小波和主分量分析方法研究思维脑电

MENTAL EEG ANALYSIS USING WAVELET TRANSFORM AND PRINCIPAL COMPONENT ANALYSIS

研究自发脑电和思维活动的关系。利用小波和主分量分析结合的WPCA算法对不同思维任务记录的六导脑电进行处理,并对思维特征的频谱能量和变化率等多指标进行综合分析和计算。结果表明WPCA算法不仅可以实现噪声的去除,而且能提高主分量的贡献率,降低输入矢量的维数。对脑电主分量的分析揭示了脑电与思维个体、思维种类、复杂度以及注意力的联系,思维任务的神经网络分类结果验证了WPCA方法研究脑电和思维的有效性,为进一步理解认知和思维过程,实现对思维的定位和分类提供了依据。

To explore the relationship between spontaneous EEG and cognitive tasks. An algorithm called WPCA, which is based on wavelet transform and principal component analysis (PCA), was used to process the six channel EEG. The indices of spectral energy and variation rate were calculated, analyzed and stated. The result shows that the proposed WPCA algorithm not only has a good character in noise removing, but also in centralizing the component energy and decreasing the data dimension. The analysis of EEG component revealed the relationship between individual EEG and the mental tasks' kinds, complexity and attention. The result of the NN classification shows the efficiency of the WPCA method. The research is applicable to localize and classify the cognitive tasks and to study the mental function .