Framework of Applying Independent Component Analysis After Compressed Sensing for Electroencephalogram Signals

This paper proposes a novel compressed sensing (CS) framework for electroencephalogram (EEG) signals with artifacts. A feature of this framework is the application of an independent component analysis (ICA) to remove the interference of artifacts after CS in a data processing unit. Therefore, we can remove the ICA processing block from the sensing unit. In the framework, we use a random sampling measurement matrix in CS to suppress the Gaussian of the compressed sensing data. Herein, the proposed framework is evaluated using raw EEG signals with a pseudo-model of an eye-blinking artifact. The comparison of normalized mean square error (NMSE) values are shown to quantitatively demonstrate the effectiveness of proposed framework.