| 1 |
|
| 2 |
C ******************************************************************* |
| 3 |
C ** THIS FORTRAN CODE IS INTENDED TO ILLUSTRATE POINTS MADE IN ** |
| 4 |
C ** THE TEXT. TO OUR KNOWLEDGE IT WORKS CORRECTLY. HOWEVER IT IS ** |
| 5 |
C ** THE RESPONSIBILITY OF THE USER TO TEST IT, IF IT IS USED IN A ** |
| 6 |
C ** RESEARCH APPLICATION. ** |
| 7 |
C ******************************************************************* |
| 8 |
|
| 9 |
C ******************************************************************* |
| 10 |
** FICHE F.37. ROUTINES TO CALCULATE FOURIER TRANSFORMS. ** |
| 11 |
C ** THREE SEPARATE ROUTINES FOR DIFFERENT APPLICATIONS. ** |
| 12 |
C ******************************************************************* |
| 13 |
|
| 14 |
|
| 15 |
|
| 16 |
SUBROUTINE FILONC ( DT, DOM, NMAX, C, CHAT ) |
| 17 |
|
| 18 |
C ******************************************************************* |
| 19 |
C ** CALCULATES THE FOURIER COSINE TRANSFORM BY FILON'S METHOD ** |
| 20 |
C ** ** |
| 21 |
C ** A CORRELATION FUNCTION, C(T), IN THE TIME DOMAIN, IS ** |
| 22 |
C ** TRANSFORMED TO A SPECTRUM CHAT(OMEGA) IN THE FREQUENCY DOMAIN.** |
| 23 |
C ** ** |
| 24 |
C ** REFERENCE: ** |
| 25 |
C ** ** |
| 26 |
C ** FILON, PROC ROY SOC EDIN, 49 38, 1928. ** |
| 27 |
C ** ** |
| 28 |
C ** PRINCIPAL VARIABLES: ** |
| 29 |
C ** ** |
| 30 |
C ** REAL C(NMAX) THE CORRELATION FUNCTION. ** |
| 31 |
C ** REAL CHAT(NMAX) THE 1-D COSINE TRANSFORM. ** |
| 32 |
C ** REAL DT TIME INTERVAL BETWEEN POINTS IN C. ** |
| 33 |
C ** REAL DOM FREQUENCY INTERVAL FOR CHAT. ** |
| 34 |
C ** INTEGER NMAX NO. OF INTERVALS ON THE TIME AXIS ** |
| 35 |
C ** REAL OMEGA THE FREQUENCY ** |
| 36 |
C ** REAL TMAX MAXIMUM TIME IN CORRL. FUNCTION ** |
| 37 |
C ** REAL ALPHA, BETA, GAMMA FILON PARAMETERS ** |
| 38 |
C ** INTEGER TAU TIME INDEX ** |
| 39 |
C ** INTEGER NU FREQUENCY INDEX ** |
| 40 |
C ** ** |
| 41 |
C ** USAGE: ** |
| 42 |
C ** ** |
| 43 |
C ** THE ROUTINE REQUIRES THAT THE NUMBER OF INTERVALS, NMAX, IS ** |
| 44 |
C ** EVEN AND CHECKS FOR THIS CONDITION. THE FIRST VALUE OF C(T) ** |
| 45 |
C ** IS AT T=0. THE MAXIMUM TIME FOR THE CORRELATION FUNCTION IS ** |
| 46 |
C ** TMAX=DT*NMAX. FOR AN ACCURATE TRANSFORM C(TMAX)=0. ** |
| 47 |
C ******************************************************************* |
| 48 |
|
| 49 |
INTEGER NMAX |
| 50 |
REAL DT, DOM, C(0:NMAX), CHAT(0:NMAX) |
| 51 |
|
| 52 |
REAL TMAX, OMEGA, THETA, SINTH, COSTH, CE, CO |
| 53 |
REAL SINSQ, COSSQ, THSQ, THCUB, ALPHA, BETA, GAMMA |
| 54 |
INTEGER TAU, NU |
| 55 |
|
| 56 |
C ******************************************************************* |
| 57 |
|
| 58 |
C ** CHECKS NMAX IS EVEN ** |
| 59 |
|
| 60 |
