(1) Choose , the number of principal axes or eigenvectors required to estimate. Consider matrix and set
    .
(2) Initialize eigenvector of size , for example, randomly;
(3) Update as ;
(4) Do the Gram-Schmidt orthogonalization process ;
(5) Normalize by dividing it by its norm: .
(6) If has not converged, go back to step 3.
(7) Increment counter and go to step 2 until equals .
Algorithm 2: Fixed-point