1063Re: [tracker2] NSR support?
- Sep 28, 2006On 9/28/06, Curt, WE7U <archer@...> wrote:
> This digi-on-distance thing is more general than the earlier idea,This is an interesting idea... one of the problems I've encountered
> so I like this even better! Less maintenance would be required over
> time for the digi settings.
> If this could be implemented we'd probably set up our SAR units with
> a digi range of 5 miles. The only thing required to do this would
> be a callsign-SSID and perhaps a timestamp, right? Timing-out or
> driving out of range removes a callsign from the list.
with my mobile is that there's too much DX to compete with. If the
folks in my neighboring towns would enforce either path hop limits
(which seem all too easy to spoof and work around with kpc3 digi's),
or distance limits, then the noise floor would go down here. The
newN-N has really done a good job, but there are still stragglers that
squeak thru with stupid paths.
The problem with limiting by distance is that some packets are
positionless.... messages and some weather reports. So yep, the trick
is to cache a station's position so that their positionless packets
are whitelisted as Curt aluded to. This may cause some problems with
internet messaging since (I think??) the message comes out the IGATE
first, then the position report. Worst case is that the next retry of
the message gets thru... but in the end, the message from my friend in
germany doesn't get digipeated even when we know his position b/c he's
4,000 miles away. Or do we digipeat all packets that are in 3rd party
format based on the IGATE callsign? Just me brainstorming aloud.
What about stations (errr idiots) that set their position to 0°N x
0°W? They don't get digipeated until they fix it.... there'd be a
learning curve for them.
The thing I really liked about NSR was that you could present a packet
to the network with no path and the digi would fill in the path for
you (that's not quite how NSR worked - I'm paraphrasing a bit). I
think path substituion for blank paths would be great. We're doing
that here in my town with javaaprs as a digi. For example, stupid
paths like W7-7 are "converted" to W2-1. I call a WIDE7-7 path
stupid, but frankly, I like the idea of sending all you can and the
digipeater being smart enough to figure out what a reasonable limit
is. This can be done with hops or with distance....
In the german packet system, if you run too long of a TXdelay you are
not permitted to connect. We could do the same for APRS with either
TXd or paths- send them a message that their path was too long, but
they'd have to have a tracker that could accept messages. Hmm...
probably better to just fix the path on the fly if the digi is the
first to hear it.... no need to fix paths that are 3 hops old.
What would be neat would be if I could somehow send a message to a
digi and have it message me back to tell me how many mS I could/should
trim off my txdelay. Anyone remember the x1j nodes that told you
if we are to have a reliable network, we have to have the tools to
control it and limit access, lest it be overwhelmed.
- << Previous post in topic Next post in topic >>