Algorithm 4 Typical principal component analysis

Input: EEG data $D=\{x_1,x_2,...,x_n\}$, low-dimensional space dimension d. Output: Projection matrix $W^{*}=(w_1,w_2,...,w_d)$.   procedure PCA(D):      Sample centering $x_i\leftarrow x_i-\frac{1}{m}{\sum }_{i=1}^m{x}_i$.      Calculate $X{X}^T$.      Eigenvalue decomposition for $X{X}^T$.      Select the largest d eigenvalues.      $W^{*}=(w_1,w_2,...,w_d)$.      New EEG data $D^{*}=W^{*T}D$.      Return: $D^{*}$.