[ncl-talk] Hilbert transform

[Dennis Shea]

Hi Thomas,

Apologies, I included a partial file 'demod.ncl' in my 1 Sept response.

I resurrected some code and have made a 'demod_cmplx' function which
will be in the 6.3.1 release of NCL.

See:  http://www.ncl.ucar.edu/future_release.shtml

===
I have made a version of the contributed.ncl  library which will
contains the function.
It is available for download:

   http://www.cgd.ucar.edu/~shea/contributed.ncl_631

It is the last function.

===
I created two examples. One matches the Bloomfield (Fourier Analysis
of Time Series: An Introduction) complex demodulation of annual
sunspot numbers.

 Click on:  Spectral analysis, Complex Demodulation



On Tue, Sep 1, 2015 at 7:21 AM, Dennis Shea <shea at ucar.edu> wrote:
> There is nothing distributed with NCL: documented or otherwise.
>
> An untested and undocumented simple demodulation function is attached.
> It used to be in the  'shea_util.ncl' library but I removed it for
> some forgotten reason.
>
> Commonly, a low pass filter is passed over the returned results. This
> is trivial for the amplitude spectrum. However, filtering a quantity
> like phase can be an issue.
>
> Unless you have theory to guide you, how do you pick the frequency?
>
> ===
> Filters .. http://www.ncl.ucar.edu/Applications/filter.shtml
> .
>
> On Wed, Aug 26, 2015 at 10:04 AM, Thomas Tobian <thomastobian at gmail.com> wrote:
>> Hi folks,
>>
>> Do we have any functions (perhaps undocumented) for computing complex
>> demodulation or a Hilbert transform ? Thanks in advance.
>>
>>
>> Cheers,
>> Thomas Tobian,
>>
>> _______________________________________________
>> ncl-talk mailing list
>> ncl-talk at ucar.edu
>> List instructions, subscriber options, unsubscribe:
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>