OpenBCM V1.07b12 (Linux)

Packet Radio Mailbox

DB0FHN

[JN59NK Nuernberg]

 Login: GUEST





  
ZL3AI  > APRDIG   14.07.04 10:32l 793 Lines 30565 Bytes #999 (0) @ WW
BID : 3587-ZL3AI
Read: GUEST
Subj: TAPR Digest, Jul 12, 1/3
Path: DB0FHN<DB0MRW<DB0ERF<DB0FBB<DB0GOS<DB0EEO<DB0RES<ON0AR<ZL2BAU<ZL2BAU<
      ZL3VML
Sent: 040714/0910Z @:ZL3VML.#80.NZL.OC #:27845 [Chch-NZ] FBB7.00i $:3587-ZL3AI
From: ZL3AI@ZL3VML.#80.NZL.OC
To  : APRDIG@WW

TAPR APRS Special Interest Group Digest for Monday, July 12, 2004.

1. Wanted : PIC code for NMEA bit bashing
2. Re: APRS 144.39 dual channel upgrade
3. Re: APRS 144.39 dual channel upgrade
4. Re: 9612+
5. Re: APRS 144.39 dual channel upgrade
6. Re: APRS 144.39 dual channel upgrade
7. Re: RE 9612+
8. Re: APRS 144.39 dual channel upgrade
9. Re: APRS 144.39 dual channel upgrade
10. Re: APRS 144.39 dual channel upgrade
11. Re: APRS 144.39 dual channel upgrade
12. Re: APRS 144.39 dual channel upgrade
13. Re: APRS 144.39 dual channel upgrade
14. Re: APRS 144.39 dual channel upgrade
15. Re: APRS 144.39 dual channel upgrade
16. Re: 9612+
17. Re: APRS 144.39 dual channel upgrade
18. Re: APRS 144.39 dual channel upgrade
19. Re: APRS 144.39 dual channel upgrade
20. Re: APRS 144.39 dual channel upgrade
21. Re: APRS 144.39 dual channel upgrade
22. Re: APRS 144.39 dual channel upgrade
23. new terminology
24. Re: APRS 144.39 dual channel upgrade
25. Re: APRS 144.39 dual channel upgrade
26. Re: new terminology
27. RE: new terminology
28. does anyone see me??
29. Re: APRS 144.39 dual channel upgrade
30. Re: APRS 144.39 dual channel upgrade
31. RE: new terminology
32. Re: APRS 144.39 dual channel upgrade
33. Re: APRS 144.39 dual channel upgrade
34. Re: APRS 144.39 dual channel upgrade
35. THD7A(G) 2mtr brick gone?
36. Re: APRS 144.39 dual channel upgrade
37. Re: does anyone see me??
38. Re: APRS 144.39 dual channel upgrade
39. Re: APRS 144.39 dual channel upgrade
40. Re: does anyone see me??
41. Re: Super Tiny DIgipeater
42. Re: Wanted : PIC code for NMEA bit bashing
43. Re: does anyone see me??
44. Re: Super Tiny DIgipeater
45. Re: Wanted : PIC code for NMEA bit bashing

----------------------------------------------------------------------

Subject: Wanted : PIC code for NMEA bit bashing
From: "Andrew Rich" <vk4tec@hotmail.com>
Date: Mon, 12 Jul 2004 20:05:12 +1000
X-Message-Number: 1

Is anyone willing to give up some code for decoding NMEA 4800 data stream
using bit bash method ? I am using the MAX232 and USART method for a
GPS display. I would rather keep the chip count down to one.

http://www.tech-software.net/gpsiilcd.htm

Cheer Andy VK4TEC

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Robert Bruninga" <bruninga@usna.edu>
Date: Mon, 12 Jul 2004 09:28:32 -0400
X-Message-Number: 2

Thanks! for the 144.99 suggestion.  I was starting to monitor 145.01 and
researching its use in this area but Ill certainly switch my target over to
144.99.

ON SECOND THOUGHT.  Since the goal of this is for a LOCAL permanent input
frequency for local FIXED use, it should not be shared with surrounding
areas or we end up with DUPES.  SO although I will begin to research 144.99
in our area for some yet-to be-determined use, I think ill not try to use
it for this alt-freq idea, since such use would BLOCK its use by anone else
in the metro area.

