Exponential window correction factors?


Are there correction factors for exponential windows?  I searched the forum and knowledge base and did not see it.


I assume that there is no need for an amplitude correction factor for exponential windowing, since the window is mostly affecting the (low amplitude) end part of our signal, and not the high energy/high amplitude part. But I did not find information to confirm this. Also, if indeed a correction factor is applied and I am making the wrong assumption?


Re: Exponential window correction factors?

Siemens Phenom Siemens Phenom
Siemens Phenom

For sure the exponential window distorts the amplitude and energy.


Generally speaking, because a FRF is the desired measurement outcome, the correction factors are not as important if the same window is used on both the input and output signal (see picture below):




Notice that the force window is not flat at the beginning.  Instead it has the same shape and decay as the exponential window used on the response acceleration signal.  This way, both the force input and accelerometer output amplitude are affected identically.


If a flat force window was used, the resulting FRF would have the wrong amplitude. Calibration factors or window corrections would have to be adjusted to reflect that the response is being lowered in amplitude while the force is not.