[aprssig] digipeter determination

[Robert Bruninga]

Yes, this technique will work to solve the problem of one strong
close digi.  But other than that, most digis are far enough away
to be able to be discriminated by antenna placement and
attenuators and directivity...   Bob

>>> Wes Johnston <aprs at kd4rdb.com> 1/7/05 9:48:53 AM >>>
I got to thinking about this this morning and I worry about using beams
to
collect data from the various 144.39 digis.  If I'm too close to one of
the
nearby digi's, it will prevent me from hearing another distant digi,
even with
a beam pointed at the other digi.

I was thinking that a more effective (although incrementally expensive
method)
would be to add an extra radio at each of the digipeater sites.  One
radio
would TX/RX on 144.39, the _ONE_ TNC at the site would work out what
needed to
be digipeated and TX (just as they do today).  The TX line from the TNC
is
hooked to the 144.39 radio AND to a 70cm radio.  Each site is assigned
it's own
70cm frequency to uplink to the super digi.  The super digi uses a
matching
radio and 1200 TNC (or sound card tnc) for each 70cm only link.

The result is that the 144.39 digis would TX simulatneously on 70cm and
144.39. 
The super digi would better be able to simultaneously RX multiple links
if they
were all on different frequencies rather than trying to RX on 144.39
using
beams to differentiate between the various surrounding digipeaters.

Wes
--



Quoting Robert Bruninga <bruninga at usna.edu>:

> >>> Wes Johnston <aprs at kd4rdb.com> 1/6/05 10:38:13 PM >>>
> >In the super lan, why do the uplinks to the super digi have
> >to be 9600?
>
> Because most TNC's cannot operate at one speed on
> RX and a different simultaneouse speed on TX.
> Thus to serve the typical user, if he RX's at 9600 you have
> to give him an input at 9600.
>
> Of course, if he wants he can send at 1200.  There are
> 8 inputs for that..  But why require 2 TNC's per user
> when one will do...
>
> > super digi collects up to 8 channels of 1200 baud,
> >aggregates into one 9600 output.... this is a cheaper
> >config, eh?
>
> That is exactly the whole concept!  But then you have
> to provide a 9600 baud input for those users that
> are using 9600 baud to receive so that they dont
> have to run two TNC's..
> Bob
>
>