Swarm: demeaning before computing FFTs?

[Branden Christensen]

Hi all:

Is the signal demeaned (around zero) first before computing the FFTs and plotting the spectrograms? Or is it assumed that the data streaming off the digitizer is demeaned or at least more or less evenly distributed around zero? I ask this because not all digitizers demean the signal and one of the 24-bit digitizers in my network outputs a raw signal with counts very close to the top range of the 24-bit digitizer (8+ million counts). This results in bogus looking spectrograms in Swarm. When the signal is demeaned in other software packages (e.g., SQLX), the spectograms look normal.

Saludos, 

Branden Christensen

Director, OSOP

Sign up for the OSOP mailing list: http://www.osop.com.pa/about/osop-mailing-list/

LA SALSA VIVE...Gózala. Cali 2014. 

OSOP

--

[Paul Friberg]

Demeaning in an FFT sense is only dealing with the DC offset at 0 Hertz. This should not mess up a spectrogram, except for the 0 Hz value.

Could it be that the digitizer has a clipped signal and that is messing things up, but I guess that would have been obvious.

Interesting, I have not heard of a digitizer that demeans the data by default.

Regards,

Paul

--
--
You received this message because you are subscribed to the Google
Groups "Earthworm Community Forum" group.
 
To post to this group, send an email to earthwo...@googlegroups.com
 
To unsubscribe from this group, send an email to
earthworm_for...@googlegroups.com
 
For more options, visit this group at
http://groups.google.com/group/earthworm_forum?hl=en

---
You received this message because you are subscribed to the Google Groups "Earthworm Community Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to earthworm_for...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
===================================
Paul Friberg   p.fr...@isti.com
ISTI==Instrumental Software Technologies, Inc.
Phone 845.256.9290 FAX 845.256.9299

[Branden Christensen]

Hi Paul:

Thanks for your email. 

To clarify, digitizers, at least the ones we design at OSOP, do not demean the signal in real-time. We just design the digitizer so that the output range is more or less evenly distributed around zero. I would have to talk to our engineers, but I think this is done by a simple subtraction of a known value.

I have one instrument we designed as a prototype a few years ago that outputs data in real-time with counts way up in the millions. When processing the data, I generally demean the signal back to reality before continuing with processing. I was wondering if Swarm did this automatically or not because the spectrograms generated by swarm look bogus, like the scale is completely off:

Thank you for any insights. And see you at AGU in a few months!

Saludos, 

Branden Christensen

Director, OSOP

Sign up for the OSOP mailing list: http://www.osop.com.pa/about/osop-mailing-list/

LA SALSA VIVE...Gózala. Cali 2014. 

OSOP, S.A.

--

[Mitchell M Withers (mwithers)]

I've found it to be pretty rare to have zero mean data from a digitizer.  Seismometers drift all the time.  Its always best to demean, detrend, and window to smooth the edges before doing an fft

---

From: earthwo...@googlegroups.com <earthwo...@googlegroups.com> on behalf of Branden Christensen <branden.c...@osop.com.pa>
Sent: Saturday, October 18, 2014 4:32 PM
To: Earthworm Community Forum
Subject: Re: [Earthworm Forum] Swarm: demeaning before computing FFTs?

 

[Parker, Thomas]

Can you update your version of swarm and let me know how it looks? I've fixed a couple of bugs which caused trouble for spectrograms with some data sources. You can download swarm 2.3.4.9 at:

http://volcanoes.usgs.gov/software/swarm/download.php

*************************************

Tom Parker

U.S. Geological Survey

Alaska Volcano Observatory

4200 University Drive

Anchorage, AK 99508

tpa...@usgs.gov

907-786-7030 (phone)

907-786-7401 (fax)

[Branden Christensen]

Hi Tom:

I get the same result using the newest version (2.3.4.9). Would you like me to end you a small miniSEED file?

Saludos, 

Branden Christensen

Director, OSOP

Sign up for the OSOP mailing list: http://www.osop.com.pa/about/osop-mailing-list/

LA SALSA VIVE...Gózala. Cali 2014. 

OSOP, S.A.

--

[Parker, Thomas]

Thanks for the bug report and test data. I just pushed out swarm 2.3.4.11 which will demean and detrend the signal before FFT. The spectrograms are now looking much better when fed a signal with a large offset. Please give it a shot and let me know how it goes.

--Tom

*************************************

Tom Parker

U.S. Geological Survey

Alaska Volcano Observatory

4200 University Drive

Anchorage, AK 99508

tpa...@usgs.gov

907-786-7030 (phone)

907-786-7401 (fax)

[Branden Christensen]

Thomas:

THANK YOU!

This is great. Works like a charm. 

Saludos, 

Branden Christensen

Director, OSOP

Sign up for the OSOP mailing list: http://www.osop.com.pa/about/osop-mailing-list/

LA SALSA VIVE...Gózala. Cali 2014. 

OSOP, S.A.

---

[Jean-Marie SAUREL]

Hello,

Regarding the de-meaning out of the digitizer, this is an option on
Nanometrics digitizers and is made in fact by an high-pass filter.

I wouldn't be surprised if other digitizers provide also the option to
have digital high-pass filters before the stream output.

Regards.

Jean-Marie.

Le 20/10/2014 22:11, Branden Christensen a écrit :
> Thomas:
>
>
> THANK YOU!
>
> This is great. Works like a charm.
>
>
> Saludos,
>
>

> *Branden Christensen*

> Director, OSOP
> Sign up for the OSOP mailing list:
> http://www.osop.com.pa/about/osop-mailing-list/
> LA SALSA VIVE...Gózala. Cali 2014.
>
> OSOP, S.A.
> ---
>

> --
> --
> You received this message because you are subscribed to the Google
> Groups "Earthworm Community Forum" group.
>
> To post to this group, send an email to earthwo...@googlegroups.com
>
> To unsubscribe from this group, send an email to
> earthworm_for...@googlegroups.com
>
> For more options, visit this group at
> http://groups.google.com/group/earthworm_forum?hl=en
>
> ---
> You received this message because you are subscribed to the Google
> Groups "Earthworm Community Forum" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to earthworm_for...@googlegroups.com

> <mailto:earthworm_for...@googlegroups.com>.

> For more options, visit https://groups.google.com/d/optout.


--

--------------------------------------
Institut de Physique du Globe de Paris
Observatoire Volcanologique et Sismologique de Martinique
Morne des Cadets
97250 Fonds St Denis
0596784141
Martinique - FWI

[Mitchell M Withers (mwithers)]

I don't think I'd take advantage of that feature. There could be useful information in the long period drift of the instrument. And I'd like to know when the offsets are getting large so I can recenter if necessary.

________________________________________
From: earthwo...@googlegroups.com <earthwo...@googlegroups.com> on behalf of Jean-Marie SAUREL <sau...@ovmp.martinique.univ-ag.fr>
Sent: Friday, October 24, 2014 8:31 AM
To: earthwo...@googlegroups.com

Subject: Re: [Earthworm Forum] Swarm: demeaning before computing FFTs?

Hello,

Regards.

Jean-Marie.

To unsubscribe from this group and stop receiving emails from it, send an email to earthworm_for...@googlegroups.com.