Using LabVIEW's FFT

[Orbital]

The LabVIEW FFT uses the usual eponential term:

e^(-i*2*pi*k*n / N)

Is there a way that I could modify this term to include a constant and otherwise perform the FFT as usual? i.e. replace the above exponential term with something including a constant, "c":

e^(-i*2*pi*k*n*c / N)

I'm actually using a 2D FFT on a square 2D array but hopefully a solution to the 1D case above would be easy to apply.

If this is more of a math than programming problem, a point in the right direction would still be greatly appreciated!

[johnsold · Accepted solution]

The only parameter you can control is N, the number of samples in the dataset.  You can try making 1/N' = c/N.  See if that will give you the results you expect. If c < 1, then N' > N and padding of the input data will be required.  For the opposite case a reduction of the input is needed.  Be careful about how the data is reduced so that you do not lose information.

Lynn

[Orbital]

Thanks very much!

I'm not sure if I still need it or not but I think that will work if I do.