>I wondered how far I could get with this, as I remember Tony
>Nixon's ID badge problem. I think Peter Peres said that any
>radio used would go through walls floors and ceilings, hence
>my question as to whether you really can restrict RF to limits
>at low cost and low fuss

For such a short distance (20ft) the best range limiting is to attenuate
the signal as much as you need and use desensitized reecivers. You will
have to use some form of signal quality detection system in the receiver.
See my other posting on this in this thread. You could detect S/N by
comparing the shape ('cleanness') of the received pulses in that system,
in the time domain. I'd like to remind that a simple technique to decide
whether a digital signal is 1 or 0 with noise over it, over a period of
time, is to sample it at a constant fast rate and use two accumulators for
0 and 1. Then (or rather meanwhile) you normalize them. The result is the
average signal level and a S/N number that can be used as a threshhold for
squelch (the S/N number can be obtained by incrementing a different set
of accumulators for too short and too long pulses in the input).

With the ILS style modulation I described in the other posting you should
end up with 25% 1's on one side 50% 1's on the other and 75% in the
middle. Beware reflections etc at short range from people and car bodies
and fences and garbage cans and posts and whatever else Murphy decides to
throw at you. A large van with metal cab parked in the wrong place is a
classic showbuster in my experience.

good luck,

Peter

--
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