FMR simulation signal sudden drop

[jiawen xu]

Recently, I conducted an FMR simulation for the materials to be used. My testing method was as follows:
1. Add an external magnetic field Bz = 2T, then relax.
2. Add a pulsed magnetic field B_ext = vector(0.001 * sinc(2*pi*(f_max) * (t-t0)), 0, Bz)
Set the maximum frequency to 120e9 Hertz.
Finally, after performing a fast Fourier transform (FFT) simulation on mx, I found that the amplitude of mx was not zero when F = 0 in the low-frequency region. Therefore, I changed to perform FFT on myy and found that the curve did not gradually decrease to 0 in the high-frequency region but suddenly dropped nearby.
Is this type of spectrum reasonable? If not, how should I modify my program? What is the cause of this? What measures should I take to make my spectrum gradually decrease to 0 in the high-frequency region? I hope all the professors can give me guidance. Thank you in advance!

[Антон Луценко]

Hi, if you take a look at the spectrum of the sinc pulse itself, you can see that it is actually rectangular, not gradually decreasing to zero. Your resulting spectrum therefore looks quite reasonable. 

At f=0, you get average value of the data across time. 

[jiawen xu]

Thank you very much for your reply. After modifying the correct damping coefficient, I obtained the appropriate spectrum. I have another question. When I processed the data in the post-processing stage, were there any differences in the handling of mx and my? Which one is more appropriate?