Though on the other hand, in non-metro or linear areas where  one digi
would cover the population density without ever having any need for a
second tier of alt frequencies, then maybe 144.99 is the way to go...  Sure
makes it easy to set up...

Bob

Though

>>><wes@johnston.net> 7/11/04 3:45:17 PM >>>
Yes, I was only disagreeing with the WX conversion... the rest sounded fine
to me.  That's why it sounded like I agreed with most of what you said... I
did!  I must've missed the part about unconverted WX only on the alternate
frequency (it makes perfect sense now that I have "read between the lines"
and figured out that WX station = Home station = Alternate digi input
frequency).  In _that_ case, this sounds great.  Since only one digi will
be listening on the alt input frequency, then there would be no possibility
of doubling up on converted / non converted packets.  Full steam ahead.  My
kpc4 just became more valueable. Of course the kpc9612's with the extra
1200 baud modem header get real interesting really quickly too.

The slick part about this alt frequency is that 144.39 + 600khz'ish gets us
right into the largely unused 144.90 - 145.09 packet sub band.  The simple
(and likely prized alt input frequency will be 144.99 because all the home
stations have to do is to transmit 144.39 with the standard plus offset.  I
guess the gentleman's rule will need to be that no digipeater will tx on
144.99.  And the home stations can specify their local digi's callsign in
their outbound path.  This just gets more and more interesting.

Wes

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Robert Bruninga" <bruninga@usna.edu>
Date: Mon, 12 Jul 2004 09:41:42 -0400
X-Message-Number: 3

>>>"Richard Amirault" 7/11/04 9:24:53 PM >>>
>Disadvantages [of the 144.39 dual channel upgrade]:
>
>Reduced access to digis for mobiles ... 
>(or am I missing something here?)

Must have missed it.  This significantly improves channel access to mobiles
because it removes any contention on 144.39 from all the home and WX
stations who will all input to the digi over on the local input channel.

AND when their packets are bundled and digipeated over onto the 144.39
channel, the digis will use true CSMA so that again, they won't interfere
with mobiles at all.

de WB4APR, Bob

----------------------------------------------------------------------

Subject: Re: 9612+
From: "Michael Huslig" <mhuslig@kantronics.com>
Date: Mon, 12 Jul 2004 08:43:41 -0500
X-Message-Number: 4

We developed both 1200 and 9600 add-in modems for the 9612+ using the header
pins that ran down the center of the board.  But interest was so low that we
dropped them and stopped installing the header pins.  When we had to
redesign for the MX modem chip, we removed the traces for the header pins
altogether.

Mike

----- Original Message ----- 
From: <wes@johnston.net>

>Was it my imagination or did I read somewhere that the kpc9612plus supported
an
>optional extra 1200 baud modem?  I can't find a reference to it on their web
>pages anywhere...
>Wes

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Robert Bruninga" <bruninga@usna.edu>
Date: Mon, 12 Jul 2004 10:10:03 -0400
X-Message-Number: 5

>>><wes@johnston.net> 7/12/04 9:47:40 AM >>>
>Dupes on 144.99 caused by multiple digi's hearing the 
>same home station should be elliminated b/c the users 
>would specify their local digi by callsign instead of
>the generic WIDE.  

Wow, this is true!  Good idea.

>Of course if two digipeaters in nearby towns hear the 
>same station, then that station has caused a hidden 
>transmitter problem to it's neighbor, and [nothing]
>is going help that..... (sigh...)

But since home stations only have to hit their ONE local input channel
digi, then low power from the base stations would help...

>You know, Bob, what is really needed is for the mobiles 
>to move to this alternate input freq.  

Yes, I considered that, but it completely destroys the universality of APRS
as a national asset on 144.39 for anyone at any time.   No, the FIXED
stations are the ones that should move, since they dont move, and always
use the same digi...

But good thoughts.

One DOWN side of all this, however, is that the MAJORITY of all the load on
144.39 is NOT LOCAL, but DX junk.  SO just moving the local FIXED stations
and WX inputs off, will not have quite the big impact as it could.  Buf if
the other areas also do it, at least the TXD amortization remains as an
improvement, since it helps the digis use time more efficiently...

