[aprssig] High density timeslotting

[Robert Bruninga]

>>> scott at opentrac.org 06/08/06 11:14 PM >>>
>Right now, timeslotting... The practical limit would seem to 
>be about 2-3 seconds separation between trackers.
>...Digipeating would present an interesting challenge.

Remember the concept we discussed back when you first joined the
SIG about having digipeaters bundle packets for objects so that
we could amortize the TXdelays?   This is kind of separate from
your original question above, but one of the problems with
bundling is that the sender, expects an immediate digipeat or
sometimes he will send again and again until he hears his packet.

How about this wrinkle?  The  DIGI's only bundle dumb-trackers.
Those are not listening for their return packet, so they do not
care if they are bundled or not.   The digi, simply gathers up
all tracker packets and then once every 30 seconds (or whenever
it gets seven) which ever comes first, then sends them out in
one packet.  For short Mic-E packets where the TXD is fully
half the length of the packet, this could double throughput
for events with a large number of trackers...

Delay should not be more than 30 seconds though, since then
it starts impacting the latency standards in APRS.

de Wb4APR, Bob