[aprssig] 10m & 1000 Hz shift

[Jim Lux]

At 10:00 AM 11/27/2005, aprssig-request at lists.tapr.org wrote:


>Message: 7
>Date: Sun, 27 Nov 2005 12:19:11 -0500
>From: "Robert Bruninga" <bruninga at usna.edu>
>Subject: Re: [aprssig] APRS trackers on 10m (updated)
>To: <aprssig at lists.tapr.org>, <sean at rimboy.com>
>Message-ID: <s389a461.092 at FSGWHUB.usna.edu>
>Content-Type: text/plain; charset=US-ASCII
>
>Updated design goals for a 10m APRS tracker:
> >
> > 1) Xtal controlled 25W tracker on 28.187 MHz
> > 2) Operates straight FSK at 1200 baud 1000 Hz shift
> > 3) uses trimed down CB antennas for cheap mobiles
> > 4) XTAL controled RX can be built with just an RF
> >     amp and an FSK receiver chip -OR-
>* 5) ANY normal AX.25 TNC and an SSB RX will work too
>* 6) Tracker PIC processor INTERROGATES other
>*     VOICE radios in the car and includes the present
>*     OPERATING frequency in the position report.
>* 7) Posit should be the extremely short Mic-E format
>* 8) PIC should do automatic time-slot assignment
>       (1-to-60) according to the sum of the last 2 bytes
>       of the call plus the SSID.



Differential propagation will likely cause numerous problems with 1000 Hz 
shift on any sort of skywave 10m path, especially at 1200 
symbols/second.  Don't try to use a modulation designed for wireline comms 
and its peculiarities on a skywave path.

There are two basic strategies:

1) If you aren't looking for a two way path, then a smart receiver (at your 
Igate?) that can tolerate huge frequency offsets might be a better approach.

2) If you need good frequency control: You've got a GPS receiver hooked up, 
for gosh sakes.. use the 1pps to measure the crystal frequency in your 
transmitter, and adjust the audio frequencies you're generating so that 
you're right on.  Even receivers that don't have an explicit 1pps output 
often send out the data stream synchronized to the 1pps.

Trying to shoehorn it into some other format because you happen to have 
legacy modems is probably a bad plan.

Jim