Bob

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "AE5PL Lists" <HamLists@ametx.com>
Date: Mon, 12 Jul 2004 09:12:57 -0500
X-Message-Number: 6

>-----Original Message-----
>From: Robert Bruninga
>
>2) The FIXED station users would TX on 145.01 (only used as
>an example) but still receive on 144.39.  They would not lose

So much for CSMA on the alternate channel if the fixed stations are only
going to monitor the APRS channel.

>SOFTWARE IMPROVEMENT #1.
>
>Then the next level of improvemnt would be for there to be a
>processor that collects all the non-standard APRS WX reports
>and weatherless position and positionless weather, and
>converts them all to the standard COMPLETE WX format before
>dumping them in a bundle.
>Being smart, it can always wait till it has
>5in a bundle making for a nice neat channel burst.

Some of hose fixed stations are the IGates.  If they are only listening
on 144.39, then they are most likely interfering with each other on the
alternate channel.  If fixed stations are monitoring both frequencies
(very ugly, indeed, with the software available), then you will be
introducing the same problem seen on APRS-IS with Mic-E conversion:
multiple packets representing the same data with different dupe check
signatures.  Except this time you are introducing them on RF.

While the ideas are admirable, both have severe network implications
that IMO make them untenable.

73,

Pete Loveall AE5PL
mailto:pete@ae5pl.net

----------------------------------------------------------------------

Subject: Re: RE 9612+
From: "Neville A. Cross" <yn1ncv@arrl.net>
Date: Mon, 12 Jul 2004 09:09:18 -0500
X-Message-Number: 7

Wes,

I had that question two weeks ago. I wrote to Kantronics. Heres is their 
answer:

----- MSG form Kantronics -----------------
Hi Neville,
Kantronics developed the expansion boards for the KPC-9612+ in 1998.  In
the following year, we sold 7 boards.  In 2000, the boards were
discontinued.

With Regards,

Cheryl Seiwald
Kantronics
P - 785.842.7745
F - 785.842.2031
www.kantronics.com

-----------------end of MGS -------------------

I hope this save time and effort to you.

Neville
YN1NCV

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: wes@johnston.net
Date: Mon, 12 Jul 2004 11:12:28 -0400 (EDT)
X-Message-Number: 8

On Mon, 12 Jul 2004 10:10:03 -0400, "Robert Bruninga" wrote:

>One DOWN side of all this, however, is that the MAJORITY
>of all the load on 144.39 is NOT LOCAL, but DX junk.  SO
>just moving the local FIXED stations and WX inputs off,
>will not have quite the big impact as it could.  Buf if the

This is kinda what I was getting at last week with the suggestion that we
simply cripple 144.39 with digined digipeaters.  Digined can strip the
paths out of packets thereby limiting a station's hops to ONE on 144.39.
Then there is no out of town junk DX to contend with.  All the out of town
DX junk would be diverted to some 440 frequency with each of those digi's
having alternate local input frequencies(just as we've been discussing
alternate input frequencies for 144.39 this morning).  Add to my idea you
idea of TXD amortization to make 9k6 more effiencent, we'd have a winner
(abeit an expensive winner).

I think this alternate input freq for locals on 144.39 is a good first step
though... and with the appropriate TNC to TNC wiring, two TNCs can share
one radio and not desense one another ... all without expensive 2m
cavities.

Gosh, I never thought in a zillion years APRS would get so active that I'd
have to refer to distant station's reports as DX junk.... grin.
Wes

ham callsign: kd4rdb
find me: http://wesvan.zapto.org

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "AE5PL Lists" <HamLists@ametx.com>
Date: Mon, 12 Jul 2004 10:37:05 -0500
X-Message-Number: 9

>-----Original Message-----
>From: wes@johnston.net
>
>This is kinda what I was getting at last week with the
>suggestion that we simply cripple 144.39 with digined
>digipeaters.  Digined can strip the paths out of packets
>thereby limiting a station's hops to ONE on 144.39.  Then

I went a simpler route with javAPRSDigi which does not modify the packets
in any way: if the path can go or has gone more than X hops, simply don't
digipeat the packet.  javAPRSDigi is an adjunct to javAPRSSrvr.  My
reasoning was also pretty simple: I won't try to dictate to others what
path to use, I simply won't let my station be used to contribute to the QRM
on the frequency which is caused by long paths.

