LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

FFT conversion to spectrum display?

Man I need to bone up on spectral analysis. I would hope this is an easy one but I really have no clue.

I have an application where I am reading some FFT data from a file and displaying to a graph. Heres where my ignorance kicks in full speed. I *thought* that an FFT was the same as a spectrum analyzer but I guess not. The lower frequencies on the graph are almost always full scale where the higher frequencies fall off in what looks like a logrithmic fashion. While this does give the effect of a spectrum analyzer it leaves most of the graph unwritten to which looks funny.

Ive been doing some reading on FFTs and spectrum analyzers but, so far, I have not found the correlation if any. The main gist of the question is though if I already have some FFT data is there a way to convert this to what I want?

Included is a quick picture of the array and the formatting I am putting it through to stick it on the graph so it wont move...
0 Kudos
Message 1 of 3
(3,168 Views)
I can't speak to any of the high-level mathematics, but it seems that you could do a reverse FFT transform on the data, convert it to a waveform, and spectrum analyze it in LabVIEW. I'm assuming that by "spectrum analyzing" it you mean that you want to display the signal on an amplitude vs. frequency graph. Below I've listed the necessary VIs to implement the process I described above:

Inverse FFT: All Functions»Analyze»Signal Processing»Frequency Domain
Build Waveform: All Functions»Waveform
FFT Power Spectrum: All Functions»Analyze»Waveform Measurements

That last function will return a waveform in the amplitude/frequency domain, which you can then route directly to a graph. Hope this helps!

Regards,
E. Sulzer
Applications Engineer
National Instruments
0 Kudos
Message 2 of 3
(3,142 Views)
FFT data is essentially spectrum analyser data. The inverse of the sample length determines the frequency resolution and the Nyquist theorem determines the max frequency that can be input before aliasing. FFT data when graphed looks like a bunch of vertical bards that get shorter as you move to the right.

If it were network analyser data, they should all be less than one, unless you want an oscillator.
0 Kudos
Message 3 of 3
(3,140 Views)