EEG artifact suppression based on SOBI based ICA using wavelet thresholding

EEG is sensitive to certain irrelevant sources as well as artifacts. Various studies have been carried out to detect and denoise the EEG artifacts while retaining the useful information from the original signal but still more improved analysis is required for suppression of artifacts in corrupted EEG data. This paper utilizes a blind source separation algorithm based on the second-order blind identification (SOBI) and then wavelet denoising with soft thresholding is carried out. Experimental observations show that the present method yields better suppression of artifacts in terms of the evaluation parameters of RMSE (Root Mean Square Error) and PSNR (Peak signal to noise ratio).