小波相关分析在脑-计算机接口系统中的研究

Research on the Wavelet Coherence in Brain-computer Interface

脑电信号本身很微弱、并伴有很强的背景噪声,但其中蕴藏着多种生理现象,有着非常重要的临床价值。针对特定思维诱发脑电信号的特点,提出一种确定其分布情况及提取其波形的方法。首先使用离散小波变换对诱发脑电信号进行多层分解,然后使用小波奇异点理论和小波相关分布,准确地确定诱发脑电信号的分布情况,并根据分析的结果重构出诱发脑电波形。在实际的脑-计算机接口系统中,为确保系统的准确性提供了有利的保证。

EEG signals are very weak, generally accompanied by strong background noises, but they reflect various physiological phenomena and make great sense in the clinical condition. Based on the characteristics of the specific thinking-evoked EEG signals, an approach is proposed to determine their distribution and pick up their waveform out of strong noises. The EEG signals are decomposed by the way of discrete wavelet transform. And then, using the wavelet singularity theory and the wavelet coherence analysis, the distribution of the evoked signals can be found out. As a result, according to this analysis result, the evoked signals can be picked up. In the practical brain-computer interface, this approach proves to be an effective way to ensure the accuracy.