i'm using the Hilbert transform block on an ADAU1446 for envelope calculation.
It really has a phase difference of 90° between its outputs, but you cannot call
any of the two output a "zero degrees" output, because it shows a
non-constant group delay respect to input.
Maybe i could solve this using a fir filter implementation of HT, but it would need
too much memory resources.
So, my questions are:
-is it possible to make an all-pass filter to compensate one of the two outputs so it
gets phase-linear in respect to the HT transform block?
-to have an idea of the phase response of one of the two outputs of the HT block,
i used a stimulus/probe block, but it seems not to work (i get no amplitude or phase trace).
I could use an external vector analyzer....but could you provide some details on this
or have i to use hardware instrumentation?