|
|
Re: Iterative Match Filterings
Posted:
Apr 6, 2012 3:13 PM
|
|
> >> What you've been describing isn't a
> >> matched filtering process,
> >It works just fine on Excel, the recovery of the original signal
> >waveform as well as the method outlined in the OP to recover the
> >original amplitude. So unless there is some other filter out there
> >that does what I'm doing, it's time to claim this puppy in a method
> >patent.
> Knock yourself out, but if you call it a matched filter or describe it
> as a matched filtering process you'll just be confusing people.
Humor.
I'd love to flatter myself into thinking I discovered a new signal
processing filter. After all, I've been saying all along that there
is low hanging fruit all over the place in engineering applications.
But the reality is fundamentally new filters from a mathematics POV
are very rare. They figured out everything related to convolutions in
signal processing decades ago.
> >EXTRA CREDIT:
>
> >If you multiply the FFTs of a noisy signal and its kernel on the SPICE
> >electronics simulator -- this is faster and easier than Excel -- what
> >would this represent?
>
> >a. a match filter
>
> >b. a wiener filter
>
> >c. a kalman filter
>
> >Bret Cahill
>
> As so many others have said, you have a problem communicating.
Can you think of more than one interpretation of what was posted
above?
Bret Cahill
|
|
