[aprssig] APRS Message Idea

[Robert Bruninga]

The APRS decay algorithim is simple and works for
every packet, whether it is a message, and object
or a position.    New info is transmitted immediately,
then 8 secs later, then 16, then 30, then 1 min, then
2 mins, etc down to eventually once every 30 minutes.

THus new data is rapidly retried and old data takes
up little channel capacity.  There are no user settings
to mess up and no way someone can send things
once a minute for ever.  It works in tactical real-time
immediate events and long term friendly QSO's.  The big 
advantage is in messages, of course, where keyboard 
users are talking in real time.

on a 50% propabilty one hop channel, (on average,
where say all messages take 2 packets) then in APRSdos,
user lines are typoically delivered in  under 16 seconds
per line.  But in a simplistic, retry-1-per-minute aproach, 
it takes two minutes per line...  

Now go over two hops where it might typically take 4 TX
packets for success.  With APRSdos, this will be achieved
in under a minute where as the simplistic method will take
an agonizing 4 minutes per  line ...   Bob

>>> jdw at eng.uah.edu 3/3/05 9:52:05 AM >>>

On Mar 3, 2005, at 8:21 AM, Robert Bruninga wrote:

> It should have been in all code.

I just took a look at the spec (1.0.1 was the latest I found at 
tapr.org) and googled for "aprs message decay algorithm" and found 
nothing that would be useful for me if I were implementing a client.  
If it's not in the spec, you can't blame clients for not implementing 
it.

> But then it decays down to one retry every 30
> minutes until the station eventually comes on the air.

What I was looking for was the maximum number of times a message would

be sent (assuming the message wasn't ACKed).  I assume there is some 
maximum number of times a message will be transmitted, otherwise the 
entire network would have degraded into nothing but messages and 
collisions.

> With APRSdos the message will always eventually get
> delivered.

Does this mean it retries forever?

-Jason
kg4wsv


_______________________________________________
aprssig mailing list
aprssig at lists.tapr.org 
https://lists.tapr.org/cgi-bin/mailman/listinfo/aprssig