Adaptive Principal Component Analysis Based Recursive Least Squares for Artifact Removal of EEG Signals

摘要

Artifacts or noise sources increase the difficulty in analyzing the EEG and to obtaining neural activity. In this paper, an adaptive principal component analysis based recursive least squares algorithm is proposed to remove the artifacts. The algorithm is designed to adaptively derive a relatively small number of decorrelated linear combinations of a set of random zero-mean variables while retaining as much of the information from the original variables as possible. The proposed method was tested in real EEG records acquired from seven subjects. In our experimental study, we show that our proposed method can effectively enhance the spike for all subjects. It is concluded that the proposed method reduces the common artifacts present in EEG signals without removing significant information embedded in these records.