Re: limited area digipeater


James Ewen
 

The concept is called geo-fencing. I proposed the concept at least 5 years ago, if not 10 years ago. (Lynn you're off the hook, it was a proposal to the general APRS community, not just for you).

The concept is to limit the area being covered by a specific digipeater. This would allow digipeater placement where there is significant overlap with neighboring digipeaters, but where there's a hole needing to be fillled. Think a valley in the mountains that is a dead spot, but the valleys on either side are covered by neighboring digipeaters. A mountaintop that can see the dead spot, but also overlaps heavily with the neighboring digipeaters could cause issues due to the overlap. By putting a geo-fence on the new digipeater, limiting it to only the deaf valley would keep the interference to a minimum.

As with viscous digipeating, geo-fencing should only apply on the first hop. You can't limit to a specific area and still handle packets from outside the area being digipeated by neighboring digis.

Now add in another concept called dynamic digipeating, and you can deal with packets travelling further than the network can handle. The digipeater needs to keep a running total of the channel load over the last X minutes, and the distance to each station. The idea is to support the closest users while ignoring users further away.

If a channel is completely loaded with 40 users, we want to keep track of the 40 closest users. We will digipeat them, but not digipeat those further away.

The digipeater determines the distance limit dynamically based on the users it hears and the activity level. If the frequency is very quiet, you might hear the digipeater handling packets from 200 miles away. However, if you are in a busy city, your local digipeater may only handle packets for people just within the city, and ignore those who are further away.

This means that a dynamic digipeater will handle as many packets as it can reliably, droppung packets from stations far afield in order to support its local users.

In really quiet areas, the digipeaters will handle the long distance packets because they don't negatively impact the network.

All of these concepts were dismissed by Bob as being disruptive to APRS by disrupting the reliabiility of the network.

These concepts however do the exact opposite, they work to improve the reliability of the network by ensuring a better chance of packet transmissions being copied properly on a well tuned network.

Using random collisions and heterodynes to limit packet handling does not lend itself to reliable communications. Planned and implemented controlled packet handling is a better solution.

Chew on that for a bit! :)

Sent from my BlackBerry device on the Rogers Wireless Network

From: "Steve Daniels" <steve@...>
Sender: aprsisce@...
Date: Fri, 14 Feb 2014 00:17:39 -0000
To: <aprsisce@...>
ReplyTo: aprsisce@...
Subject: [aprsisce] limited area digipeater

Just setting up a crossband digipeater mainly to support Raynet in my local area. The 2M frequencies being close together.

That’s going OK.

But came across software that allows limiting the RF area that is covered. That would be a very handy function for APRSIS32

Basically using the standard filter settings to limit what is digipeated by area. That would be very handy as I often get packets bouncing down the country

 

Steve Daniels

Amateur Radio Callsign G6UIM

Torbay Freecycle  Owner

http://uk.groups.yahoo.com/group/torbay_freecycle

APRSISCE/32 Beta tester and WIKI editor http://aprsisce.wikidot.com

 

 

Join APRSISCE@groups.io to automatically receive all group messages.