[OZAPRS] Analysis of VK1 APRS traffic.

[Robert Bruninga]

> Hunting on the net it looks like 50% to 65% is a reasonable 
> maximum utilisation rate for a p-persistence CSMA network 
> like our APRS network...

Not meaning to nit-pick, but...

I'm not sure about the details of your network, but APRS is
generally not a CSMA network to the Users.  Say there are 30
users in the vicinity of one digi, and 30 in the vicinity of
each of say 3 surrounding digis.  Of the 120 users sharing the
channel maybe the mobiles (who hve the highest rate packets) can
never hear more than 1 or 2 other stations direct.  Hence their
CSMA performance is on the order of 1% or so.  Even base
stations might only hear 5 or 6 or so direct, hence, their CSMA
performance is only on the order of about 5%, still not any CSMA
to speak about.

Of course, everyone hears their own-digi, but that only protects
themselves from wasting a TX packet, and it does nothing to
prevent eachof them from coliding with any other.

Further, in a large area with lots of surrounding digis (We have
on the order of 50 in the surrounding 200 miles, then 95% or
more of all traffic relayed by any one digi is from a long way
away, so again, even if a local could hear EVERYONE in his
surropunding area and do perfect CSMA to avoid a collision with
all of his other locals, then again, he would only avoid about
5% of the total potential for collisioins.

So APRS is more like a NON-CSMA ALOHA network.  Hence, the
throughput is about 18%, which is the theoretical for an ALOHA
channel.

Thus, the most important control over the QRM and reliability of
the channel to local users is to -prevent- incomming DX packets
from far out of area.  That is the major incentive of the New-N
paradigm.

> then our channel utilisation averages 39% which would
certainly make  
> it hard for mobile stations to get into Mt Ginini from outside
the  
> ACT (ie weaker signals) especially as both Peter, VK1NPW and I
are  
> quite strong into Mt Ginini.

Again, I am completely blind to your area, so ignore this next
comment if it does not apply.  But if I was in an area with a
very tall mountain top digi that covered a well populated area
very well, and yet that digi was seeing too much from out of
area (it had constant QRM on its input that none of the local
users can hear), one thing to drastically give priority to local
users is to put an input on a LOCAL frequency that digipeats to
the MAIN frequency.  This way, the local users get priority
every time.

Besides, what is a local digi for, except to serve the local
users.  And in some areas, we find that only about 5% or less of
the traffic carried by that digi is ORIGINATED locally to that
digi.  This means even if those 5% local users get *priority* at
the expense of outside DX, it really does not affect the overall
throughptut but by 5%, yet gives locals 100% priority with only
5% of the competition that had before.

Good luck
Bob WB4APR
 > 
> Comments?
> 
> 
> Carl,
> vk1kcm.
> 
> _______________________________________________
> Ozaprs mailing list
> Ozaprs at aprs.net.au
> http://aprs.net.au/mailman/listinfo/ozaprs
> 
> _______________________________________________
> Ozaprs mailing list
> Ozaprs at aprs.net.au
> http://aprs.net.au/mailman/listinfo/ozaprs
> 

_______________________________________________
Ozaprs mailing list
Ozaprs at aprs.net.au
http://aprs.net.au/mailman/listinfo/ozaprs
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://second.aprs.net.au/pipermail/ozaprs/attachments/20070216/2e8d84e5/attachment.htm