Just a different way to approach the problem.  The Digi_Ned approach, as I
understand it, will still digi the packet but with the modified path. The
javAPRSDigi approach won't digi the packet at all.  Also note that since
javAPRSDigi is an adjunct to javAPRSSrvr via javAPRSIGate, all the packets
are being gated to APRS-IS at that point as well.

BTW, before anyone says we don't need CSMA on the fixed station alternate
frequency, I am on the fringe of the DFW area and I normally see about 20
stations direct in a 30 minute period.  Most are fixed. Also, I am not
referring to you, Wes with the above statement.  Just taking this
opportunity to throw this statement in.

73,

Pete Loveall AE5PL
mailto:pete@ae5pl.net

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Richard Amirault" <ramirault@erols.com>
Date: Mon, 12 Jul 2004 11:59:26 -0400
X-Message-Number: 10

I don't follow ... under this new proposal ... are the home stations
listening on .39? or are they listening on the new freq.?  Are the mobiles
on .39 or not?  If the home stations are not listening on .39 (and the
mobiles are on .39) then how can mobile access to digi's not be affected?

Richard in Boston, MA, USA
N1JDU

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Robert Bruninga" <bruninga@usna.edu>
Date: Mon, 12 Jul 2004 12:33:32 -0400
X-Message-Number: 11

>>>"AE5PL Lists" <HamLists@ametx.com> 7/12/04 10:12:57 AM >>>
>>2) The FIXED station users would TX on 145.01
>>(for example) but still receive on 144.39.
>
>So much for CSMA on the alternate channel if the fixed 
>stations are only going to monitor the APRS channel.

Agreed, but the number of say 30-minute fixed stations and 10 minute WX
stations for any given alt-input are so small that the probability of a
collision is miniscule... And with regards to the IGates you mention below,
THey also fall under this SMALL number criteria and so they too will see an
IMPROVED performance since they also are getting full CSMA at the digi on
144.39 when their packets are brought over from the alt channel....

>Some of hose fixed stations are the IGates.  If they are 
>only listening on 144.39, then they are most likely interfering 
>with each other on the alternate channel. 

I cannot imagine more than at most two IGates being on the same channel.
Remember, this is a local thing.  If there are more igates than that for
any one digi area, then there are too many igates... which would be another
problem worth fixing... not encouraging...

>If fixed stations are monitoring both frequencies
>(very ugly, indeed, with the software available), then
>you will be introducing the same problem seen on
>APRS-IS with Mic-E conversion:

Agree completely.  There is no intention to monitor both.

Bob, WB4APR

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Robert Bruninga" <bruninga@usna.edu>
Date: Mon, 12 Jul 2004 13:02:35 -0400
X-Message-Number: 12

Concerning the lack of CSMA on the alternate channel:

1) Lets assue currently most digis hear 3 others.  And most users are using
an average 3 hop path.  THis means that any one digi is seeing traffic from
about 10 others or that 90% of everything it hears is NOT local.

2)  Now then, currently any FIXED station that transmits on 144.39 will
have a chance of collision that is 90% not local, or said another way, only
a 10% chance of colliding with another local.

3) But of all the packets on 144.39 on average, less than HALF are from
fixed stations, so comparing traffic for collisions, the probability of
colliding with another FIXED local station is probably less than 5%...

OR ANOTHER WAY:

Now how many LOCAL FIXED stations are there in one's local digi area? Lets
say on the order of 50?.  Lets say 5 are WX at 5 minutes and the other 45
are at 30 mins. That's 75 packets in 30 minutes or 1 packet every 24
seconds.  This also is less than about a 5% probability of collision.

So it seems to me that even without CSMA on the alternate channel, the
advantages of getting a guaranteed CSMA when it is brought over to 144.39
is a WIN-WIN situation.

de Wb4APR, Bob

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Robert Bruninga" <bruninga@usna.edu>
Date: Mon, 12 Jul 2004 13:15:31 -0400
X-Message-Number: 13

>>>"AE5PL Lists" <HamLists@ametx.com> 7/12/04 11:37:05 AM >>>
>BTW, before anyone says we don't need CSMA on the 
>fixed station alternate frequency, I am on the fringe of 
>the DFW area and I normally see about 20 stations 
>direct in a 30 minute period.

