[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Discuss-gnuradio] Squelch block, AGC function and scanner example
From: |
Martin Dvh |
Subject: |
Re: [Discuss-gnuradio] Squelch block, AGC function and scanner example |
Date: |
Wed, 19 Jan 2005 20:04:08 +0100 |
User-agent: |
Mozilla Thunderbird 0.9 (X11/20041124) |
Frank Brickle wrote:
Martin Dvh wrote:
> I hope also to find a way to determine what kind of signals they are
(FM, AM, TV, FSK)
>
> This way you get a list of everything there is to receive with your
current frontend.
A non-trivial task. Robust classification in this setting means doing
selective channel downconversion and making repeated sequential spectral
observations on the channels.
I was thinking of the following:
Find parts of the spectrum that are significantly above the noise level (peaks)
and add these to a list.
After that process them one at a time (estimate bandwidth and do selective
channel downconversion)
Then try AM-demod, FM-demod, (LSB-demod, USB-demod)
If one of these has a outputlevel above a certain threshold then mark it as
AM-like, FM/FSK-like etc.
All of this does not have to be in realtime. Just save the
powerspectra/original signal and process them later.
In broadband, this usually requires maintaining a substantial delay line
from which to grab and process the selected channels. In commercial
systems like the Agilent E3238S the delay line has been the one of the
most expensive components.
If you're interested, I wrote a brief expository article about automatic
signal classification using probability models of power spectra that
appeared in the December 2003 QEX.
Thanks for the tip.
Do you know if QEX is available in europe. (I live in the Netherlands)?
If I order a back-issue-article from the united states it takes a while to get
here and shipping costs tend to rise.
Greetings,
Martin