[aprssig] Viscous digipeating (was: Airborne Digis)

[Pentti Gronlund]

Bob Poortinga writes:

> Just because a certain digipeater didn't hear it, doesn't mean that is
> wasn't digipeated.  This "solution" adds needless congestion to the
> APRS frequency.  Every digipeater which hears a packet that needs
> digipeating should digipeat it without delay.  The FM capture effect
> takes care of collisions.

When viscous digipeating is applied liberally instead of plain-jane
stupid WIDE1-1 digipeaters, you will get rid of most of the congestion.

> Also, if two digipeaters are using "viscous" digipeating, how do you
> determine how long to wait?  If they both wait the same time, you
> still end up with a collision.

They don't. AX.25 devices never do, they add some random delay before
transmitting.

I am truly sorry but I do not believe in FM AFSK simulcasting without
right-on-the-target frequency and timing standards. All you will get
is howling and corrupted packets. Been there, etc.

We use 64 for persistence and 100 for slottime at most places. It en-
ables the patchy network to send the packet from one digipeater at a
time. The beauty in this is that also the downlink works quite perfect 
with only 5-10 watts of power.

I have to confess that we got rid of most congestion a long time ago 
by making the digipeaters reject position packets from too far away.
We also stopped gating message traffic from the internet, as none of 
the current igate software packages can effectively filter out message 
traffic to non-local stations.

Benjamin OH3BK
-- 
		Live Reports from the Taxman's Paradise!