Algorithm 5 CCA based denoising of EEG signal [72]

Input: X: EEG data matrix Input: Y: matrix of auxiliary variables (e.g., EOG or ECG data) Output: Z: matrix of cleaned EEG data Center and whiten the X. Initialize the weight matrices A and B randomly. repeat     Compute the canonical weights $w_a$ by maximizing the correlation between X and Y with respect to A. ▹$w_a$: weights used to linearly combine the EEG signal for one component     Compute the canonical weights $w_b$ by maximizing the correlation between X and Y with respect to B.▹$w_b$:weights used to linearly combine the auxiliary signal for one component     Update the weight matrices A and B. until convergence Compute the cleaned data as $Z=A^{T}X$. return Z