Issue with CST's AR filter

As mentioned above, I have a problem with the AR filter of CST MWS. I have read several manuals concerning the post-processing for resonant structures. But I'm not quite sure, what value to insert to the "First timestep" field.
Is it the value where the excitation signal has vanished OR where the transmitted signal has its maximum?
Thanks,
thestranger268

The AR filter is a "prediction" tool. Perhaps the first time step should be at a time where you can see a definite decaying trend in your "reflected/transmitted" signal.

Thank you. By putting the first time step to the output signal's maximum the solver predicts the signal decay quite fast. Sorry, if this fact was obvious but the CST manual clearly says that the first time step is the value where the excitation (!) signal has vanished...
But thanks for the quick reply =)