1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
|
/*************************** Fraunhofer IIS FDK Tools **********************
(C) Copyright Fraunhofer IIS (2003)
All Rights Reserved
Please be advised that this software and/or program delivery is
Confidential Information of Fraunhofer and subject to and covered by the
Fraunhofer IIS Software Evaluation Agreement
between Google Inc. and Fraunhofer
effective and in full force since March 1, 2012.
You may use this software and/or program only under the terms and
conditions described in the above mentioned Fraunhofer IIS Software
Evaluation Agreement. Any other and/or further use requires a separate agreement.
$Id$
Author(s): M. Lohwasser, M. Gayer
Description:
This software and/or program is protected by copyright law and international
treaties. Any reproduction or distribution of this software and/or program,
or any portion of it, may result in severe civil and criminal penalties, and
will be prosecuted to the maximum extent possible under law.
******************************************************************************/
#include "fft_rad2.h"
#include "scramble.h"
#define __FFT_RAD2_CPP__
#if defined(__arm__) /* cppp replaced: elif */
#include "arm/fft_rad2_arm.cpp"
#elif defined(__GNUC__) && defined(__mips__) && defined(__mips_dsp) /* cppp replaced: elif */
#include "mips/fft_rad2_mips.cpp"
#endif
/*****************************************************************************
functionname: dit_fft (analysis)
description: dit-tukey-algorithm
scrambles data at entry
i.e. loop is made with scrambled data
returns:
input:
output:
*****************************************************************************/
#ifndef FUNCTION_dit_fft
void dit_fft(FIXP_DBL *x, const INT ldn, const FIXP_STP *trigdata, const INT trigDataSize)
{
const INT n=1<<ldn;
INT trigstep,i,ldm;
scramble(x,n);
/*
* 1+2 stage radix 4
*/
for (i=0;i<n*2;i+=8)
{
FIXP_DBL a00, a10, a20, a30;
a00 = (x[i + 0] + x[i + 2])>>1; /* Re A + Re B */
a10 = (x[i + 4] + x[i + 6])>>1; /* Re C + Re D */
a20 = (x[i + 1] + x[i + 3])>>1; /* Im A + Im B */
a30 = (x[i + 5] + x[i + 7])>>1; /* Im C + Im D */
x[i + 0] = a00 + a10; /* Re A' = Re A + Re B + Re C + Re D */
x[i + 4] = a00 - a10; /* Re C' = Re A + Re B - Re C - Re D */
x[i + 1] = a20 + a30; /* Im A' = Im A + Im B + Im C + Im D */
x[i + 5] = a20 - a30; /* Im C' = Im A + Im B - Im C - Im D */
a00 = a00 - x[i + 2]; /* Re A - Re B */
a10 = a10 - x[i + 6]; /* Re C - Re D */
a20 = a20 - x[i + 3]; /* Im A - Im B */
a30 = a30 - x[i + 7]; /* Im C - Im D */
x[i + 2] = a00 + a30; /* Re B' = Re A - Re B + Im C - Im D */
x[i + 6] = a00 - a30; /* Re D' = Re A - Re B - Im C + Im D */
x[i + 3] = a20 - a10; /* Im B' = Im A - Im B - Re C + Re D */
x[i + 7] = a20 + a10; /* Im D' = Im A - Im B + Re C - Re D */
}
for(ldm=3; ldm<=ldn; ++ldm)
{
INT m=(1<<ldm);
INT mh=(m>>1);
INT j,r;
trigstep=((trigDataSize << 2)>>ldm);
FDK_ASSERT(trigstep > 0);
/* Do first iteration with c=1.0 and s=0.0 separately to avoid loosing to much precision.
