Wiener filter for independent component analysis signals
Wei Hao, Xiuhong Wang · 2017
Electroencephalographic (EEG) recordings are widely used in the analysis of the brain signals, the EEG data are recorded in combination with background activities like noise, and with artifacts of physiological or technical origins. Independent component analysis (ICA) is a method that allows blind separation of sources. This technique includes FastICA, which is used here to identify artifacts from EEG. Even when ICA separates the signals into components, they could still be mixed with noise at certain level. A Wiener Filter can be applied to each artifactual component, and finally cleaner artifacts could be obtained. All of the work will be implemented in Matlab.