you say: " "All" is hard to do, but you can record big pieces of the
spectrum."
So, theoretically, with the right equipment it would be possible? I.E.
the sampling rate but that could be fixed with a mega-fast computer that
could sample at immense frequencies? Or does a really high sample rate
still not produce the dynamic range necessary to capture all of the RF
spectrum? If not, what part of the recording process IS responsible for
the dynamic range?
If it is possible, and also assuming enough storage-space is available
(and the rate hard-drive capacity is increasing it's not difficult to
imagine), we could have amazing video/radio recorders - no need to set
the channel, just press record! awesome!
Or is the sheer amount of information way too much for this to ever be
likely? Surely not, RF 'capacity' is a fixed, finite number, whereas HD
capacity is increasing in a non-linear fashion. Or is it more
fundamental than that?
I'm really interested in the project as a whole though, and $550 rollers
doesn't sound a bad price for the necessary equipment. How accessible is
the project? I mean, I develop software for a living but not mega
low-level stuff. Is it a requirement to understand fully electronics?
Sorry for going on too much.
Thanks for your excellent reply,
Mat
On Tue, 2006-08-29 at 16:08 +0000, Eric Blossom wrote:
> > On Tue, Aug 29, 2006 at 01:14:03PM +0100, mat holton wrote:
> > > Hi there,
> > >
> > > Perhaps you could resolve a discussion we were having in the office
> > > today. I have always wondered whether it would be possible to record in
> > > some form (analogue or digital - analogue would seem to make more sense
> > > to me) ALL the data that my radio-antenna receives and then play it back
> > > into a radio receiver. Effectively recording every single station
> > > simultaneously?
> >
> > "All" is hard to do, but you can record big pieces of the spectrum.
> >
> > There are two limitations involved:
> >
> > 1) the sampling rate of your system (must be > 2 x the bandwidth of interest)
> > 2) the dynamic range of all of the signals that you are trying
> > to record. You need enough bits in your ADC to represent the
> > weakest signal and the strongest signal at the same time.
> >
> > > Is this not what the GNUradio project is sort of doing?
> >
> > We sometimes record pieces of the spectrum, but generally we try to
> > deal with the data in real time.
> >
> > > Oh, and as a result of searching for this answer, I've become aware of
> > > this pretty cool project, so...is the only way to use the software to
> > > purchase a $1300 board? I know about using a sample instead though. What
> > > makes this board so expensive? I mean, you can buy a digital radio for
> > > $50 (in uk i've seen them for 15 quid) can't you?
> >
> > These days our preferred h/w is the "Universal Software Radio
> > Peripheral" (USRP). See http://www.ettus.com/
> >
> > > So, can you answer my 'simple' question?
> >
> > Hope this helped.
> >
> > If you've got an interest in this kind of stuff, a good place to start
> > would be to subscribe to the discuss-gnuradio mailing list.
> > http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
> >
> > Eric
> >
> >