【摘要】 脑电信号和眼电信号存在频谱混叠，目前的单通道脑电信号中眼电伪迹去除方法容易造成脑电信号失真。提出一种基于经验小波变换(EWT)和改进的自适应噪声完备经验模态分解(ICEEMDAN)的单通道脑电信号眼电伪迹去除算法。首先使用EWT将单通道脑电信号分解为δ频段和高频段信号，再用ICEEMDAN将δ频段信号自适应分解为多维本征模态函数(IMFs),设置样本熵阈值自动去除眼电伪迹信号，最后重构得到滤波后的脑电信号。基于半模拟脑电数据和真实脑电数据开展实验，结果表明所提算法相比于已有算法能够在去除眼电伪迹的同时更好地保留原始脑电信息。更多还原

【Abstract】 There is spectrum aliasing between electroencephalogram(EEG)and electrooculogram(EOG)signals, and existing EOG artifacts removing methods for single-channel EEG signals lead to the loss of useful EEG information. A novel EOG artifacts removal method for single-channel EEG signals is proposed based on empirical wavelet transform(EWT)and improved complete ensemble empirical mode decomposition with adaptive noise(ICEEMDAN). Firstly, the EOG artifacts that contaminate single-channel EEG signals are decomposed into δ rhythm and high rhythm sub-signals by EWT,and then the δ rhythm sub-signal is adaptively decomposed into multi-dimensional intrinsic mode functions(IMFs)by ICEEMDAN. Finally, the sample entropy threshold is set, IMFs with EOG artifacts are automatically removed, and the EEG signals are reconstructed. Experiments are performed on semi-simulated EEG data and actual EEG data. The results demonstrate that, compared with existing methods, the proposed method can effectively remove the EOG artifacts and retain the original EEG information with better performance.更多还原