《仪器仪表学报》2010年第11期2447-2452,共6页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金资助(60874102);863项目(2008AA04Z212)资助

摘　　要：为了消除肌电信号中的噪声并且保留信号的细节信息,本文提出了基于小波域隐马尔科夫模型的肌电信号消噪方法。该方法利用隐马尔科夫模型对表面肌电信号小波分解后的小波系数之间的相关性进行建模,运用训练模型算法(Expectation-Maximization algorithm,EM算法)估计出该模型的参数,以贝叶斯估计得到真实信号的小波系数,通过重构实现肌电信号的滤波。实验结果表明该方法能有效地去除肌电信号中的噪声,对进一步的特征提取和模式识别创造了良好的条件。In order to eliminate noise in EMG and hold details of the signal, Wavelet-domain EMG de-noising using hidden Markov model is proposed. The correlation between wavelet coefficients is modeled with HMM. The model parameters are estimated using Expectation-Maximization algorithm and the wavelet coefficients of the real signal are computed using Bayesian estimation. At last, de-noised EMG is achieved through reconstructing the wavelet coefficients. Experiment result shows that noise can be eliminated effectively with the method proposed in this paper, which is benefit for further extraction and pattern recognition of the signal.

关 键 词：肌电信号 小波域 隐马尔科夫 EM算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]