Subject: Re: Fourier doesn't do nonlinear very well
X-Newsreader: Microsoft Outlook Express 6.00.2800.1106
Date: Thu, 24 Oct 2002 14:01:09 -0500
NNTP-Posting-Date: Thu, 24 Oct 2002 13:37:33 EST
"John Larkin" wrote in message
> On Wed, 23 Oct 2002 21:04:20 -0500, "JD" wrote:
> >The fact is that the world is mostly nonlinear diffeqs, and it is really incompetent
> > (amazingly so) that real world EEs try to always think that Fourier can work
> >in that environment.
> Just pick up your cell phone (even the *connectors* are seriously
> nonlinear devices at a cell site) or fly a 747 to China to see how
> "really incompetent" engineers "always" are.
I was speaking of the subset that feel too awful comfortable in
the Fourier (or frequency) domain. It is a great tool for simplification,
but it is best not to overly simplify. Some people in these discussions
were obviously contemplating their navels, rather than being grounded
One anecdote in my past: an RF engineer at AT&T was confused
about the 1/2 frequency components from an oscillator circuit.
I took a look, and had to explain to him about the parametric
effects in the circuit (which gave him the necessary insight.) He
knew about parametric effects, and had probably even designed
some really fancy parametric amps (or somesuch), but his mind
was stuck 'in the box' of being a technician-head rather than
taking an engineering mentality. It was probably a brain-fart
(that all people have), but still, this being stuck in the Fourier
(or frequency) domain is really incredible.