[aprssig] Draft Copy of Thesis on APRS

[Tom Hayward]

On Sat, Dec 13, 2014 at 4:38 PM, Kenneth Finnegan
<kennethfinnegan2007 at gmail.com> wrote:
> I don't follow the argument for why you would want to allow identical
> packets through the network more often than every 30 seconds. What's the
> advantage of allowing repeated packets every 6 seconds on RF?

Sometimes it would be nice to have a very fast beacon rate for fast
moving objects during a special event. It's not a big deal though, so
out of respect for the network I use longer rates.

An interesting compromise would be to have smart digis that only
digipeat one of each packet type per dupe cycle. Currently, the source
+ entire payload is dupe-filtered. My idea would be to dupe filter on
source + type, so only one position, one status, one weather, etc.,
per station would get through each dupe period, even if the position
or weather changed. If you want to get really fancy, the digi could
dynamically modify this dupe period based on ALOHA calculation.

If the digis could filter like this, one guy running a fast rate
wouldn't impact a wide area. And the people near him would have a more
accurate location. Kind of like proportional pathing, but enforced at
a network level. (I already use proportional pathing or short/empty
paths when I want to beacon fast/often to the local area.) If you
consider one of the core principles of APRS that local information is
the most valuable to the user, this seems like a smart way to improve
performance both locally (accuracy) and wide area (scalability).

Sorry for the tangential brainstorm.

Tom KD7LXL