Doing the math yields about one packet every 90 seconds.  Seems about like
only a 1.1% probabilty of collision... if done on the alternate input
channel... Compared to the very *high* probability of collision on
144.39...

Remember, we are simply moving the FIXED benign stations off.  In times of
stress, and if a fixed station begins to increase his rate because an
operator is present, then that operator can decide whether to stay on the
alt channel or to move back to 144.39 In any case keeping the non high-rate
stations off of 144.39 will cause a net win for everyone...

Bob

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Robert Bruninga" <bruninga@usna.edu>
Date: Mon, 12 Jul 2004 13:24:10 -0400
X-Message-Number: 14

I don't follow ... under this new proposal ... are the home stations
listening on .39? or are they listening on the new freq.?  Are the mobiles
on .39 or not?  If the home stations are not listening on .39 (and the
mobiles are on .39) then how can mobile access to digi's not be affected?

Richard in Boston, MA, USA N1JDU wrote:
>
>I don't follow ... under this new proposal ... are the home 
>stations listening on .39? or are they listening.... etc???

Everyone listens on 144.39.  
Mobiles continue to TX on 144.39
only fixed and WX stations TX on 145.xxx
but their packets are digipeated over to 144.39 for further routing.

Bob

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "AE5PL Lists" <HamLists@ametx.com>
Date: Mon, 12 Jul 2004 12:31:39 -0500
X-Message-Number: 15

Your math is flawed.  You are assuming fixed stations transmit one
packet every 30 minutes, period.  In fact, most fixed stations I see
produce 2 to 4 packets (not including objects) every 30 minutes (posits,
status, and ID's).

Second, you do not figure IGates into your equation.  There are multiple
IGates (like path settings, we can't force someone to turn off their
IGate) that would be in the range of this mythical alternate channel
digi.  Those IGates will, by definition, be interfering with each other.

Finally, of the 20 stations I mentioned there are 6 weather stations.
In a previous post, you claimed that over half of the packets on 144.39
are from mobile stations.  In fact, in the DFW area, the number is much
less.  As Wes pointed out, the biggest problem is not fixed vs. mobile,
but paths that are excessive for the local environment.

Your proposal is asking everyone with fixed stations in a local area to
transmit "in the blind" on the alternate frequency.  In addition to
being bad network design, this is of questionable legal standing which
has been hashed over many times in the past on this SIG.  As I said
before, your intent is admirable but the implementation is not.

73,

Pete Loveall AE5PL
mailto:pete@ae5pl.net

----------------------------------------------------------------------

Subject: Re: 9612+
From: "Michael Huslig" <mhuslig@kantronics.com>
Date: Mon, 12 Jul 2004 12:52:17 -0500
X-Message-Number: 16

There was also a lot of firmware to handle the third port, which hasn't
tracked with our most recent updates.  My guess is the CEO would say no,
but it never hurts to ask.

Mike

----- Original Message ----- 
From: "Rick Green" <rtg@aapsc.com>
To: "Michael Huslig" <mhuslig@kantronics.com>
Sent: Monday, July 12, 2004 10:31 AM
Subject: Re: [aprssig] Re: 9612+

>On Mon, 12 Jul 2004, Michael Huslig wrote:
>
>>We developed both 1200 and 9600 add-in modems for the 9612+ using the
header
>>pins that ran down the center of the board.  But interest was so low
that we
>>dropped them and stopped installing the header pins.  When we had to
>>redesign for the MX modem chip, we removed the traces for the header pins
>>altogether.
>>
>>Mike
>>
> WHat would it take to induce Kantronics to release the designs to someone
>interested in building their own?  What if TAPR or some other small entity
>(Byonics?, Hivalueradio? Far Circuits?) were interested in funding a small
>run of PCBs and marketing them with a schematic and parts list?
>
>--
>Rick Green
>
>"They that can give up essential liberty to obtain a little
> temporary safety, deserve neither liberty nor safety."
>                 -Benjamin Franklin

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Richard Amirault" <ramirault@erols.com>
Date: Mon, 12 Jul 2004 13:53:07 -0400
X-Message-Number: 17

----- Original Message ----- 
From: "Robert Bruninga"

>Everyone listens on 144.39.
>Mobiles continue to TX on 144.39
>only fixed and WX stations TX on 145.xxx
>    but their packets are digipeated over
>    to 144.39 for further routing.

