High frequency cut off (wds.m)

by DJ_USA last modified Jun 26, 2007 05:27 PM
Up to Waves

High frequency cut off (wds.m)

Posted by DJ_USA at June 25. 2007
Dear all,
I am using Nortek Matlab programs to calculate directional waves developed by Dr. Lee Gordon (PUV method).

My question is how to calculate the scale factor which controls high frequency cut off.
I have read Wave2001 paper written by Dr. Gordon, but the paper did not exactly show which is the scale factor.

I checked the program, wds.m, and the maximum scale factor (i.e. maxfac) to decide the cut off frequency is linked to the following equation.

ac=find(coshkh.^2) > maxfac);

Does that mean the scale factor shown in Wave2001 paper is coshkh.^2 ?
I am wondering if anyone could answer this question.
I checked the forums and I could not find information regarding this question.

Sincerely,

Daijiro Kobashi
Louisiana State University
Current state: Being created

High frequency cut off (wds.m)

Posted by DJ_USA at June 26. 2007
Hi Daijiro Kobashi,

I actually had to download the paper you mentioned from our support page to understand what Lee Gordon was doing myself.  

The scale factor is basically the transfer function that converts the energy spectra for the measurements to the surface energy density.  From the equation that you are showing it appears that the cutoff frequency is determined when this scale factor exceeds the value set by maxfac.  I would look into where this value is set in the code.

Alternatively, you could search for the "inflection point" that is indicated in the paper. Basically you look for a local minimum in the energy spectrum above a given frequency (for example 0.15 Hz).  This is what is done in the software for Quickwave and it seems to work quite well.  It is also less sensitve to variable noise floors. <img src=" />;)'>

Kind regards,
Torstein
Current state: Being created

High frequency cut off (wds.m)

Posted by DJ_USA at June 26. 2007
Quote (Torstein @ June 26 2007,13:17)
Alternatively, you could search for the "inflection point" that is indicated in the paper. Basically you look for a local minimum in the energy spectrum above a given frequency (for example 0.15 Hz).  This is what is done in the software for Quickwave and it seems to work quite well.  It is also less sensitve to variable noise floors. <img src=" />;)'>

Dear Dr. Torstein,
Thanks for your help and your time.

I calculated frequencies of the inflection points and the frequencies change each burst (between 0.2 Hz and 0.32 Hz).

Does Nortek Quickwave software use constant cut off frequency (averaged for all bursts) or variable cut off frequencies for each burst?
I am wondering which one I should use.

Yours Sincerely,

Daijiro Kobashi
Louisiana State University
Current state: Being created
Powered by Ploneboard
Document Actions
Log in


Forgot your password?
New user?