y = ifft(x)
y = ifft(x,n)
ifft(x)
is the inverse fast Fourier transform of vector x
.
ifft(x,n)
is the n
-point inverse FFT.
x
, ifft(fft(x))
equals x
to within roundoff error. If x
is real, ifft(fft(x))
may have small imaginary parts.
ifft(x)
is the same as the algorithm for fft(x)
, except for a sign change and a scale factor of n = length(x)
. So the execution time is fastest when n
is a power of 2 and slowest when n
is a large prime.
fft
,fft2
,fftshift
,filter
dftmtx
, freqz
,specplot
, andspectrum
in the Signal Processing Toolbox
(c) Copyright 1994 by The MathWorks, Inc.