Sorry if this has been addressed (I don't read every last message on the
list) but ... so does that mean a home station will digipeat mobiles they
hear on .39 over to the new freq? If they don't then mobiles will loose that
very imports first digi (RELAY) into the (main) DIGI ... And if they will be
then won't the mobiles packet get to the DIGI twice on some occasions? ....
And wll the home stations be transmiting "in the blind" on a freq it is not
listening to?

Richard in Boston, MA, USA
N1JDU

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Larry Cerney" <lcerney@viawest.net>
Date: Mon, 12 Jul 2004 12:12:40 -0600
X-Message-Number: 18

Please help me understand this.  My station which is Fixed and WX transmits
on 145.xx in the blind to be received by a node which gathers some 5 beacons
and retransmits them in bulk back onto 144.39.  Is that correct?  If so, I
think I understand this is to reduce the chance of collisions with hidden
stations and to reduce the total number of packets on 144.39 for any given
time.  Is this correct?

The reason I ask is, and here on the front ranges of the Rocky Mountains we
don't seem to have a problem with an inordinate number of collisions, it
seems to me that channel utilization will not decrease that much just by
bundling up Fixed and WX station beacons, removing their headers and
retransmitting them back in bulk. 

Am I missing something here?

Larry
K0ANI

"There is a theory which states that if ever anyone discovers exactly what
the Universe is for and why it is here, it will instantly disappear and be
replaced by something even more bizarre and inexplicable. 
 
There is another theory which states that this has already happened."
 
Douglas Adams (1952-2001)

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Larry Cerney" <lcerney@viawest.net>
Date: Mon, 12 Jul 2004 12:27:17 -0600
X-Message-Number: 19

If the problem is junk DX getting into the local nets, perhaps we should do
a better job of educating APRS users as to the concept of local vs. DX.  I
have asked on a number of occasions users who had path of Wide7-7 and
Wide5-5 if they really needed the send their home locations out to other
APRS networks over a 5 or 6 state area when their positions were available
to the world at the first Igate they reached.  Most stations would then back
off to a more reasonable and local path of Wide or Wide2-2. 

Bob wrote a good primer on APRS and how Digi's work with great tips on how
to set up your station that I share whenever I am asked about setting up a
APRS station.  You can find it here:
http://web.usna.navy.mil/~bruninga/aprs/fix14439.html

Pass it along to those who might need a little insight.

73...
Larry
K0ANI

"The wireless telegraph is not difficult to understand. The ordinary
telegraph is like a very long cat. You pull the tail in New York, and
it meows in Los Angeles. The wireless is the same, only without the
cat." 
-Albert Einstein (1879-1955)

----------------------------------------------------------------------

Subject: Re: APRS 144.39 dual channel upgrade
From: "Robert Bruninga" <bruninga@usna.edu>
Date: Mon, 12 Jul 2004 14:28:30 -0400
X-Message-Number: 20

>>>"AE5PL Lists"  7/12/04 1:31:39 PM >>>
>Your math is flawed.  You are assuming fixed stations 
>transmit one packet every 30 minutes...  In fact, most 
>fixed  stations I see produce 2 to 4 packets every 
>30 minutes  (posits, status, and ID's).

Thanks.   I stand corrected:
1) Two packets every 30 minutes yields a 2% instead of a 1% probability of
    collision on the alt channel.
2) If benign stations are sending out ID packets, then lets KILL THAT NOW.
    THEY ARE USELESS! and do not belong on APRS in any case!

Still, a 98% probability (instead of 99%) of getting one's LOCAL packet
onto his LOCAL network with 100% CSMA at the digi on 144.39 is a BIG WIN in
my book...

