Clearly, I got gr.freq_xlating_fir_filter_ccc working!

Well, after a little bit of mucking about, I was able to down-shift the
segment I needed into a range that
is audible to the human ear. Here’s a snippet:

http://www.sbrac.org/files/vlf.mp3

(I have it in FLAC and OGG and WAV as well, just figured most folks
would have a MP3 player :slight_smile: ).

It’s about 40 seconds of a segment of the VLF band, between 23KHz and
28KHz, downshifted so that
23KHz is at DC. The dominant “tone” is the U.S. Navy VLF station
called “NAA”, located in Cutler, Maine,
about 850km from me. Down in the mud, there’s also the NML station
that is in North Dakota, about
1600km from me. The clicks 'n pops are 'spherics mostly caused by
distant lightning.

I pre-process the entire input bandwidth (DC-48KHz) to remove all
harmonics of 60Hz, prior to doing
anything else with it. That cleans up the spectrum considerably! The
RMS signal power for up to 4
narrow-band VLF stations is also recorded in real-time, to measure
ionospheric absorption.

Doing this audio-shifting trick was more for amusement than any
scientific value.

The whole digital portion of the signal processing chain is done in Gnu
Radio. I have a magnetic loop
antenna outside the house, about 300’ of wire wound onto a 1.8M square
frame, and I’m using
a cheap Behringer microphone pre-amp as a low-noise amplifier in front
of the sound card. If the
sound card was a little quieter, I wouldn’t need the pre-amp, but
there’s just a little too much noise
on the card to allow the signals to be really clean, so an audio
pre-amp for VLF radio signals!

Cheers

Marcus L.
Principal Investigator
Shirleys Bay Radio Astronomy Consortium

On Thu, Apr 22, 2010 at 1:07 AM, Marcus D. Leech [email protected]
wrote:

Well, after a little bit of mucking about, I was able to down-shift the
segment I needed into a range that
is audible to the human ear. Here’s a snippet:

http://www.sbrac.org/files/vlf.mp3

Marcus,
That’s great! Glad you got it working.

Since you who are experienced with all of this stuff had issues using
the freq_xlating_fir_filter, then it’s bound to be a source of
confusion for others. Would it be possible for you to add some notes
to the documentation of the class header file to help explain it
better?

Thanks,
Tom