Denoising and rhythms extraction of EEG under +Gz acceleration based on wavelet packet transform

Filtering and feature extraction are very important in the analysis and study of EEG signal under +Gz acceleration. In this study, a new filter of different frequency characteristics of EEG signal under +Gz acceleration is constructed and four kinds of rhythms of EEG signal are extracted by using wavelet packet transformation. EEG under different G loads is analyzed and compared, and then EEG dynamic characteristics are studied in order to analyze its advantages and disadvantages. Experimental results show that wavelet packet method can effectively suppress interference bands in EEG, such as EMG, power and so on, and effectively reflect the dynamic characteristics of different rhythms, exhibiting good characteristics. The proposed method is also applicable for analyzing and studying other dynamic biomedical signals.