sftpack, a FORTRAN90 code which carries out some "slow" Fourier transforms, that is, Fourier transforms without the techniques that allow for very fast computation.
The intention is to make the underlying calculation clear, and to provide a simple check and comparison of timing and usage for the fast calculations.
The fast Fourier transform is a highly efficient procedure. However, the coding of the algorithm, and the conventions for indexing, ordering, and normalization, can make it quite difficult to understand what is being done, or even to see how to use the software correctly. This means that a coding error or mistaken use can be undetected.
The computer code and data files described and made available on this web page are distributed under the GNU LGPL license.
sftpack is available in a C version and a C++ version and a FORTRAN90 version and a MATLAB version and a Python version.
C8LIB, a FORTRAN90 code which implements certain elementary functions for double precision complex variables;
COSINE_TRANSFORM, a FORTRAN90 code which demonstrates some simple properties of the discrete cosine transform (DCT).
FFT_OPENMP, a FORTRAN90 code which demonstrates the computation of a Fast Fourier Transform in parallel, using OpenMP.
FFT_SERIAL, a FORTRAN90 code which demonstrates the computation of a Fast Fourier Transform, and is intended as a starting point for implementing an OpenMP parallel version.
FFTPACK5, a FORTRAN90 code which implements the Fast Fourier Transform.
HAAR, a FORTRAN90 code which computes the Haar transform of data.
SINE_TRANSFORM, a FORTRAN90 code which demonstrates some simple properties of the discrete sine transform.
WAVELET, a FORTRAN90 code which does some simple wavelet calculations;