[aprssig] THOUGHTS ON RELIEVING APRS CONGESTION ON 144.39

[Tom Hayward]

On Wed, Nov 17, 2010 at 08:11, Wes Johnston, AI4PX <wes at ai4px.com> wrote:
> Not to be a wet blanket, but the tmd700 does not allow for setting txdelay,
> and it's horrendous 500ms delay makes 9600 baud useless for trying to
> conserve bandwidth.

Yes, 9600 baud isn't much faster than 1200 baud for total airtime, but
by using an altnet instead of just a single frequency, the bandwidth
doubles. It's about channels, not baudrate.

> Why use an alt freq that is so close to 144.39?  If you hook up two TNCS to
> one radio on two different frequencies, one goes deaf when the other TX's.
> What good did that do you?  I'm in the middle of TXing a 9k6 packet and the
> 1200baud side transmits and the 9k6 radio doesn't hear the end of my
> packet.  I just don't see where this bought you anything.

You're thinking as a single digipeater, not a network. We have noticed
in western Washington that if one digipeater happens to be deaf when
you're packet comes through, it's usually digipeater by another
digipeater. APRS is designed to be redundant like this, and it works.

> Why not shoot for the alt input frequency of 144.99?

Adding 144.35 to a 144.39 system is cheap. The antenna and the notch
filter do not need retuning.

> Another idea that was batted around a few years ago was to bundle packets at
> the digipeater site.

I'm in the process of writing some software that will use this idea
for cross-gating. Every 10 seconds or so, packets will be cross-gated
between networks, with no path, in one big bundle (there may be other
limits too, we'll experiment). This will allow users of the altnet to
see users of 144.39 and vise versa without adding too much congestion
back into the network (that's the goal anyway, again, we'll
experiment).

Tom KD7LXL