[aprssig] WRAP-UP MISSOULA MARATHON APRS SUPPORT

Bob Bruninga bruninga at usna.edu
Tue Jul 10 16:02:42 EDT 2012


But I was not talking about overloading the channel for others.  I was
talking specifically about overloading a private event with no other
traffic, and colliding packets with your own participants.  At 7 trackers
and 30 second intervals, and 2 digipeaters, that is 42 packets per minute.
With each packet taking about a second, it is impossible for all of those
packets to be received reliably.

The casual observer will get the feeling that everyone is getting fresh
data, but it is the collided packets from the weaker or more distant tracker
that are not being heard reliably because most of its packets are being
clobbered by the closer trackers.  The "gaps" in coverage may not be a range
issue at all, its just that those packets are weaker and are being clobbered
by stronger packets closer in.

The ALOHA channel has been well  understood since the 60's when it has been
proven that any loading over 18% of channel busy only makes throughput
worse.  Said another way, unless you hear nearly 80% quiet between packets,
then you can be assured that you are not hearing everything, and that many
of the packets are suffering collisions.

For 7 trackers and 2 digipeaters, there is no way I would use 30 second
reporting if I wanted reliability.  Generally, I consider a channel (even if
it is private and exclusive) to not be able to support more than 6
one-minute trackers with only 1 hop.  12 trackers would have to do 2 minute
rate to be reliable.  Adding two hops will guarantee to make things worse.

Now of course, IF your event used no digipeaters and all stations could hear
everyone else so that carrier detect could work, then you could double these
figures.

Etc..

Bob, Wb4APR

-----Original Message-----
From: aprssig-bounces at tapr.org [mailto:aprssig-bounces at tapr.org] On Behalf
Of Herb Gerhardt
Sent: Tuesday, July 10, 2012 2:36 PM
To: 'TAPR APRS Mailing List'
Subject: Re: [aprssig] WRAP-UP MISSOULA MARATHON APRS SUPPORT

You folks must think outside the box.  Every place is a unique location.
Right now within 20 miles of Missoula  there are a total of 8 APRS stations
on my map and only two digipeaters, so transmitting at 30 sec is fine and
dandy FOR THIS LOCATION, it won't saturate the APRS network.

Our recommended path settings are for the entire US and local paths which
transmit more often are just fine in some areas.  These areas not only have
few users, they have very few digipeaters or home stations that will digi
the WIDE1-1 path, so only a few transmissions will hit a digi.  More often
might be better in some remote areas of our country.......

Also keep in mind that special events and Search and Rescue missions have
priority over normal local usage.  These are usually of a short duration and
for SAR, a life might be at stake, so more often transmissions to make sure
you are being heard, are justified.


Herb, KB7UVC
NW APRS Group, West Sound Coordinator
Our WEB Site:  http://www.nwaprs.info



> -----Original Message-----
> From: aprssig-bounces at tapr.org [mailto:aprssig-bounces at tapr.org] On Behalf
Of
> Brett Friermood
> Sent: Tuesday, July 10, 2012 10:01 AM
> To: TAPR APRS Mailing List
> Subject: Re: [aprssig] WRAP-UP MISSOULA MARATHON APRS SUPPORT
>
> > I count 7 trackers going at 30 second rate which is really pushing the
> > Aloha limit on a 1200 baud channel with digipeats. You may have lost a
> > lot of packets due to collisions...
>
> What was the reasoning for 30 second transmit rates?
>
> At a far off world record pace of 15 mph that is 1/4 mile traveled per
minute, which
> equates to the many times on this list recommended minimum beacon rate.
Ded
> reckoning will easily handle that even on an extremely curvy course.
>
> _______________________________________________
> aprssig mailing list
> aprssig at tapr.org
> https://www.tapr.org/cgi-bin/mailman/listinfo/aprssig


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





More information about the aprssig mailing list