LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

FFT spatial (X and Y data)

  Dear,
 
I would like to make a FFT (Fast Fourier Transform) spatial, not like a 2D image but with X and Y data control. 
I want link my X (not a time) and Y data to the FFT vi, but I can put just my Y data.
 
Can we make just FFT with time in X with Labview?
 
The link to the LV FFT help
 
Regards
  Julien
0 Kudos
Message 1 of 10
(4,078 Views)

Hi,

I’m not sure to really understand your problem. On the link you have a function to calculate 2D FFT! Then if you want to calculate the FFT of your X array and your Y array you can do it really easily: split your array in 2 arrays, one for X and one for Y, calculate independently the 2 FFTs and built a new array.

Let me know if I understand your question and else please post your VI and give more details about your application.

    Benjamin R.


Senior LabVIEW Developer @Neosoft


0 Kudos
Message 2 of 10
(4,056 Views)

Hi BenjaminR,

I have just X data to transform but I don't would like a FFT of time but a FFT of position (Y data).

My graph before the FFT is like this image.

 

If you require more information, let me know.

  Massif

0 Kudos
Message 3 of 10
(4,049 Views)

Hi,

Unfortunately, you don’t have any functions in LabVIEW to do that. You have to code this using math theory about Fourier Transform.

http://zone.ni.com/devzone/cda/ph/p/id/121

http://zone.ni.com/devzone/cda/ph/p/id/87

    Benjamin R.


Senior LabVIEW Developer @Neosoft


0 Kudos
Message 4 of 10
(4,044 Views)
If you only want the power spectrum of your unevely sampled data, you could use the "Unevenly Sampled Signal Spectrum".
 
 
Alternatively, you could evenly resample your data by interpolation before feeding it to the FFT.
 
BTW: I cannot see your picture. Can you attach a VI with some typical data in a diagram constant?


Message Edited by altenbach on 04-25-2008 08:24 AM
Message 5 of 10
(4,036 Views)
    Dear,
Thank you for the response.
I have joined a VI with data inside.
Regard
  Julien
0 Kudos
Message 6 of 10
(4,012 Views)
Actually the FFT and power spectrum VIs in LabVIEW does not assume the x-axis of your input data.  You can use the VI on either time-based data or spaced based data.  The difference is how you interpret the x-axis of the output of the FFT results.
 
When you apply power spectrum calculation on a time-based waveform, you get the power of the signal vs frequency.  The unit of frequency is in Hz, which means cycles per seconds. When you apply power spectrum calculation on a space-based waveform, you get the power of the signal vs spatial frequency.  The unit of spactial frequency is in cycles per meter (Or cycles per "other unit you use for position measurement, for example, mm").
 
Back to the history, when Fourier invented the Fourier Transform, he did applied the transform on a spaced-based data, which is the temperature distrubution on a metal beam. 
 
You data is sampled equally in space. So I compute the "deta position" with the first and second element in the position array and connect this deta to the dt input of the power spectrum VI.  Then the df output is the spatial frequency.  If the unit of your positin is in mm, the unit of the spactial frequency is cycles per mm.
 
Your data contains two scans (forward and backward).  You had better separate it to contain only the forward or backward data so that there is less leakage in the result power spectrum.
 
Please see the attached VI.
0 Kudos
Message 7 of 10
(3,996 Views)
Dear Nann,
 
Thank you for this completed explanation.
My data is sampled equally in space for this application, but the futures applications will have a speed who change.
Definitely, I have to make the code for this FFT.
 
Regards
 
  Massif
0 Kudos
Message 8 of 10
(3,983 Views)
Nann, can you give  me the code for labview 8.2 please.
0 Kudos
Message 9 of 10
(3,979 Views)
Here is the 8.2 VI.
0 Kudos
Message 10 of 10
(3,959 Views)