Research Article
Removal of Muscle Artifacts from Single-Channel EEG Based on Ensemble Empirical Mode Decomposition and Multiset Canonical Correlation Analysis
Algorithm 1
The EEMD-MCCA Algorithm.
Input: the single-channel EEG signal with size . | Output: the reconstructed EEG signal after muscle artifact removal. | The First Step: | (1) for do | (2) Add independent identically distributed white noise to the single-channel EEG ; | (3) Apply EMD to the above noisy signal and derive a set of IMFs by (1)–(10), denoted as ; | (4) end for | (5) Obtain an ensemble of IMF sets ’s; | (6) Calculate a set of averaged IMFs as the final decomposition, that is ; | The Second Step: | (7) temporally delayed versions of the matrix are generated according to (12), that is ; | (8) Apply MCCA to the data sets and extract the underlying sources in ; | (9) Set the sources corresponding to muscle artifacts (with low autocorrelation) to zero; | (10) Return the cleaned multichannel signals by passing the source matrix through the mixing matrix ; | (11) Reconstruct the single-channel EEG signal by summing the recovered IMFs in the matrix . |
|