[aprssig] High density timeslotting

[scott at opentrac.org]

> channel.  We are likely going to stop digipeating in channel 
> next year 
> and have that handled by a 9600 baud UHF network.  But to be able to 
> squeeze more trackers into a smaller slot is something that would be 
> valuable to us.

Considering how many KPC-3's and things you've got, it seems like it might
be easier in your case to use two or three channels for trackers, and
aggregate those onto the 9600 baud UHF network.

But yeah, B2V is the sort of thing I'm thinking of.  I've also got users
interested in hang gliders / paragliders / sailplanes and such, and for them
it's even more important to have frequent updates.

I've given a little thought to an on-air coordination protocol that could be
used with RX-capable units like the Tracker2.  You could implement a
DHCP-like protocol where the trackers would get their timeslot assignments
from a central source.  That would require knowing in advance how many slots
are going to be used, though, unless you were to update all of the units as
needed to expand the cycle time.  I suppose you'd want to allocate one time
slot for each coordination transmitter, or maybe keep them on a 600 kHz
split so the trackers would get commands and coordination on one frequency
and transmit on another.

Now if ALL trackers had accurate GPS time references, we could implement
slotted ALOHA and double the channel capacity, more or less.  Better than
that, if the digis transmitted information about what slots they'd heard
traffic (or collisions) in, trackers could spread themselves out within the
net cycle time and have even less probability of collisions.

Scott
N1VG