On Thu, Jun 10, 2010 at 9:27 AM, n8deu <[email protected]> wrote:
>

> James, I must apologize here and read closer. My response was
> in respect reading a ~15 second delay in the original post which
> triggered my reaction. It was the original post by KC8RYW and
> chopped off in susequent posts that referenced the following:

No need to apologize. I saw the original suggestion of 15 second, and
that you picked up on the problem with using such a large delay value.

> How are you going to identify the low power tracker? I think a low
> power tracker would already be using WIDE1-1 for its first hop.

By definition, only low powered trackers needing help from fill-in
digipeaters use a path of WIDE1-1 as a first hop. Of course in reality
people have screwed that up, and people use WIDE1-1 on 50 watt
mobiles, or from airborne platforms that obviously do not need help
from fill-in digipeaters.

> If a "fill-in" digi digipeats anything other than a missed WIDE1-1 on
> the first hop it may add additional congestion under some conditions,
> because it cannot hear that the station may have already raised 2
> distance wide area digipeaters.

Well, again by definition a fill-in digipeater only acts upon the
WIDE1-1 alias (and probably it's own callsign). A fill-in digipeater
will not act upon WIDE2-2 or WIDE2-1, or anything other than WIDE1-1.

> I am certainly in favor of a "fill-in" that does not digipeat a WIDE1-1
> hop if a WIDEn-N or other digi has already digipeated the first hop.

Which is what we would be looking to create.

> How do you propose a smart "fill-in" would properly select a low
> powered tracker from a high powered station using WIDE2-2 for
> their path?

As defined above, a station asking for a hop of WIDE2-2 would not be
acted upon by the fill-in digipeater. WIDE2-2 is not WIDE1-1, and the
fill-in only acts upon WIDE1-1.

> A band opening would be an intesting event if a smart "fill-in" acts
> on the first hop of a WIDE2-2 path.

This would be the case of a fill-in digipeater being misconfigured,
and rather than being a fill-in digipeater, it would be configured as
a full digipeater.

> If a DX station raised all the smart "fill-in" stations under this scenario
> I think RF bandwidth would be wasted on a wider scale on distant networks.

Yes, indeed you would be wasting bandwidth.

However look at it this way... the way that a properly implemented
network should work. (All of the issues you have raised rely upon bad
timing choices, or improperly implemented digipeater settings.)

Let's say that we have 3 main digipeaters that act upon WIDE2-2,
WIDE2-1, and WIDE1-1 (a common setup for a digipeater that limits the
number of hops to 2 or less. Let's also say that we have a 3 more
fill-in digipeaters within this network in places where there are
spots that have less than stellar RX coverage. (The real reason to put
fill-in digipeaters out in the field).

A regular station running WIDE2-2 would only trigger the main
digipeaters. Perhaps only one on the first hop and the others on the
subsequent hop, or perhaps 2 or 3 on the first hop request, depending
upon topology.

A low powered station running WIDE1-1,WIDE2-1 might trigger a fill-in
digipeater on the first hop, and then one or more main digipeaters on
the second hop. It might also trigger a main digipeater on the first
hop if within range. It might possibly trigger one or more fill-in
digipeaters, and one or more main digipeaters on that first hop as
well. It all depends upon who can hear that station.

Now replace the regular fill-in digipeaters with SMART fill-in digipeaters.

If the low powered station can only be heard by one of the SMART
fill-ins, the fill-in will wait a second or two, and not hearing a
main digipeater reacting, it digipeats the packet, boosting it to the
main digipeater(s).

If the low powered station is heard by a SMART fill-in, and a main
digipeater, the main digipeater digipeats the packet. The SMART
digipeater hears that digipeated packet, and drops the request in its
queue.

Now let's think about a balloon tracker flying over this network using
a path of WIDE1-1,WIDE2-1 (Which unfortunately some groups feel is not
only acceptable, but necessary). EVERY digipeater, main and fill-in
will hear the initial packet. EVERY digipeater will act.

If the fill-in digipeaters are the SMART version, only the 3 main
digipeaters respond, while the SMART fill-in digipeaters hold off for
that second or two. They obviously hear the packets from the main
digipeaters, and drop their digipeat request.

The balloon concept would be similar to a high powered tracker running
WIDE1-1 as a first hop, and hitting many digipeaters at once, or a
band opening where propagation enhancements make many digipeaters hear
the packet.

Remember that the fill-in digipeater concept is designed to help
onlythose that need a boost. The current hardware available ALWAYS
helps (only WIDE1-1 requests), whether it is needed or not. Adding the
SMART routine in there allows the fill-in to boost only the stations
that NEED help, and stay quiet when they don't need help.

I hope that helps get the concept straightened out. It's difficult to
make decisions on how new algorithms will integrate into a system when
you don't have a firm grasp on the current algorithms and how they
work.

James
VE6SRV

Reply via email to