Hi John
   I also have been using digital filtering with one of my computers with an smt-8
style vertical. I tried IIR filtering but ran into numerical stability problems. I
found better luck with multipole FIR filtering. I made a program in Basic to
create the filter coefficients for a desired shape and then have an option in my
data acquisition program to use the filter file created. It works quite well on a
386. What I like is you can create various filter responses in different frequency
ranges and if this doesn't work for the noise you just change the coeficients not
the hardware. I am trying to have the program create it's own filter based on the
long term background noise. Currently this consideration is included in my trigger
routine.
Regards
Barry

John Hernlund wrote:

> On Tue, 1 Aug 2000, Larry Conklin wrote:
> > Second, I routinely see a very long period background noise.  If I run the
> > Winquake FFT routine on a "no event" record from my system, I see a broad
> > peak around 100 seconds with several spikes in the gram around 60 - 120
> > seconds.  I wouldn't be surprised if a lot of this is thermal noise, but it
> > would be nice to know if what I'm seeing is reasonably typical of seismic
> > noise.  The noise level does seem lower at night, when there isn't any
> > activity in the house.  I understand that there are continuout microseisms
> > with a period of around 6 seconds (?).  They aren't obvious in my records,
> > perhapse because I don't have the LF gain set high enough.  My LF noise
> > level produces peaks that average around 20 or so from a 12 bit A-D
> > converter.
>
> Larry,
>    One thing not discussed often on the list is digital noise filtering.  If
> you have some programming experience, many routines for this type of filtering
> are available for free (especially in fortran).  If you see a time span in
> your seismogram that you think might be hiding an event, you can try filtering
> that section of data.  The simplest and most encountered type of noise filter
> is the Weiner filter.  It uses the mathematical idea that the noise and the
> untainted signal are not correlated.  The power spectrum of the noise and the
> untainted signal can often be easily estimated from the power spectrum
> (modulus squared frequency spectrum) of a digital record.  Knowledge of this
> then leads to the construction of the filter, which when applied to the data
> can often have amazing results.  If you or anyone else wants more info on
> this, let me know and I can post some pdf files describing the numerical
> routines and theory.

__________________________________________________________

Public Seismic Network Mailing List (PSN-L)