[aprssig] Fixing older KPC-3 Duping Digis!
Bob Bruninga
bruninga at usna.edu
Wed Aug 31 13:55:03 EDT 2011
> Yes, it would.
> What format and means do you suggest?
This is the best news since 2004 New-N Paradigm!
* In 2004 we standardized APRS on WIDEn-N (including W1-1,W2-1)
* In 2005 We standardized Digi localinfo Beacons to include
- Digi info format (W2, SSn-n.) (or W3, etc)
- Local Freq Objects for Voice, IRLP and Echolink, etc
* Now, in 2011 we solve the old KPC-3 problem!
Instead of the "W2, SSn-N..." format for the digibeacon, we now suggest for
these old KPC-3's to use a new format "W2-2 SSn-N...". Where the "W2-2" is
the Key to mean "Do not use W1-1 in this area. Use only W2-2 (or W3-3 if
ok). If every old KPC-3 with the dupe problem used this standard beacon,
then we can now educate people that when they see that specific "Wx-X..."
nomenclature that it means to only use that path because that digi is a
duping KPC-3. SO basically we are changing the "W2" to a "W2-2" as a subtle
but standardized way to convey this info to users (or W3-3 in some areas).
Remember, under the 2005 http://aprs.org/localinfo.html recommendations,
what comes out of a digi to its immediate users are:
1) The digi beacon every 10 minutes DIRECT
2) The digi beacon every 60 minutes via 2 hops *(see below)
3) Local FREQ object(a) every 10m direct
4) Local FREQ object(b) every 10m direct
5) Local Freq object(c) every 10m direct
Problem is that my guess is fewer than 1 in 3 digis ever even upgraded to
the 2005 Local-Info initiative (beacons 3,4,5 above). But this is a good
time to do all of that.
Would that work?
The goal everywhere is that if your radio has been on for 10 minutes, your
APRS radio should capture all the local info of value, the DIGI, Voice,
Echolink and IRLP or other info.
NOTE: I think there should be two options for the DX beacon (#2 above). As
shown, it is for large sparse networks like New Mexico or something where
you want to see the overall network of digis (few people out there). But
for dense areas like back-east, I think making the DX beacon every 30
minutes is no more total packets but then at least every 30 minutes lets you
see the next digi beyond the one you are currently in.
Back east, waiting an hour for that info is too long to wait.
Bob, Wb4APR
-----Original Message-----
From: Tom Russo [mailto:russo at bogodyn.org]
Sent: Wednesday, August 31, 2011 1:17 PM
To: Bob Bruninga
Cc: 'TAPR APRS Mailing List'
Subject: Re: [aprssig] APRS talk in Los Alamos?
On Wed, Aug 31, 2011 at 01:10:21PM -0400, we recorded a bogon-computron
collision of the <bruninga at usna.edu> flavor, containing:
> Tom,
>
> EXCELLENT explanation!
>
> Now would this problem go away if people used the WIDE2-2 path in that
area
> instead of WIDE1-1,WIDE2-1?? I think so, because then the DUPE filter
would
> work on the 2nd hop?
Yes, it would.
> If this fixes it, then our goal would be to get ALL such OLD KPC-3's to
have
> a BEACON that says something like: "Use W2-2." Especiall in an area whre
> most of the TNC's have this problem?
The recommendation to use W2-2 is in fact on the web page of the club that
maintains the digis (www.urfmsi.org). I agree that it would be a good idea
for those digis to have some kind of standard beacon that would be
easily recognized by users. What format and means do you suggest? I could
possibly pass it on to the fellow who maintains the digis. It might be some
time, though, before it gets implemented.
> -----Original Message-----
> From: aprssig-bounces at tapr.org [mailto:aprssig-bounces at tapr.org] On Behalf
> Of Tom Russo
> Sent: Tuesday, August 30, 2011 8:18 PM
> To: TAPR APRS Mailing List
> Subject: Re: [aprssig] APRS talk in Los Alamos?
>
> On Tue, Aug 30, 2011 at 05:05:25PM -0600, we recorded a bogon-computron
> collision of the <ws5n at wildblue.net> flavor, containing:
> > <html><head><meta name="Generator" content="PSI HTML/CSS Generator"/>
>
> [you might want to adjust your mail program to send only plain text email
to
> APRSSIG --- those of us using old-fashioned text mail programs get to see
> all the HTML markup, which I've hand-edited out in my reply]
>
> >While there I hope you can monitor the APRS traffic around the city.
> When I drive in to Albuquerque, I start hearing "MY POS" repeats
> sometimes three or four times each time I beacon. I have heard it as
> much as 7 times! I am running WIDE1-1,WIDE2-1 in the mobile. I
> have my suspicions as to what is happening, but I don't have the expertise
> to analyze the situation and make recommendations. I'd love to hear
> your take on it, if you see the same behavior.
>
> That's because all the nearby digis are ancient KPC-3 TNCs with old
> firmware,
> with broken handling of WIDEn-N: they fail to mark the final hop as used,
> so if WIDEn-N were implemented normally the path you are using would only
> be a one-hop path:
>
> first transmit: WIDE1-1,WIDE2-1
> First digipeat: LAMOSC*,WIDE1,WIDE2-1
> First digipeat(2): SANDIA*,WIDE1,WIDE2-1
> First digipeat(3): CAPILA*,WIDE1,WIDE2-1
> First digipeat(4): ELKMTN*,WIDE1,WIDE2-1
> (etc. for every digi that happens to hear you)
> End of line, because nobody will digipeat this (WIDE1 is showing as the
> first unused digi, and nobody will do anything with that).
>
> SO, the work-around is that all of these digis have a workaround in place:
> MYALIAS is set to WIDE1-1, with all the problems that go with it ---
> notably,
> no dupe checking on the first hop. So you get THIS mess instead:
>
> first transmit: WIDE1-1,WIDE2-1
> First hop: LAMOSC*,WIDE2-1
> First hop(2): SANDIA*,WIDE2-1
> First hop(3): CAPILA*,WIDE2-1
> First hop(4): ELKMTN*,,WIDE2-1
> and then
> Second hop: LAMOSC*,SANDIA*,WIDE2
> Second hop: LAMOSC*,CAPILA*,WIDE2
> Second hop: LAMOSC*,ELKMTN*,WIDE2
> Second hop: SANDIA*,LAMOSC*,WIDE2
> Second hop: SANDIA*,CAPILA*,WIDE2
> Second hop: SANDIA*,ELKMTN*,WIDE2
> and so on and so on.
>
>
> That is, if you use WIDE1-1 you're no better than the Bad Old Days of
> RELAY,WIDE, thanks to outdated hardware with outdated firmware.
>
> The KPC-3's are all donated hardware from WA5IHL who used to maintain a
> large
> number of normal packet nodes with them, and repurposed them in the early
> days of APRS. It is not reasonable to expect that they'll all be replaced
> in any sort of short order with more modern hardware/firmware, so we're
> stuck
> with the local advice NOT to use WIDE1-1,WIDE2-1 in the central New Mexico
> area --- just use WIDE2-2 so that all dupe checking is actually done
> properly.
> With WIDE2-2, no ping-ponging is possible, you get fewer "MY POS" beeps
> on your D7X0, and the channel gets less congestion.
--
Tom Russo KM5VY SAR502 DM64ux http://www.swcp.com/~russo/
Tijeras, NM QRPL#1592 K2#398 SOC#236
http://kevan.org/brain.cgi?DDTNM
"One man alone can be pretty dumb sometimes, but for real bona fide
stupidity, there ain't nothin' can beat teamwork." - Edward Abbey
More information about the aprssig
mailing list