[aprssig] APRS low-power-local ALT input channel
nc8q
nc8q-1 at gelm.net
Mon Sep 27 19:21:32 EDT 2004
----- Original Message -----
From: "Henk de Groot" <henk.de.groot at hetnet.nl>
To: "TAPR APRS Mailing List" <aprssig at lists.tapr.org>; "TAPR APRS Mailing List" <aprssig at lists.tapr.org>
Sent: Monday, September 27, 2004 3:48 PM
Subject: RE: [aprssig] APRS low-power-local ALT input channel
> At 10:38 27-9-2004 -0500, Wes Johnston wrote:
> >And finally, if CSMA works so well, why did they switch to DAMA in europe?
>
> I can answer that! CSMA works well if you play by the rules.
>
> Channel access pseudo code:
>
> 1) if CARRIER DETECTED then goto 2); else goto 3)
> 2) wait SLOTTIME time; goto 1)
> 3) draw a random number between 0 and 255
> 4) if random number > PERSIST then goto 2)
> 5) START TRANSMISSION; goto 1)
>
> Now what values of PERSIST and SLOTTIME should you use. That's exactly the
> problem in Europe with people who do not understand this channel access
> method. SLOTTIME is most of the time set to 10, which is 100ms, in most
> stations this is set okay. The ideal PERSIST value however depends on the
> number of users on the channel. If you are alone for example a value of 255
> is okay, you never have to leave room for another station since you are
> alone. If there are however 4 users, PERSIST should be set to 64, giving
> you a 1/4 chance of transmission. The other stations should also use this,
> giving everyboy an equal share.
When I did BBS<>BBS forwarding on a non-user channel. I used
(256/n)-1 where 'n' is the number of BBSs and nodes sharing the channel.
If there are only 2 BBSs/nodes they can each use 255 as identical keyup
collisions would be taken care of by the numbered "I" frames & SLOTTIME.
3 stations = 127 PERSIST, 4 stations = 84, 5 stations use 63, ...
However, IMHO, APRS stations have such little data to send
and such infrequent transmissions,
that a PERSISTance of 1 should be adequate.
However, this may not work with APRS's "UI" frames. I am not familiar
with "UI" frames that do not have error correction present at a higher level.
<snip>
3s, Chuck
> Kind regards,
>
> Henk.
>
>
>
> _______________________________________________
> aprssig mailing list
> aprssig at lists.tapr.org
> https://lists.tapr.org/cgi-bin/mailman/listinfo/aprssig
>
>
More information about the aprssig
mailing list