In this paper, we examine the FFT of sequences x„ = xN + „ with period N that satisfy certain symmetric relations. It is known that one can take advantage of these relations in order to reduce the computing time required by the FFT. For example, the time required for the FFT of a real sequence is about half that required for the FFT of a complex sequence. Also, the time required for the FFT of a real even sequence x„ = xN_n requires about half that required for a real sequence. We first define a class of five symmetries that are shown to be "closed" in the sense that if x„ has any one of the symmetries, then no additional symmetries are generated in the course of the FFT. Symmetric FFTs are developed that take advantage of these intermediate symmetries. They do not require the traditional preand postprocessing associated with symmetric FFTs and, as a consequence, they are somewhat more efficient and general than existing symmetric FFTs.
[1]
P. Swarztrauber.
THE METHODS OF CYCLIC REDUCTION, FOURIER ANALYSIS AND THE FACR ALGORITHM FOR THE DISCRETE SOLUTION OF POISSON'S EQUATION ON A RECTANGLE*
,
1977
.
[2]
C. Temperton.
Fast Mixed-Radix Real Fourier Transforms
,
1983
.
[3]
Peter D. Welch,et al.
The fast Fourier transform algorithm: Programming considerations in the calculation of sine, cosine and Laplace transforms☆
,
1970
.
[4]
Paul N. Swarztrauber,et al.
FFT algorithms for vector computers
,
1984,
Parallel Comput..
[5]
Jean Dollimore.
Some Algorithms for Use with the Fast Fourier Transform
,
1973
.
[6]
B. Fornberg.
A vector implementation of the Fast Fourier Transform
,
1981
.
[7]
Roger W. Hockney,et al.
A Fast Direct Solution of Poisson's Equation Using Fourier Analysis
,
1965,
JACM.