Message-ID: <5337.200103161327@gemini>
From: "James Moritz" <j.r.moritz@herts.ac.uk>
Organization: University of Hertfordshire
To: rsgb_lf_group@blacksheep.org
Date: Fri, 16 Mar 2001 13:31:53 +0000
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Content-Transfer-Encoding: 8bit
Subject: LF: "Wolf" BPSK tests anyone?
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Dear LF Group,

Since transatlantic QRSS activity seems to have ceased for the 
time being, I thought I would give KK7KA's "Wolf" a try. This is 
essentially BPSK, but the encoding and decoding scheme is 
designed so that a weak signal can be effectively integrated over a 
long period. The current version of the software operates with a 
fixed-length 15 character message; with good SNR, this will 
decode in 24 seconds, but with a weak signal, the software will 
operate on up to about 25 minutes of received signal, hopefully 
leading to a significant improvement in performance. The US 
Lowfers seem to have been doing quite well with it. There is quite 
a bit of info on http://www.scgroup.com/ham/wolf.html, and also 
http://www.highnoonfilm.com/xmgr/updates/wolf.htm .

I have now tested my Decca TX with a specially built BPSK keyer 
which has the accurate timing required by Wolf. I am able to 
receive my own signals, and decode them successfully, so I know 
it works! Both the carrier frequency, and the timing of the data 
should be within 1 part in 10^7, which seems to be more than 
adequate.

The requirements for receiving this mode are basically just 
receiver plus soundcard/PC, but great care is needed in initially 
calibrating both the system frequency offsets, and the errors in the 
soundcard sampling rate, in order to successfully decode weak 
signals. This is critical for successful operation. The software 
includes some utilities to help do this. The signal processing is "off 
line"; first you record a period of received signal at 8k samples/sec 
using the software of you choice (I used DL4YHF's Spectrum Lab, 
which conveniently allows you to display a spectrogram at the 
same time). Then, you run the recorded .wav file through the Wolf 
program, together with the appropriate calibration factors. The 
program prints the decoded message. This can be repeated if 
neccessary, until you are happy with the results.

I gather VA3LK is also nearly ready to start trials using this mode, 
so if anyone is interested in trying some tests, I would be happy to 
oblige. The most important thing seems to be which frequency to 
use; from this end, the obvious choice seems to be either in the 
135.9 - 136.0kHz, or 137.4 - 137.6kHz ranges. I can vary ERP 
from about 1W down to 1mW, or less if you like, and set frequency 
in 0.1Hz steps.  Let me know what signal would best suit your 
needs. I should be able to be QRV most evenings over the next 
couple of weeks. 

Cheers, Jim Moritz
73 de M0BMU