Beware: The impact on the overal FFT precision is rather large. */
{
j = 0;
for(r=0; r<n; r+=m)
{
INT t1 = (r+j)<<1;
INT t2 = t1 + (mh<<1);
FIXP_DBL vr,vi,ur,ui;
//cplxMultDiv2(&vi, &vr, x[t2+1], x[t2], (FIXP_SGL)1.0, (FIXP_SGL)0.0);
vi = x[t2+1]>>1;
vr = x[t2]>>1;
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur+vr;
x[t1+1] = ui+vi;
x[t2] = ur-vr;
x[t2+1] = ui-vi;
t1 += mh;
t2 = t1+(mh<<1);
//cplxMultDiv2(&vr, &vi, x[t2+1], x[t2], (FIXP_SGL)1.0, (FIXP_SGL)0.0);
vr = x[t2+1]>>1;
vi = x[t2]>>1;
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur+vr;
x[t1+1] = ui-vi;
x[t2] = ur-vr;
x[t2+1] = ui+vi;
}
}
for(j=1; j<mh/4; ++j)
{
FIXP_STP cs;
cs = trigdata[j*trigstep];
for(r=0; r<n; r+=m)
{
INT t1 = (r+j)<<1;
INT t2 = t1 + (mh<<1);
FIXP_DBL vr,vi,ur,ui;
cplxMultDiv2(&vi, &vr, x[t2+1], x[t2], cs);
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur+vr;
x[t1+1] = ui+vi;
x[t2] = ur-vr;
x[t2+1] = ui-vi;
t1 += mh;
t2 = t1+(mh<<1);
cplxMultDiv2(&vr, &vi, x[t2+1], x[t2], cs);
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur+vr;
x[t1+1] = ui-vi;
x[t2] = ur-vr;
x[t2+1] = ui+vi;
/* Same as above but for t1,t2 with j>mh/4 and thus cs swapped */
t1 = (r+mh/2-j)<<1;
t2 = t1 + (mh<<1);
cplxMultDiv2(&vi, &vr, x[t2], x[t2+1], cs);
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur+vr;
x[t1+1] = ui-vi;
x[t2] = ur-vr;
x[t2+1] = ui+vi;
t1 += mh;
t2 = t1+(mh<<1);
cplxMultDiv2(&vr, &vi, x[t2], x[t2+1], cs);
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur-vr;
x[t1+1] = ui-vi;
x[t2] = ur+vr;
x[t2+1] = ui+vi;
}
}
{
j = mh/4;
for(r=0; r<n; r+=m)
{
INT t1 = (r+j)<<1;
INT t2 = t1 + (mh<<1);
FIXP_DBL vr,vi,ur,ui;
cplxMultDiv2(&vi, &vr, x[t2+1], x[t2], STC(0x5a82799a), STC(0x5a82799a));
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur+vr;
x[t1+1] = ui+vi;
x[t2] = ur-vr;
x[t2+1] = ui-vi;
t1 += mh;
t2 = t1+(mh<<1);
cplxMultDiv2(&vr, &vi, x[t2+1], x[t2], STC(0x5a82799a), STC(0x5a82799a));
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur+vr;
x[t1+1] = ui-vi;
x[t2] = ur-vr;
x[t2+1] = ui+vi;
}
}
}
}
#endif
/*****************************************************************************
functionname: dit_ifft (synthesis)
description: dit-tukey-algorithm
scrambles data at entry
i.e. loop is made with scrambled data
returns:
input:
output:
*****************************************************************************/
#if !defined(FUNCTION_dit_ifft)
void dit_ifft(FIXP_DBL *x, const INT ldn, const FIXP_STP *trigdata, const INT trigDataSize)
{
const INT n=1<<ldn;
INT trigstep,i,ldm;
scramble(x,n);
/*
1+2 stage radix 4
*/
for (i=0;i<n*2;i+=8)
{
FIXP_DBL a0, a1, a2, a3, a00, a10, a20, a30;
a00 = (x[i + 0] + x[i + 2])>>1; /* Re A + Re B */
a10 = (x[i + 4] + x[i + 6])>>1; /* Re C + Re D */
a20 = (x[i + 1] + x[i + 3])>>1; /* Im A + Im B */
a30 = (x[i + 5] + x[i + 7])>>1; /* Im C + Im D */
a0 = (x[i + 0] - x[i + 2])>>1; /* Re A - Re B */
a2 = (x[i + 4] - x[i + 6])>>1; /* Re C - Re D */
a3 = (x[i + 1] - x[i + 3])>>1; /* Im A - Im B */
a1 = (x[i + 5] - x[i + 7])>>1; /* Im C - Im D */
x[i + 0] = a00 + a10; /* Re A' = Re A + Re B + Re C + Re D */
x[i + 4] = a00 - a10; /* Re C' = Re A + Re B - Re C - Re D */
x[i + 1] = a20 + a30; /* Im A' = Im A + Im B + Im C + Im D */
x[i + 5] = a20 - a30; /* Im C' = Im A + Im B - Im C - Im D */
x[i + 2] = a0 - a1; /* Re B' = Re A - Re B - Im C + Im D */
x[i + 6] = a0 + a1; /* Re D' = Re A - Re B + Im C - Im D */
x[i + 3] = a3 + a2; /* Im B' = Im A - Im B + Re C - Re D */
x[i + 7] = a3 - a2; /* Im D' = Im A - Im B - Re C + Re D */
}
for(ldm=3; ldm<=ldn; ++ldm)
{
const INT m=(1<<ldm);
const INT mh=(m>>1);
INT j,r;
trigstep=((trigDataSize << 2)>>ldm);
{
j = 0;
for(r=0; r<n; r+=m)
{
INT t1 = (r+j)<<1;
INT t2 = t1 + (mh<<1);
FIXP_DBL vr,vi,ur,ui;
//cplxMultDiv2(&vr, &vi, x[t2], x[t2+1], FL2FXCONST_SGL(1.0), (FIXP_SGL)0.0);
vi = x[t2+1]>>1;
vr = x[t2]>>1;
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur+vr;
x[t1+1] = ui+vi;
x[t2] = ur-vr;
x[t2+1] = ui-vi;
t1 += mh;
t2 = t1+(mh<<1);
//cplxMultDiv2(&vi, &vr, x[t2], x[t2+1], FL2FXCONST_SGL(1.0), FL2FXCONST_SGL(0.0));
vr = x[t2+1]>>1;
vi = x[t2]>>1;
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur-vr;
x[t1+1] = ui+vi;
x[t2] = ur+vr;
x[t2+1] = ui-vi;
}
}
for(j=1; j<mh/4; ++j)
{
FIXP_STP cs;
cs = trigdata[j*trigstep];
for(r=0; r<n; r+=m)
{
INT t1 = (r+j)<<1;
INT t2 = t1 + (mh<<1);
FIXP_DBL vr,vi,ur,ui;
cplxMultDiv2(&vr, &vi, x[t2], x[t2+1], cs);
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur+vr;
x[t1+1] = ui+vi;
x[t2] = ur-vr;
x[t2+1] = ui-vi;
t1 += mh;
t2 = t1+(mh<<1);
cplxMultDiv2(&vi, &vr, x[t2], x[t2+1], cs);
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur-vr;
x[t1+1] = ui+vi;
x[t2] = ur+vr;
x[t2+1] = ui-vi;
/* Same as above but for t1,t2 with j>mh/4 and thus cs swapped */
t1 = (r+mh/2-j)<<1;
t2 = t1 + (mh<<1);
cplxMultDiv2(&vr, &vi, x[t2+1], x[t2], cs);
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur-vr;
x[t1+1] = ui+vi;
x[t2] = ur+vr;
x[t2+1] = ui-vi;
t1 += mh;
t2 = t1+(mh<<1);
cplxMultDiv2(&vi, &vr, x[t2+1], x[t2], cs);
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur-vr;
x[t1+1] = ui-vi;
x[t2] = ur+vr;
x[t2+1] = ui+vi;
}
}
{
j = mh/4;
for(r=0; r<n; r+=m)
{
INT t1 = (r+mh/2-j)<<1;
INT t2 = t1 + (mh<<1);
FIXP_DBL vr,vi,ur,ui;
cplxMultDiv2(&vr, &vi, x[t2], x[t2+1], STC(0x5a82799a), STC(0x5a82799a));
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur+vr;
x[t1+1] = ui+vi;
x[t2] = ur-vr;
x[t2+1] = ui-vi;
t1 += mh;
t2 = t1+(mh<<1);
cplxMultDiv2(&vi, &vr, x[t2], x[t2+1], STC(0x5a82799a), STC(0x5a82799a));
ur = x[t1]>>1;
ui = x[t1+1]>>1;
x[t1] = ur-vr;
x[t1+1] = ui+vi;
x[t2] = ur+vr;
x[t2+1] = ui-vi;
}
}
}
}
#endif
|
