| Re: IDL FFT (spec -> interferogram) [message #30160 is a reply to message #30029] |
Fri, 05 April 2002 01:17  |
Randall Skelton
Messages: 169
Registered: October 2000
|
Senior Member |
|
|
On Thu, 4 Apr 2002, Paul van Delst wrote:
> Randall Skelton wrote:
>>
>> Hi all,
>>
>> Having read through all of the FFT posts that google groups keeps, I am no
>> closer to understanding why I am unable to transform a spectrum into an
>> interferogram using IDL. All of the data files, procedures, and pictures
>> of this are at http://tulip.atm.ox.ac.uk/~rhskelto/fft-help/
>
> all your spec data is zero.
Not quite. In the real domain there is a single sharp peak while in the
imaginary part there is a sharp dispersion curve-- both are centered about
the index 256 in the 512 point case (or 1024 in the 2048 point case).
Experimentally, this can be thought of as a the result of inputting a
mode stabilized laser into an interferometer.
>
>> The result 'idl_igm' contains twice the number of points (minus 2)
>> because of the required reflection about the Nyquist frequency.
>> Moreover, the result appears to be modulated (almost like a frequency
>> chirp)?
>
> This looks like the correct result to me - i.e. an interferogram of a
> spectrum. Do you want the envelope of this?
I'm not entirely sure what you mean by the "envelope of this?" I am
trying to examine the center-burst of the resulting interferogram as you
did with the AIRS data in your FFT comparison. I would expect a single
line spectrum to give rise to a center-burst interferogram (am I off base
here?).
The procedure shown for AIRS (and in fft_to_interferogram.pro) does not
seem to work for me:
IDL> n = 2048 ; could have used the 512 case
IDL> spec = dcomplexarr(n)
IDL> read_cmplx, 'spec2048.dat', spec
IDL> real_data = double(spec)
IDL> imag_data = imaginary(spec)
IDL> real_part = [ real_data, REVERSE( real_data[ 1 : n - 2 ] ) ]
IDL> imag_part = [ imag_data, -1.0 * REVERSE( imag_data[ 1 : n - 2 ] ) ]
IDL> cxs = COMPLEX( real_part, imag_part )
IDL> ifg = FFT( cxs, /INVERSE )
IDL> ifg = shift( ifg, -1 * ( n - 1 ) )
IDL> plot, ifg
In this case, 'ifg' does not look correct. I have been out of the lab and
sitting in front of a computer for over a year now but this isn't what I
remember seeing on the scope when I did these sorts of things.
The following example, however, does give what I would expect. Note that
instead folding about the Nyquist frequency with the imaginary part
rotated, I simply change the sign of all the even indexed points (i.e.
reflecting even points about the x-axis).
IDL> for i=0, n-1, 2 do spec[i] = -spec[i]
IDL> ifg2 = fft(spec)
IDL> for i=0, n-1, 2 do ifg2[i] = -ifg2[i]
IDL> plot, ifg2
To be perfectly honest, I'm not exactly sure why this works (yet). If
anyone has any insight, I'd love to hear it! Otherwise, I'm off to the
engineering library...
Cheers,
Randall
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
