I've tried according to the paper ''NUMERICAL SIMULATION OF THE ROTATIONALLY SAMPLED WIND SPEED FOR LARGE WIND TURBINES'' by BURLIBASA, CEANGA, my normalized correlation is working correctly according to the paper, but I am not a getting a smooth large peak's as shown in the paper. In the paper it has been told to do the fast Fourier transform of the auto-correlation and I have used 'fft' command in matlab. Can anyone help explain why I am not getting the same graph?