William Sommerwerck wrote:

Yes, I'm aware of the mathematical definition of minimum
phase but haven't seen where it defines what is minimized
in a mathematical so as to give it the name. All roots being
inside the unit circle may minimize the radius of the locus
enclosing them I don't think that is what minimum refers to.


Minimum simply means that any other network with the same frequency response
would have the same or greater phase shift. That's all.


And at all frequencies, eh? I guess that's simple enough.
I'm ashamed it's the first time I've seen it stated so
simply. Maybe because it's too obvious to state and I was
looking instead for something too subtle. :-)

Now I just have to satisfy myself that reflecting any zero
from inside to outside the unit circle (which preserves the
magnitude response) will cause only an increase in the
amount of phase shift at any point around the unit circle.
Not at all clear at this point.

long pause

Well, I did that via the kind tutelege of the folks in
comp.dsp, which is why this post sat so long in my
"drafts" folder. :-)


Thanks,

Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein