Sean,
Thank you for the reply. I should go back to my client to see if they
can give me a more specific definition, but the approach I thought of
was as you described, subtracting the difference in times at the same
amplitude. But, one problem is that there could be many places where
the waveform crossed the same amplitude. I actually tried programming
this but I did not find a good way to handle that problem.
Intuitively, one could look at a waveform and a time-shifted waveform
of a similar shape, and say that the sum of all the time differences
between similar points would be the total error. However, this
concept is based on the shape of the waveform and I am not sure how
one would approach programming it.
Allen
