IIRC the Nyquist criteria strictly relates to the bandwidth of the signal,
not the maximum frequency (effectively you can use an alias if you know
enough about it).
Therefore you can sample at "sub Nyquist" rates provided your filters and
signal knowledge are sufficient.
RP



>If it is a fairly narrowband signal, centered on 50kHz, maybe you could
mix
>it down to a lower frequency, say 1kHz, then do the processing, in the
same
>way that a radio mixes the RF down to an IF where the gain/filtering
stages
>are easier. As long as the "IF" is high enough that any side bands are
>preserved you should be OK.

This is an area in which I know enough to be dangerous, so no warranty on
the comments ...

If the signal is band limited, can't you simply subsample at a divisor of
the IF rate, thereby doing the mixing and A/D conversion in one step.


Jack

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics