NAG Library Routine Document

c06rff (fft_cosine)


    1  Purpose
    7  Accuracy


c06rff computes the discrete Fourier cosine transforms of m sequences of real data values. The elements of each sequence and its transform are stored contiguously.


Fortran Interface
Subroutine c06rff ( m, n, x, ifail)
Integer, Intent (In):: m, n
Integer, Intent (Inout):: ifail
Real (Kind=nag_wp), Intent (Inout):: x(0:n,m)
C Header Interface
#include nagmk26.h
void  c06rff_ (const Integer *m, const Integer *n, double x[], Integer *ifail)


Given m sequences of n+1 real data values xjp , for j=0,1,,n and p=1,2,,m, c06rff simultaneously calculates the Fourier cosine transforms of all the sequences defined by
x^ k p = 2n 12 x0p + j=1 n-1 xjp × cos jk πn + 12 -1k xnp ,   k= 0, 1, , n ​ and ​ p= 1, 2, , m .  
(Note the scale factor 2n  in this definition.)
This transform is also known as type-I DCT.
Since the Fourier cosine transform defined above is its own inverse, two consecutive calls of c06rff will restore the original data.
The transform calculated by this routine can be used to solve Poisson's equation when the derivative of the solution is specified at both left and right boundaries (see Swarztrauber (1977)).
The routine uses a variant of the fast Fourier transform (FFT) algorithm (see Brigham (1974)) known as the Stockham self-sorting algorithm, described in Temperton (1983), together with pre- and post-processing stages described in Swarztrauber (1982). Special coding is provided for the factors 2, 3, 4 and 5.


Brigham E O (1974) The Fast Fourier Transform Prentice–Hall
Swarztrauber P N (1977) The methods of cyclic reduction, Fourier analysis and the FACR algorithm for the discrete solution of Poisson's equation on a rectangle SIAM Rev. 19(3) 490–501
Swarztrauber P N (1982) Vectorizing the FFT's Parallel Computation (ed G Rodrique) 51–83 Academic Press
Temperton C (1983) Fast mixed-radix real Fourier transforms J. Comput. Phys. 52 340–350


1:     m – IntegerInput
On entry: m, the number of sequences to be transformed.
Constraint: m1.
2:     n – IntegerInput
On entry: one less than the number of real values in each sequence, i.e., the number of values in each sequence is n+1.
Constraint: n1.
3:     x0:nm – Real (Kind=nag_wp) arrayInput/Output
On entry: the data values of the pth sequence to be transformed, denoted by xjp, for j=0,1,,n and p=1,2,,m, must be stored in xjp.
On exit: the n+1 components of the pth Fourier cosine transform, denoted by x^kp, for k=0,1,,n and p=1,2,,m, are stored in xkp, overwriting the corresponding original values.
4:     ifail – IntegerInput/Output
On entry: ifail must be set to 0, -1​ or ​1. If you are unfamiliar with this argument you should refer to Section 3.4 in How to Use the NAG Library and its Documentation for details.
For environments where it might be inappropriate to halt program execution when an error is detected, the value -1​ or ​1 is recommended. If the output of error messages is undesirable, then the value 1 is recommended. Otherwise, if you are not familiar with this argument, the recommended value is 0. When the value -1​ or ​1 is used it is essential to test the value of ifail on exit.
On exit: ifail=0 unless the routine detects an error or a warning has been flagged (see Section 6).

Error Indicators and Warnings

If on entry ifail=0 or -1, explanatory error messages are output on the current error message unit (as defined by x04aaf).
Errors or warnings detected by the routine:
On entry, m=value.
Constraint: m1.
On entry, n=value.
Constraint: n1.
An internal error has occurred in this routine. Check the routine call and any array sizes. If the call is correct then please contact NAG for assistance.
An unexpected error has been triggered by this routine. Please contact NAG.
See Section 3.9 in How to Use the NAG Library and its Documentation for further information.
Your licence key may have expired or may not have been installed correctly.
See Section 3.8 in How to Use the NAG Library and its Documentation for further information.
Dynamic memory allocation failed.
See Section 3.7 in How to Use the NAG Library and its Documentation for further information.


Some indication of accuracy can be obtained by performing a subsequent inverse transform and comparing the results with the original sequence (in exact arithmetic they would be identical).

Parallelism and Performance

c06rff is threaded by NAG for parallel execution in multithreaded implementations of the NAG Library.
Please consult the X06 Chapter Introduction for information on how to control and interrogate the OpenMP environment used within this routine. Please also consult the Users' Note for your implementation for any additional implementation-specific information.

Further Comments

The time taken by c06rff is approximately proportional to nm logn, but also depends on the factors of n. c06rff is fastest if the only prime factors of n are 2, 3 and 5, and is particularly slow if n is a large prime, or has large prime factors. Workspace of order On is internally allocated by this routine.


This example reads in sequences of real data values and prints their Fourier cosine transforms (as computed by c06rff). It then calls c06rff again and prints the results which may be compared with the original sequence.

Program Text

Program Text (c06rffe.f90)

Program Data

Program Data (c06rffe.d)

Program Results

Program Results (c06rffe.r)

© The Numerical Algorithms Group Ltd, Oxford, UK. 2017