IF ( MOD ( NMAX, 2 ) .NE. 0 ) THEN |
| 61 |
|
| 62 |
STOP ' NMAX SHOULD BE EVEN ' |
| 63 |
|
| 64 |
ENDIF |
| 65 |
|
| 66 |
TMAX = REAL ( NMAX ) * DT |
| 67 |
|
| 68 |
C ** LOOP OVER OMEGA ** |
| 69 |
|
| 70 |
DO 30 NU = 0, NMAX |
| 71 |
|
| 72 |
OMEGA = REAL ( NU ) * DOM |
| 73 |
THETA = OMEGA * DT |
| 74 |
|
| 75 |
C ** CALCULATE THE FILON PARAMETERS ** |
| 76 |
|
| 77 |
SINTH = SIN ( THETA ) |
| 78 |
COSTH = COS ( THETA ) |
| 79 |
SINSQ = SINTH * SINTH |
| 80 |
COSSQ = COSTH * COSTH |
| 81 |
THSQ = THETA * THETA |
| 82 |
THCUB = THSQ * THETA |
| 83 |
|
| 84 |
IF ( THETA. EQ. 0.0 ) THEN |
| 85 |
|
| 86 |
ALPHA = 0.0 |
| 87 |
BETA = 2.0 / 3.0 |
| 88 |
GAMMA = 4.0 / 3.0 |
| 89 |
|
| 90 |
ELSE |
| 91 |
|
| 92 |
ALPHA = ( 1.0 / THCUB ) |
| 93 |
: * ( THSQ + THETA * SINTH * COSTH - 2.0 * SINSQ ) |
| 94 |
BETA = ( 2.0 / THCUB ) |
| 95 |
: * ( THETA * ( 1.0 + COSSQ ) -2.0 * SINTH * COSTH ) |
| 96 |
GAMMA = ( 4.0 / THCUB ) * ( SINTH - THETA * COSTH ) |
| 97 |
|
| 98 |
ENDIF |
| 99 |
|
| 100 |
C ** DO THE SUM OVER THE EVEN ORDINATES ** |
| 101 |
|
| 102 |
CE = 0.0 |
| 103 |
|
| 104 |
DO 10 TAU = 0, NMAX, 2 |
| 105 |
|
| 106 |
CE = CE + C(TAU) * COS ( THETA * REAL ( TAU ) ) |
| 107 |
|
| 108 |
10 CONTINUE |
| 109 |
|
| 110 |
C ** SUBTRACT HALF THE FIRST AND LAST TERMS ** |
| 111 |
|
| 112 |
CE = CE - 0.5 * ( C(0) + C(NMAX) * COS ( OMEGA * TMAX ) ) |
| 113 |
|
| 114 |
C ** DO THE SUM OVER THE ODD ORDINATES ** |
| 115 |
|
| 116 |
CO = 0.0 |
| 117 |
|
| 118 |
DO 20 TAU = 1, NMAX - 1, 2 |
| 119 |
|
| 120 |
CO = CO + C(TAU) * COS ( THETA * REAL ( TAU ) ) |
| 121 |
|
| 122 |
20 CONTINUE |
| 123 |
|
| 124 |
C ** FACTOR OF TWO FOR THE REAL COSINE TRANSFORM ** |
| 125 |
|
| 126 |
CHAT(NU) = 2.0 * ( ALPHA * C(NMAX) * SIN ( OMEGA * TMAX ) |
| 127 |
: + BETA * CE + GAMMA * CO ) * DT |
| 128 |
|
| 129 |
30 CONTINUE |
| 130 |
|
| 131 |
RETURN |
| 132 |
END |
| 133 |
|
| 134 |
|
| 135 |
|
| 136 |
SUBROUTINE LADO ( DT, NMAX, C, CHAT ) |
| 137 |
|
| 138 |
C ******************************************************************* |
| 139 |
C ** CALCULATES THE FOURIER COSINE TRANSFORM BY LADO'S METHOD ** |
| 140 |
C ** ** |
| 141 |
C ** A CORRELATION FUNCTION, C(T), IN THE TIME DOMAIN, IS ** |
| 142 |
C ** TRANSFORMED TO A SPECTRUM CHAT(OMEGA) IN THE FREQUENCY DOMAIN.** |
| 143 |
C ** ** |
| 144 |
C ** REFERENCE: ** |
| 145 |
C ** ** |
| 146 |
C ** LADO, J COMPUT PHYS, 8 417, 1971. ** |
| 147 |
C ** ** |
| 148 |
C ** PRINCIPAL VARIABLES: ** |
| 149 |
C ** ** |
| 150 |
C ** REAL C(NMAX) THE CORRELATION FUNCTION. ** |
| 151 |
C ** REAL CHAT(NMAX) THE 1-D COSINE TRANSFORM. ** |
| 152 |
C ** REAL DT TIME INTERVAL BETWEEN POINTS IN C. ** |
| 153 |
C ** REAL DOM FREQUENCY INTERVAL BETWEEN POINTS IN CHAT.** |
