An Estimation Method for Multi-channel EEG Data Based on Canonical Correlation Analysis

Electroencephalogram (EEG) signal is often contaminated by electronic noise as well as movement artifacts. This paper presented an algorithm based on Canonical correlation analysis (CCA) to estimate multichannel EEG data. Different from previous studies, in which CCA was mainly used to detect the invariant features specific to each brain state, in this paper, the canonical variates computed by CCA were used to reconstruct the multi-channel EEG data. Firstly, two data sets, EEG signals and the reference signals based on prior knowledge were constructed. Next, canonical variates were computed by projecting the two data sets onto basis vectors. Finally, a least squares solution was used to estimate the multichannel EEG data. The experiment results suggested that the algorithm is capable of reconstructing the actual specific components with high quality. We also hint future possible application of the algorithm in the estimation of functional connectivity patterns at the end of the paper.