On Mar 18, 2005, at 10:17 AM, rwmcgwier@comcast.net wrote:
> The dynamic range of the audio coming out of the modern
> transceiver from
> ANY port, headphone, speaker, line level output, is less than
> that afforded you
> by an 8 bit audio card.
I will try to find time to record the audio amplitude variations
across the band in some future contest and also study how many bits
an RTTY signal (plus noise) needs to be above the LSB of an A/D
converter for it to be "close enough" to the optimal case.
I will also try to record the strength of the really weak signals
that come across the Pacific when the 20m band is "dead."
On top of the audio variation, hard clipping a signal before passing
it through a matched filter is obviously not optimal, so you will
need a few more bits than just letting a weak signal tickle the least
significant bit of the A/D. The only analyses I have seen are based
on the AWGN case, so it is worth looking at more real world cases.
The case of quantized FSK can be readily simulated with the sound
files from VE3NEA (especially Alex' multipath and selective fading
files) and are probably worth doing, before going to actual recorded
signals off the band.
BTW, here is another vote for the SDR-1000.
With the SDR-1000, even good two channel 24 bits A/D converter do not
have enough I and Q channel balance across the passband to deliver
the kinds of dynamic range that the SDR-1000 is capable of, but the
"radio" does have a pulse generator in the front end to give you the
capability of equalizing the A/D converters. Macintosh users will be
happy to learn that there will soon be an alternative to using a
parallel port to control the SDR-1000, which means you can run it on
any modern Macs. I don't think that it is listed at Flex-Radio's web
page yet but I have been playing with a USB (as in Universal Serial
Bus and not upper sideband) adapter from them (you do have to write
code yourself to download the adapter's firmware -- but it is just
the standard Anchor chip set).
73
Chen, W7AY
_______________________________________________
RTTY mailing list
RTTY@contesting.com
http://lists.contesting.com/mailman/listinfo/rtty
|