On Fri, 24 May 2002, Lawrence Lile wrote:

>I suppose that I could get myself a "known good " unit (we do have some),
>and play around with a short loop and my attenuator until I got the known
>good unit to pick up the signal, and then set that as a standard.  Not very
>scientific, though...

The BER is affected by S/N before the slicer. You can use a scope to see
this and avoid waiting. This is very 'scientific' assuming you integrate
the signal over the desired period. In the end, it is the BER on the
recovered data signal that matters, everything else is academic
hair-splitting. Since there is a defined relation between BER and SINAD
for a fixed bandwidth and detection method the attenuator setting for
reading 50% BER will tell you everything you need to know about the noise
margin of your receivers for your given receiving conditions. Hint:
raising the properly oriented antenna on a boom (tuned ferrite stick)
several meters above ground works wonders.

$0.02

Peter

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.