On Jun 30, 11:15=A0pm, "Steven G. Johnson" <stev...@[EMAIL PROTECTED]
> wrote:
> On Jun 29, 12:59 pm, Ron N <ron.nichol...@[EMAIL PROTECTED]
> wrote:
>
> > On Jun 29, 8:10 am, Rick Lyons <R.Lyons@[EMAIL PROTECTED]
> wrote:
>
> > > Sadly, no one can perform continuous
> > > signal processing on a computer.
>
> > Of course you can. =A0There's lots of symbolic math software.
> > Of course you need a to do this on tractable closed form
> > symbolic functions instead of arbitrary sampled data.
>
> You don't have to do it with symbolic math; there are plenty of
> numerical techniques that can compute such things to any desired
> accuracy (given enough computing power) (although which technique you
> use will depend on what kind of function you are transforming).
> However, there's not enough information about the original poster's
> needs to give sensible advice in this case.
Sounds like you're re-inventing DSP. I understood Rick's comment to
imply that once you represent a continuous signal in a form suitable
for the application of numerical techniques, it's no longer actually a
continuous signal.
|