| 154 |
C ** INTEGER NMAX NO. OF INTERVALS ON THE TIME AXIS ** |
| 155 |
C ** ** |
| 156 |
C ** USAGE: ** |
| 157 |
C ** ** |
| 158 |
C ** THE CORRELATION FUNCTION IS REQUIRED AT HALF INTEGER ** |
| 159 |
C ** INTERVALS, I.E. C(T), T=(TAU-0.5)*DT FOR TAU=1 .. NMAX. ** |
| 160 |
C ** THE COSINE TRANSFORM IS RETURNED AT HALF INTERVALS, I.E. ** |
| 161 |
C ** CHAT(OMEGA), OMEGA=(NU-0.5)*DOM FOR NU = 1 .. NMAX. ** |
| 162 |
C ******************************************************************* |
| 163 |
|
| 164 |
INTEGER NMAX |
| 165 |
REAL DT, C(NMAX), CHAT(NMAX) |
| 166 |
|
| 167 |
INTEGER TAU, NU |
| 168 |
REAL TAUH, NUH, NMAXH, PI, SUM |
| 169 |
PARAMETER ( PI = 3.1415927 ) |
| 170 |
|
| 171 |
C ******************************************************************* |
| 172 |
|
| 173 |
NMAXH = REAL ( NMAX ) - 0.5 |
| 174 |
|
| 175 |
C ** LOOP OVER OMEGA ** |
| 176 |
|
| 177 |
DO 20 NU = 1, NMAX |
| 178 |
|
| 179 |
NUH = REAL ( NU ) - 0.5 |
| 180 |
SUM = 0.0 |
| 181 |
|
| 182 |
C ** LOOP OVER T ** |
| 183 |
|
| 184 |
DO 10 TAU = 1, NMAX |
| 185 |
|
| 186 |
TAUH = REAL ( TAU ) - 0.5 |
| 187 |
SUM = SUM + C(TAU) * COS ( TAUH * NUH * PI / NMAXH ) |
| 188 |
|
| 189 |
10 CONTINUE |
| 190 |
|
| 191 |
C ** FACTOR OF TWO FOR THE REAL COSINE TRANSFORM ** |
| 192 |
|
| 193 |
CHAT(NU) = 2.0 * DT * SUM |
| 194 |
|
| 195 |
20 CONTINUE |
| 196 |
|
| 197 |
RETURN |
| 198 |
END |
| 199 |
|
| 200 |
|
| 201 |
|
| 202 |
SUBROUTINE FILONS ( DR, DK, NMAX, H, HHAT ) |
| 203 |
|
| 204 |
C ******************************************************************* |
| 205 |
C ** FOURIER SINE TRANSFORM BY FILON'S METHOD ** |
| 206 |
C ** ** |
| 207 |
C ** A SPATIAL CORRELATION FUNCTION, H(R), IS TRANSFORMED TO ** |
| 208 |
C ** HHAT(K) IN RECIPROCAL SPACE. ** |
| 209 |
C ** ** |
| 210 |
C ** REFERENCE: ** |
| 211 |
C ** ** |
| 212 |
C ** FILON, PROC ROY SOC EDIN, 49 38, 1928. ** |
| 213 |
C ** ** |
| 214 |
C ** PRINCIPAL VARIABLES: ** |
| 215 |
C ** ** |
| 216 |
C ** REAL KVEC THE WAVENUMBER ** |
| 217 |
C ** REAL RMAX MAXIMUM DIST IN CORREL. FUNCTION ** |
| 218 |
C ** REAL ALPHA, BETA, GAMMA FILON PARAMETERS ** |
| 219 |
C ** REAL H(NMAX) THE CORRELATION FUNCTION ** |
| 220 |
C ** REAL HHAT(NMAX) THE 3-D TRANSFORM ** |
| 221 |
C ** REAL DR INTERVAL BETWEEN POINTS IN H ** |
| 222 |
C ** REAL DK INTERVAL BETWEEN POINTS IN HHAT ** |
| 223 |
C ** INTEGER NMAX NO. OF INTERVALS ** |
| 224 |
C ** ** |
| 225 |
C ** USAGE: ** |
| 226 |
C ** ** |
| 227 |
C ** THE ROUTINE REQUIRES THAT THE NUMBER OF INTERVALS, NMAX, IS ** |
| 228 |
C ** EVEN AND CHECKS FOR THIS CONDITION. THE FIRST VALUE OF H(R) ** |
| 229 |
C ** IS AT R=0. THE MAXIMUM R FOR THE CORRELATION FUNCTION IS ** |
| 230 |
