Content-Transfer-Encoding: 8bit
Date: Fri, 6 Aug 1999 14:59:02 +0100
Message-ID: <199908061359.OAA30093@post.interalpha.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
To: rsgb_lf_group@blacksheep.org
From: drassew2@interalpha.co.uk
Subject: LF: GBR at 16kHz
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

In a previous EMail, Rik Strobe stated that the frequency of the GBR
transmission had moved from 16kHz to 15.75kHz.  This figure was in error.
The actual frequency is centred on 15.98 kHz ie 20 Hz low.

I spoke to the group of people responsible for this transmission and they
confirmed that is was moved slightly from 16kHz a few years ago when the
transmission changed from FSK modulation to MSK. The 16kHz had referred to
one of the two FSK tone frequencies and the designated frequency was changed
to centre the energy of the new signal in the old nominated passband.

The characteristics of the signal are are:

300 baud Minimum Shift Keying (MSK) at 15.980 kHz.   This can be seen by
using Spectrogram with 16K FFT and averaging set to 100.   The MSK produces
a wide Gaussian shaped main lobe, approximately 100 Hz wide at the 3dB
points, with deep nulls at +/- 150Hz.   It is these null that allow the
signals precise frequency to be measured as they are exactly symmetrical
about the centre.

Unfortunately, the absence of a carrier component, coupled with the wide
main lobe, precludes the long term phase monitoring I had intended to do by
mixing down to zero frequency then storing I/Q values at a 4Hz sampling
rate.  The various time transmissions at 60 - 77.5kHz remain the best
signals on which to demonstrate the Doppler shifts at dawn and dusk.   It
would be possible to use the GBR signal but would involve de-spreading the
MSK code - not a trivial task.

Pure amplitude measurements are of course still valid by taking the whole
signal in its 200 Hz bandwidth

Andy  G4JNT