[aprssig] RE: Widen-N works well in Denver.

[Robert Bruninga]

Larry,
Yes, You may be quite right.  And I am glad you are looking
at it closely.  I agreee that so many of the stations I saw
were mobiles.  A very large number.  Thus, many of them
are not on the air at the same time.

But conversly, most of them are on the air during commute
time and then the channel will be very much full.  And I agree,
if your HT tracker works most of the time, then that is
very good proof of reliability.  Unless of course, your travels 
are closer to the digi than most everyone else, then it simply
means that you get in because you are stronger.

It depends on what people expect of the network.  If they 
want it to work for mobiles most of the time, then you have to 
eliminate enough of the out-of-area packets to leave room
for all your mobiles at the same time. etc..

Goog luck.
Bob


>>> "Larry Cerney" <lcerney at viawest.net> 12/23/04 8:32 PM >>>
Hi Bob, et al;

Thanks for your interest in the APRS environment here in Colorado.  You
raised some interesting points.  But I think you don't understand the
environment here at all.

With just a first blush at an analysis of what you imply and what I see,
I
think you're wrong.

After reading your email, I did two things using FINDU as you did. 
First I
ran a "Stations Near" report for 240 hours and out of the 50 stations
reported only 32 were RF with the rest of the stations getting to FINDU
via
TPCXX weather stations.  Of the remaining 32 stations, many were just
occasional stations with some stations not having reported for many
days.
FINDU reported at the end of the list "Overall rate: 202.5 packets/hr
3.4
packets/min" which I really don't understand.

I then ran another report on "Stations Near" for the last 1 hour and it
reported over a 120 stations with the farthest being 199.8 miles and
FINDU
reporting "Overall rate: 923.0 packets/hr 15.4 packets/min" which is
about
what I had reported in my first email.  Also included in the >120
stations
is a large number (~40)of WX station that never are heard on RF. 

In the next few days I try to see if what you're saying is true, but I
know
many of the APRS stations along the Front Range here and I don't see all
that much traffic.  While my location isn't on a mountaintop, I am on
top of
a ridge that sees clearly from Pikes Peak in Colorado Springs in the
South
and Longs Peak West of Fort Collins in the North with a clear view out
to
the horizon NE, E, and SE.  I can also hear directly the N0WBW-4 WIDE
27.2
miles NW of Denver, the N0KZ-5 WIDE 57.3 miles South of me and the
WA0BAG-5
WIDE 54.7 miles North of me.  So, I hear all they hear with only one hop
and
when I listen to the 144.39 APRS frequency I only hear the 15 to 20
packets
a minute I reported in the earlier email.

As for the low powered trackers getting in to the network, I run a 2
watt HT
with an OpenTracker, K0ANI-14, 24/7.  Do I always get into the network,
no.
But I bet I get in while mobile >85% of the time.

I'll do some more analysis on the subject, but I just don't see what
you're
saying to be the "Reality" we see here.

73....
Larry
K0ANI



"The only reason for time is so that everything doesn't happen at once."

--Albert Einstein 

Then something is terribly wrong with that software or
something...  Looking at FINDU.com  shows more than 
100 APRS stations (not counting CWwx) within 46 miles 
of your station and ALL within view of only one digi.  
It is impossbile for your network to be working reliably 
if all of those  stations are on the air and using more 
than 1 digi hop. (and everyone that I checked is running 
4 hops or so!)

>There are many holes in the coverage ..., but the WIDEs
> we have here,...on the Front Range of the Rockies, are great.
>The 60 stations that UI-ALOHA used in its calculations included 
>about 10 stations with the farthest station being 515 miles away 

What about the other 90 ?

>We do get QRM from stations outside the area...
>The second program I ran was APRSNetSpy, by  N0KKZ, 
>which analyzes and counts packets heard in the area. 
>It... was seeing only 15 to 20 packets a minute, far below 
>the 60 that would choke an APRS network. 

Wow, Counts per minute are absolutely meaningless in
APRS in analyzing network overload.  A 100% fully overloaded
channel will have 0 per minute.  W e must not allow
packets-per-minute to be used anywhere as an analysis
of network performcnce.  If one wants to count something 
"per minute" then you have to go to the DIGI sites
and count how long the SQUELCH is open per minute

>With the APRS environment such as it is here in Denver,
>Colorado I don't see a need for "Firewalling" around the cities. 
> Maybe some day in the future, but WIDEn-N seems to be 
>working just fine.

Wow, from these simple comments, I see exactly the
opposite conclusion.  It appears to me that the network
in the Denver is suffering severe overload and reliability
for a low power tracker or urgent message must be nearly
impossible except for the strongest stations that can step
on everyone else.

To support just those 100 users in just the 46 mile radius of that
one digi, you should be getting over 1 packet per second!
The fact that you can only count 15 to 20 per minute
means that the other 60% to 75% of all packets transmittted
locally are being jammed and lost due to collisions.

Yes, you may "see" lots of stations on your map from 
over a million square miles in KS, OK, TX, NM, AZ, UT,
 WY, SD, and NE. and you might be "hearing" lots
of good packets, but what you are not seeing is all
that you are missing due to a completely overloaded
and saturated net into which locals with low power
find it impossible to use.

The success of APRS is not the number of stations
heard, but the reliability of a local  user being heard.

I'd be interested in looking into this situtation closer
to see what is really going on in Denver and whether an
HT or low-power tracker can be used.  It sounds to
me like it would be impossible due to total abuse of
WIDEn-N and high digis.

Again, I couild be all wrong.  I am kibitzing from afar,
but the data I see and the data you presented seem
to be widely divergent....

thanks.  Lets see what we can do about this.

Bob