FFT Series Function
The FFT series function is used to perform the fast Fourier
transform or the inverse fast Fourier transform of the data in the source series. This
function takes one or two source series. If only one source series is supplied,
the data in the value
component of the source series are treated as the real parts of the input into the
series function and the imaginary parts are assumed to be 0 for all input data
points; if there are two source series, the data in the first are treated as the real parts
and the data in the second are treated as the imaginary part.
Suppose that there are L data points in the series function domain,
the destination series will contains N = 2m data
points, where m is the smallest positive integer that satisfies the
condition 2m ³
L.
Normally, the FFT frequency range in the destination series is [0, 2 * (N-1)
* FN / N], where FN is the Nyquist
frequency. But if the Center at Zero check box on the FFT tab-page
of the Series Function dialog box
is
checked, the FFT frequency ranged will be changed to [-(N-2) * FN
/ N, FN] .
If the position components
of the source series contain incompatible data (e.g. the first contains 1, 2, 4,
5 and the second contains 1, 2, 3, 5) or if their contained data are not equally spaced, DataScene will use
cubic-spline
interpolation to generate equally spaced and compatible position and value components for the
source series first and then perform the calculations.
See Also
|