C ** RMAX=DR*NMAX. FOR AN ACCURATE TRANSFORM H(RMAX)=0. ** |
| 231 |
C ******************************************************************* |
| 232 |
|
| 233 |
INTEGER NMAX |
| 234 |
REAL DR, DK, H(0:NMAX), HHAT(0:NMAX) |
| 235 |
|
| 236 |
REAL RMAX, K, THETA, SINTH, COSTH |
| 237 |
REAL SINSQ, COSSQ, THSQ, THCUB, ALPHA, BETA, GAMMA |
| 238 |
REAL SE, SO, FOURPI, R |
| 239 |
INTEGER IR, IK |
| 240 |
|
| 241 |
C ******************************************************************* |
| 242 |
|
| 243 |
C ** CHECKS NMAX IS EVEN ** |
| 244 |
|
| 245 |
IF ( MOD ( NMAX, 2 ) .NE. 0 ) THEN |
| 246 |
|
| 247 |
STOP ' NMAX SHOULD BE EVEN ' |
| 248 |
|
| 249 |
ENDIF |
| 250 |
|
| 251 |
FOURPI = 16.0 * ATAN ( 1.0 ) |
| 252 |
RMAX = REAL ( NMAX ) * DR |
| 253 |
|
| 254 |
C ** LOOP OVER K ** |
| 255 |
|
| 256 |
DO 30 IK = 0, NMAX |
| 257 |
|
| 258 |
K = REAL ( IK ) * DK |
| 259 |
THETA = K * DR |
| 260 |
|
| 261 |
C ** CALCULATE THE FILON PARAMETERS ** |
| 262 |
|
| 263 |
SINTH = SIN ( THETA ) |
| 264 |
COSTH = COS ( THETA ) |
| 265 |
SINSQ = SINTH * SINTH |
| 266 |
COSSQ = COSTH * COSTH |
| 267 |
THSQ = THETA * THETA |
| 268 |
THCUB = THSQ * THETA |
| 269 |
|
| 270 |
IF ( THETA. EQ. 0.0 ) THEN |
| 271 |
|
| 272 |
ALPHA = 0.0 |
| 273 |
BETA = 2.0 / 3.0 |
| 274 |
GAMMA = 4.0 / 3.0 |
| 275 |
|
| 276 |
ELSE |
| 277 |
|
| 278 |
ALPHA = ( 1.0 / THCUB ) |
| 279 |
: * ( THSQ + THETA * SINTH * COSTH - 2.0 * SINSQ ) |
| 280 |
BETA = ( 2.0 / THCUB ) |
| 281 |
: * ( THETA * ( 1.0 + COSSQ ) -2.0 * SINTH * COSTH ) |
| 282 |
GAMMA = ( 4.0 / THCUB ) * ( SINTH - THETA * COSTH ) |
| 283 |
|
| 284 |
ENDIF |
| 285 |
|
| 286 |
C ** THE INTEGRAND IS H(R) * R FOR THE 3-D TRANSFORM ** |
| 287 |
|
| 288 |
C ** DO THE SUM OVER THE EVEN ORDINATES ** |
| 289 |
|
| 290 |
SE = 0.0 |
| 291 |
|
| 292 |
DO 10 IR = 0, NMAX, 2 |
| 293 |
|
| 294 |
R = REAL ( IR ) * DR |
| 295 |
SE = SE + H(IR) * R * SIN ( K * R ) |
| 296 |
|
| 297 |
10 CONTINUE |
| 298 |
|
| 299 |
C ** SUBTRACT HALF THE FIRST AND LAST TERMS ** |
| 300 |
C ** HERE THE FIRST TERM IS ZERO ** |
| 301 |
|
| 302 |
SE = SE - 0.5 * ( H(NMAX) * RMAX * SIN ( K * RMAX ) ) |
| 303 |
|
| 304 |
C ** DO THE SUM OVER THE ODD ORDINATES ** |
| 305 |
|
| 306 |
SO = 0.0 |
| 307 |
|
| 308 |
DO 20 IR = 1, NMAX - 1, 2 |
| 309 |
|
| 310 |
R = REAL ( IR ) * DR |
| 311 |
SO = SO + H(IR) * R * SIN ( K * R ) |
| 312 |
|
| 313 |
20 CONTINUE |
| 314 |
|
| 315 |
HHAT(IK) = ( - ALPHA * H(NMAX) * RMAX * COS ( K * RMAX) |
| 316 |
: + BETA * SE + GAMMA * SO ) * DR |
| 317 |
|
| 318 |
C ** INCLUDE NORMALISING FACTOR ** |
| 319 |
|
| 320 |
HHAT(IK) = FOURPI * HHAT(IK) / K |
| 321 |
|
| 322 |
30 CONTINUE |
| 323 |
|
| 324 |
RETURN |
| 325 |
END |
| 326 |
|
| 327 |
|
