[aprssig] RE: timeslotting on HF ? was > 15. APRS trackers on10m (Robert Bruninga)

[scott at opentrac.org]

Do randomly assigned slots really give us any particular advantage over
ALOHA?  If you have the bad luck to have a hash collision with another
station, you're going to be colliding with them ALL the time (or at least
half the time, with Curt's suggestion.)  And the probability of that is
probably higher than you think.  See
http://en.wikipedia.org/wiki/Birthday_paradox.

Now if you were listening in your chosen slot and heard that someone else
was already there, you could bump yourself to the next slot.  And maybe keep
listening for other open slots in your own rotation, so once in awhile you
can move in case someone else landed in the same slot without hearing you,
and now neither of you are detecting collisions because you're always in
exactly the same slot.  You could do the slot selection easily with a
pseudo-random number generator seeded with your callsign+ssid.  To get the
next slot, you just advance the PRNG one step, so everyone follows a
different sequence.

Without some way to detect conflicts, I don't think it's going to be an
improvement over what we've got now.

Scott
N1VG 

> -----Original Message-----
> From: aprssig-bounces at lists.tapr.org 
> [mailto:aprssig-bounces at lists.tapr.org] On Behalf Of Curt Mills
> Sent: Sunday, November 27, 2005 10:47 AM
> To: TAPR APRS Mailing List
> Subject: Re: [aprssig] RE: timeslotting on HF ? was > 15. 
> APRS trackers on10m (Robert Bruninga)
> 
> On Sun, 27 Nov 2005, Robert Bruninga wrote:
> 
> > But, since a PIC is involved, I guess,...  Why not?  The
> > onboard PIC does know the callsign of the owner, so it 
> > could then assign a slot based on an algorithm based on 
> > the call (including SSID in the algorithm).  Then in the 
> > unlikely event that there was a conflict, the stations 
> > affected could resolve it by changing one of  their SSID's.
> 
> Just switch the algorithm on alternate timeslots so that conflicts
> between callsigns using one algorithm would be very unlikely to be
> conflicts using the other algorithm.
> 
> -- 
> Curt, WE7U.				archer at eskimo dot com
> http://www.eskimo.com/~archer
>   Lotto:  A tax on people who are bad at math. - unknown
> Windows:  Microsoft's tax on computer illiterates. - WE7U.
> The world DOES revolve around me:  I picked the coordinate system!"
> 
> _______________________________________________
> aprssig mailing list
> aprssig at lists.tapr.org
> https://lists.tapr.org/cgi-bin/mailman/listinfo/aprssig
> 
>