>Second, you do not figure IGates into your equation.  
>There are multiple IGates that would be in the range of
>this mythical alternate channel digi.   Those IGates will,
>by definition, be interfering with each other.

If you have that many IGates at *each* digi, By definition they are
interfering with each other *already*!  We cannot make improvements if we
are doing stupid things on the channel and then assume we can continue to
do stupid things when we try to make improvements.

Further, by having an IGate transmitting on the ALT frequency with only a
2% probability of ANY collisions (and gaining full 100% CSMA when its
packets are digipeated over to 144.39) will be a TREMENDOUS gain for users
in that area that hope to get packets from the IGate! And this is
indepenedent of how many packets per minute this IGate generates!

If there are more than 2 Igates, for each such DIGI, then I would strongly
suggest that that problem be solved first before trying to QSY.  I would
say that any PERMANENT 24/7 IGate with  reliable upkeep would be the one
that QSY's to the alternate freq.  The others that come and go should be
discouraged from activiating their IGates or should stay on 144.39.  But
that is a separate issue... their QRM is there in either case...

>Finally, of the 20 stations I mentioned [in my example]
>there are 6 weather  stations....

Stil a hands-down winner!
3 Peets at 5 minuets      =  18 per 30 min
3 Home WXs at 10 mins     =   9 per 30 mins
14 Home stns at 2 per 30  =  28 per 30 mins
Total per 30 mins is         55 per 30 mins

Or about one packet every  33 seconds. or only a 3% probablility of
collision!  I'd take that guaranteed collision rate any day compared to
trying to launch a packet on 144.39!

> In a previous post, you claimed that over half
>of the packets on 144.39 are from mobile stations.

During Prime time, I stand by that claim.  But you are correct, that on 24
hour average the number is about 26% of all packets are mobiles...

> As Wes pointed out, the biggest problem is not fixed
>vs. mobile, but paths that are excessive for the local 
>environment.

That is completely correct.  But I am not trying to fix that problem!   I
am trying to IMPROVE the network in spite of it..

I am making a local DESIGN CHANGE to our local 144.39 to VASTLY improve
performance for everyone in our area.  This is in spite of any problems
with DX QRM.  It assures that our packets WILL GET THROUGH because our
local packets WILL GET FULL 100% CSMA to that QRM when they are
retransmitted by the local digi...

The cost (in the example of your area where you hear 20 fixed stations
direct) is a 3% probabilty of local collision.  But compared to what you
have now, I maintain that it is a hands-down winner...

>Your proposal is asking everyone with fixed stations 
>in a local area to transmit "in the blind" on the 
>alternate frequency.  In addition to being bad network 
>design,......<snip>

That statement is an uninformed red-hearing.  I have shown how it IMPROVES
network performance.  It seems impossible for you to conclude that IMPROVED
performance is bad design?

What *is* BAD design is the way we are operating 144.39 now which is having
* ALL MOBIILES
* and ALL WX STATIONS
* and ALL FIXED stations
* and ALL IGATES
* and ALL DIGIS
* and ALL DX packets from all other digis on the same input channel!!
* And trying to use that SAME channel for OUTPUT too.

It has always been fundamental ALOHA to improve throughput by operating
multiple INPUT channels at low collision rates than trying to jam
everything on one input channel at a high collision rate.  Classicaly about
4 or 5 alternate input channels are the optimum.  The proposal here is a
very simple first step to help separate out the collisions and amortize the
TXD delays... and take advantage of CSMA at the digi's that CAN hear
everything.

It's a hand's down winner if done correctly...

>this is of questionable legal standing which has been
>hashed over many times in the past on this SIG. 

Hogwash... and balderdash fodder for the uninformed...

>As I said before, your intent is admirable but the
>implementation is not.

Based on your concerns above, and the simple calculations of  fhannel
loading, I dont see the justification for such a conclusion...

And it doesn't matter if individuals like it or understand it or not.  It
is a local imporvement, available to locals to use or not.  When they see
the advantage they will use it...  If they don't, they can stay on 144.39
and take their chances with the order of magnitudes higher collision
rates...

de WB4APR, Bob

----------------------------------------------------------------------




Read previous mail | Read next mail


 01.12.2020 09:52:46lGo back Go up