cfftb.f 3.5 KB

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485868788
  1. *DECK CFFTB
  2. SUBROUTINE CFFTB (N, C, WSAVE)
  3. C***BEGIN PROLOGUE CFFTB
  4. C***SUBSIDIARY
  5. C***PURPOSE Compute the unnormalized inverse of CFFTF.
  6. C***LIBRARY SLATEC (FFTPACK)
  7. C***CATEGORY J1A2
  8. C***TYPE COMPLEX (RFFTB-S, CFFTB-C)
  9. C***KEYWORDS FFTPACK, FOURIER TRANSFORM
  10. C***AUTHOR Swarztrauber, P. N., (NCAR)
  11. C***DESCRIPTION
  12. C
  13. C ********************************************************************
  14. C * NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE *
  15. C ********************************************************************
  16. C * *
  17. C * This routine uses non-standard Fortran 77 constructs and will *
  18. C * be removed from the library at a future date. You are *
  19. C * requested to use CFFTB1. *
  20. C * *
  21. C ********************************************************************
  22. C
  23. C Subroutine CFFTB computes the backward complex discrete Fourier
  24. C transform (the Fourier synthesis). Equivalently, CFFTB computes
  25. C a complex periodic sequence from its Fourier coefficients.
  26. C The transform is defined below at output parameter C.
  27. C
  28. C A call of CFFTF followed by a call of CFFTB will multiply the
  29. C sequence by N.
  30. C
  31. C The array WSAVE which is used by subroutine CFFTB must be
  32. C initialized by calling subroutine CFFTI(N,WSAVE).
  33. C
  34. C Input Parameters
  35. C
  36. C N the length of the complex sequence C. The method is
  37. C more efficient when N is the product of small primes.
  38. C
  39. C C a complex array of length N which contains the sequence
  40. C
  41. C WSAVE a real work array which must be dimensioned at least 4*N+15
  42. C in the program that calls CFFTB. The WSAVE array must be
  43. C initialized by calling subroutine CFFTI(N,WSAVE), and a
  44. C different WSAVE array must be used for each different
  45. C value of N. This initialization does not have to be
  46. C repeated so long as N remains unchanged. Thus subsequent
  47. C transforms can be obtained faster than the first.
  48. C The same WSAVE array can be used by CFFTF and CFFTB.
  49. C
  50. C Output Parameters
  51. C
  52. C C For J=1,...,N
  53. C
  54. C C(J)=the sum from K=1,...,N of
  55. C
  56. C C(K)*EXP(I*(J-1)*(K-1)*2*PI/N)
  57. C
  58. C where I=SQRT(-1)
  59. C
  60. C WSAVE contains initialization calculations which must not be
  61. C destroyed between calls of subroutine CFFTF or CFFTB
  62. C
  63. C***REFERENCES P. N. Swarztrauber, Vectorizing the FFTs, in Parallel
  64. C Computations (G. Rodrigue, ed.), Academic Press,
  65. C 1982, pp. 51-83.
  66. C***ROUTINES CALLED CFFTB1
  67. C***REVISION HISTORY (YYMMDD)
  68. C 790601 DATE WRITTEN
  69. C 830401 Modified to use SLATEC library source file format.
  70. C 860115 Modified by Ron Boisvert to adhere to Fortran 77 by
  71. C changing dummy array size declarations (1) to (*).
  72. C 861211 REVISION DATE from Version 3.2
  73. C 881128 Modified by Dick Valent to meet prologue standards.
  74. C 891214 Prologue converted to Version 4.0 format. (BAB)
  75. C 900131 Routine changed from user-callable to subsidiary
  76. C because of non-standard Fortran 77 arguments in the
  77. C call to CFFTB1. (WRB)
  78. C 920501 Reformatted the REFERENCES section. (WRB)
  79. C***END PROLOGUE CFFTB
  80. COMPLEX C
  81. DIMENSION C(*), WSAVE(*)
  82. C***FIRST EXECUTABLE STATEMENT CFFTB
  83. IF (N .EQ. 1) RETURN
  84. IW1 = N+N+1
  85. IW2 = IW1+N+N
  86. CALL CFFTB1 (N,C,WSAVE,WSAVE(IW1),WSAVE(IW2))
  87. RETURN
  88. END