运动疲劳过程中脑电信号特征提取仿真
发布时间：2018-07-20 21:06
【摘要】：对运动疲劳过程中脑电信号特征进行准确提取,可以为运动疲劳损伤的治疗提供科学依据。对运动疲劳过程中不同层次脑电信号中的脑电波进行分解是进行脑电信号特征提取的基础,而传统方法利用可预测性的选取嵌入维数方法计算脑电信号序列的嵌入维数,利用相空间重构对整体信号特征提取,但是不能对不同层次脑电波进行分解,导致脑电信号特征提取精度差。提出基于模糊熵的运动疲劳过程中脑电信号特征提取方法。上述方法先融合于小波变换理论,将运动疲劳过程中不同层次脑电信号中的脑电波进行合理分解,计算不同层次脑电节律频带中小波系数的能量均值与均值差,将能量均值与均值差作为特征向量,构建FISHER线性分类器对运动疲劳中的意识疲劳信号分类。仿真结果表明,所提方法可以有效地完成对运动疲劳过程中脑电信号特征提取。
[Abstract]:The accurate extraction of EEG signals during exercise fatigue can provide scientific basis for the treatment of sports fatigue injury. The decomposition of EEG signals at different levels during exercise fatigue is the basis of EEG feature extraction, while the embedded dimension of EEG sequences is calculated by the traditional method, which is based on predictive selection of embedding dimension. The phase space reconstruction is used to extract the feature of the whole signal, but the EEG signal can not be decomposed at different levels, which leads to the poor precision of the feature extraction of the EEG signal. A method for feature extraction of EEG signals during exercise fatigue based on fuzzy entropy is proposed. Firstly, the method is combined with wavelet transform theory to decompose EEG waves in different levels of EEG during exercise fatigue, and calculate the difference of energy mean and mean value of wavelet coefficients in different levels of EEG rhythm frequency band. Based on the difference between mean energy and mean value as eigenvector, a fish linear classifier is constructed to classify the conscious fatigue signals in motion fatigue. Simulation results show that the proposed method can effectively extract the feature of EEG during exercise fatigue.
【作者单位】： 郑州大学西亚斯国际学院;
【基金】：基金项目:2015年河南省教育技术装备和实践教育研究(GZS084)
【分类号】：R87;TN911.7
本文编号：2134775
[Abstract]:The accurate extraction of EEG signals during exercise fatigue can provide scientific basis for the treatment of sports fatigue injury. The decomposition of EEG signals at different levels during exercise fatigue is the basis of EEG feature extraction, while the embedded dimension of EEG sequences is calculated by the traditional method, which is based on predictive selection of embedding dimension. The phase space reconstruction is used to extract the feature of the whole signal, but the EEG signal can not be decomposed at different levels, which leads to the poor precision of the feature extraction of the EEG signal. A method for feature extraction of EEG signals during exercise fatigue based on fuzzy entropy is proposed. Firstly, the method is combined with wavelet transform theory to decompose EEG waves in different levels of EEG during exercise fatigue, and calculate the difference of energy mean and mean value of wavelet coefficients in different levels of EEG rhythm frequency band. Based on the difference between mean energy and mean value as eigenvector, a fish linear classifier is constructed to classify the conscious fatigue signals in motion fatigue. Simulation results show that the proposed method can effectively extract the feature of EEG during exercise fatigue.
【作者单位】： 郑州大学西亚斯国际学院;
【基金】：基金项目:2015年河南省教育技术装备和实践教育研究(GZS084)
【分类号】：R87;TN911.7
【相似文献】
相关期刊论文 前2条
1 黄力宇,程敬之;急性轻中度缺氧对脑电信号复杂度的影响[J];航天医学与医学工程;2000年04期
2 ;[J];;年期
，本文编号：2134775
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2134775.html
