Name

sfft_apply_2d, dfft_apply_2d, cfft_apply_2d, zfft_apply_2d − Application step for fast fourier transform in two dimensions

FORMAT

Real transform: status = {S,D}FFT_APPLY_2D (input_format, output_format, direction, in, out, lda, fft_struct, ni_stride, nj_stride)

Complex transform in complex data format: status = {C,Z}FFT_APPLY_2D (input_format, output_format, direction, in, out, lda, fft_struct, ni_stride, nj_stride)

Complex transform in real data format: status = {C,Z}FFT_APPLY_2D (input_format, output_format, direction, in_real, in_imag, out_real, out_imag,
 lda, fft_struct, ni_stride, nj_stride)

Arguments

input_format, output_format
character∗(∗)
Identifies the data type of the input and the format to be used to store the data, regardless of the data type.  For example, a complex sequence can be stored in real format.

The character ’R’ specifies the format as real; the character ’C’ specifies the format as complex. As convenient, use either upper- or lowercase characters, and either spell out or abbreviate the word. 

The following table shows the valid values:

For complex data, the type of data determines what other arguments are needed.  When both the input and output data are real, the complex routines store the data as separate arrays for imaginary and real data so additional arguments are needed. 

directioncharacter∗(∗)
Specifies the operation as either the forward or inverse transform. Use ’F’ or ’f’ to specify the forward transform. Use ’B’ of ’b’ to specify the inverse transform.

in, outreal∗4 | real∗8 | complex∗8 | complex∗16
Both the arguments are two-dimensional arrays.  The input and output arrays can be the same array. The IN array contains the data to be transformed. The OUT array  contains the transformed data.

in_real, out_real, in_imag, out_imag
real∗4 | real∗8
Use these arguments when performing a complex transform on real data format and storing the result in a real data format.

ldainteger∗4
Specifies the number of rows in the IN and OUT arrays; lda >= ni. For {S,D} routines, lda >= ni+2 when the input format is ’R’ and the output format is ’C’ or the input format is ’C’ and the output format is ’R’.

fft_structrecord /dxml_s_fft_structure_2d/ for single-precision real operations
record /dxml_d_fft_structure_2d/ for double-precision real operations
record /dxml_c_fft_structure_2d/ for single-precision complex operations
record /dxml_z_fft_structure_2d/ for double-precision complex operations
The argument refers to the structure created by the _2D_INIT routine.

ni_stride, nj_strideinteger∗4
Specifies the distance between consecutive elements in a column and row in the IN and OUT arrays; the valid stride depends on the _INIT routine; ni_stride >= 1, nj_stride >= 1. 

Description

The _FFT_APPLY_2D routines compute the fast Fourier transform of two-dimensional data in either the forward or backward direction.  These routines are the second step of the three-step procedure.  They compute the fast forward or inverse Fourier transform, using the internal data structures created by the _FFT_2D_INIT subroutine. 

Use the _APPLY_2D routines with their corresponding _INIT_2D and _EXIT_2D routines. For example, use SFFT_APPLY after the SFFT_INIT and end with the  SFFT_EXIT routine.  See the DXML Reference Manual’s chapter on "Using the Signal Processing Subprograms" for an explanation of real and complex data format. 

Return Values

0DXML_SUCCESS()

11DXML_ILL_LDA()

12DXML_INS_RES()

13DXML_BAD_STRIDE()

15DXML_BAD_DIRECTION_STRING()

16DXML_BAD_FORMAT_STRING()

18 (for real transform only)
DXML_BAD_FORMAT_FOR_DIRECTION()

Examples

INCLUDE ’{file_location}DXMLDEF.FOR’
INTEGER∗4 N_I, N_J, STATUS, LDA
REAL∗8 A(1026,512), B(1026,513)
RECORD /DXML_D_FFT_STRUCTURE_2D/ FFT_STRUCT
N_I = 1024
N_J = 512
LDA = 1024
STATUS = DFFT_INIT_2D(N_I,N_J,FFT_STRUCT,.TRUE.)
STATUS = DFFT_APPLY_2D(’R’,’C’,’F’,A,B,LDA,FFT_STRUCT,1,1)
STATUS = DFFT_EXIT_2D(FFT_STRUCT)

This FORTRAN code computes the forward, two-dimensional, real FFT of a 1024x512 matrix A.  The result of the transform is stored in B in complex form. Note that the leading dimension of B is 1026 in order to hold the extra complex values (see section on data storage).  Note also that the input matrix A also requires a leading dimension of at least 1026.

INCLUDE ’{file_location}DXMLDEF.FOR’
INTEGER∗4 N_I, N_J, STATUS, LDA
COMPLEX∗16 A(1024,512), B(1024,512)
RECORD /DXML_Z_FFT_STRUCTURE_2D/ FFT_STRUCT
N_I = 1024
N_J = 512
LDA = 1024
STATUS = ZFFT_INIT_2D(N_I,N_J,FFT_STRUCT,.TRUE.)
STATUS = ZFFT_APPLY_2D(’C’,’C’,’F’,A,B,LDA,FFT_STRUCT,1,1)
STATUS = ZFFT_EXIT_2D(FFT_STRUCT)

This FORTRAN code computes the forward, two-dimensional, complex FFT of a matrix A, of dimension 1024 by 512. The result of the transform is stored in B in complex form.