Chris@82: Chris@82: Chris@82: Chris@82: Chris@82:
Chris@82:Chris@82: Next: The 1d Real-data DFT, Previous: What FFTW Really Computes, Up: What FFTW Really Computes [Contents][Index]
Chris@82:The forward (FFTW_FORWARD
) discrete Fourier transform (DFT) of a
Chris@82: 1d complex array X of size n computes an array Y,
Chris@82: where:
Chris@82:
FFTW_BACKWARD
) DFT computes:
Chris@82: FFTW computes an unnormalized transform, in that there is no coefficient Chris@82: in front of the summation in the DFT. In other words, applying the Chris@82: forward and then the backward transform will multiply the input by Chris@82: n. Chris@82:
Chris@82: Chris@82:From above, an FFTW_FORWARD
transform corresponds to a sign of
Chris@82: -1 in the exponent of the DFT. Note also that we use the
Chris@82: standard “in-order” output ordering—the k-th output
Chris@82: corresponds to the frequency k/n (or k/T, where T
Chris@82: is your total sampling period). For those who like to think in terms of
Chris@82: positive and negative frequencies, this means that the positive
Chris@82: frequencies are stored in the first half of the output and the negative
Chris@82: frequencies are stored in backwards order in the second half of the
Chris@82: output. (The frequency -k/n is the same as the frequency
Chris@82: (n-k)/n.)
Chris@82: