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
|
/* ----------------------------------------------------------------------
* Project: CMSIS DSP Library
* Title: arm_cfft_f64.c
* Description: Combined Radix Decimation in Frequency CFFT Double Precision Floating point processing function
*
* $Date: 23 April 2021
* $Revision: V1.9.0
*
* Target Processor: Cortex-M and Cortex-A cores
* -------------------------------------------------------------------- */
/*
* Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
*
* SPDX-License-Identifier: Apache-2.0
*
* Licensed under the Apache License, Version 2.0 (the License); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an AS IS BASIS, WITHOUT
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include "dsp/transform_functions.h"
#include "arm_common_tables.h"
extern void arm_radix4_butterfly_f64(
float64_t * pSrc,
uint16_t fftLen,
const float64_t * pCoef,
uint16_t twidCoefModifier);
extern void arm_bitreversal_64(
uint64_t * pSrc,
const uint16_t bitRevLen,
const uint16_t * pBitRevTable);
/**
* @} end of ComplexFFT group
*/
/* ----------------------------------------------------------------------
* Internal helper function used by the FFTs
* ---------------------------------------------------------------------- */
/*
* @brief Core function for the Double Precision floating-point CFFT butterfly process.
* @param[in, out] *pSrc points to the in-place buffer of F64 data type.
* @param[in] fftLen length of the FFT.
* @param[in] *pCoef points to the twiddle coefficient buffer.
* @param[in] twidCoefModifier twiddle coefficient modifier that supports different size FFTs with the same twiddle factor table.
* @return none.
*/
void arm_radix4_butterfly_f64(
float64_t * pSrc,
uint16_t fftLen,
const float64_t * pCoef,
uint16_t twidCoefModifier)
{
float64_t co1, co2, co3, si1, si2, si3;
uint32_t ia1, ia2, ia3;
uint32_t i0, i1, i2, i3;
uint32_t n1, n2, j, k;
float64_t t1, t2, r1, r2, s1, s2;
/* Initializations for the fft calculation */
n2 = fftLen;
n1 = n2;
for (k = fftLen; k > 1U; k >>= 2U)
{
/* Initializations for the fft calculation */
n1 = n2;
n2 >>= 2U;
ia1 = 0U;
/* FFT Calculation */
j = 0;
do
{
/* index calculation for the coefficients */
ia2 = ia1 + ia1;
ia3 = ia2 + ia1;
co1 = pCoef[ia1 * 2U];
si1 = pCoef[(ia1 * 2U) + 1U];
co2 = pCoef[ia2 * 2U];
si2 = pCoef[(ia2 * 2U) + 1U];
co3 = pCoef[ia3 * 2U];
si3 = pCoef[(ia3 * 2U) + 1U];
/* Twiddle coefficients index modifier */
ia1 = ia1 + twidCoefModifier;
i0 = j;
do
{
/* index calculation for the input as, */
/* pSrc[i0 + 0], pSrc[i0 + fftLen/4], pSrc[i0 + fftLen/2], pSrc[i0 + 3fftLen/4] */
i1 = i0 + n2;
i2 = i1 + n2;
i3 = i2 + n2;
/* xa + xc */
r1 = pSrc[(2U * i0)] + pSrc[(2U * i2)];
/* xa - xc */
r2 = pSrc[(2U * i0)] - pSrc[(2U * i2)];
/* ya + yc */
s1 = pSrc[(2U * i0) + 1U] + pSrc[(2U * i2) + 1U];
/* ya - yc */
s2 = pSrc[(2U * i0) + 1U] - pSrc[(2U * i2) + 1U];
/* xb + xd */
t1 = pSrc[2U * i1] + pSrc[2U * i3];
/* xa' = xa + xb + xc + xd */
pSrc[2U * i0] = r1 + t1;
/* xa + xc -(xb + xd) */
r1 = r1 - t1;
/* yb + yd */
t2 = pSrc[(2U * i1) + 1U] + pSrc[(2U * i3) + 1U];
/* ya' = ya + yb + yc + yd */
pSrc[(2U * i0) + 1U] = s1 + t2;
/* (ya + yc) - (yb + yd) */
s1 = s1 - t2;
/* (yb - yd) */
t1 = pSrc[(2U * i1) + 1U] - pSrc[(2U * i3) + 1U];
/* (xb - xd) */
t2 = pSrc[2U * i1] - pSrc[2U * i3];
/* xc' = (xa-xb+xc-xd)co2 + (ya-yb+yc-yd)(si2) */
pSrc[2U * i1] = (r1 * co2) + (s1 * si2);
/* yc' = (ya-yb+yc-yd)co2 - (xa-xb+xc-xd)(si2) */
pSrc[(2U * i1) + 1U] = (s1 * co2) - (r1 * si2);
/* (xa - xc) + (yb - yd) */
r1 = r2 + t1;
/* (xa - xc) - (yb - yd) */
r2 = r2 - t1;
/* (ya - yc) - (xb - xd) */
s1 = s2 - t2;
/* (ya - yc) + (xb - xd) */
s2 = s2 + t2;
/* xb' = (xa+yb-xc-yd)co1 + (ya-xb-yc+xd)(si1) */
pSrc[2U * i2] = (r1 * co1) + (s1 * si1);
/* yb' = (ya-xb-yc+xd)co1 - (xa+yb-xc-yd)(si1) */
pSrc[(2U * i2) + 1U] = (s1 * co1) - (r1 * si1);
/* xd' = (xa-yb-xc+yd)co3 + (ya+xb-yc-xd)(si3) */
pSrc[2U * i3] = (r2 * co3) + (s2 * si3);
/* yd' = (ya+xb-yc-xd)co3 - (xa-yb-xc+yd)(si3) */
pSrc[(2U * i3) + 1U] = (s2 * co3) - (r2 * si3);
i0 += n1;
} while ( i0 < fftLen);
j++;
} while (j <= (n2 - 1U));
twidCoefModifier <<= 2U;
}
}
/*
* @brief Core function for the Double Precision floating-point CFFT butterfly process.
* @param[in, out] *pSrc points to the in-place buffer of F64 data type.
* @param[in] fftLen length of the FFT.
* @param[in] *pCoef points to the twiddle coefficient buffer.
* @param[in] twidCoefModifier twiddle coefficient modifier that supports different size FFTs with the same twiddle factor table.
* @return none.
*/
void arm_cfft_radix4by2_f64(
float64_t * pSrc,
uint32_t fftLen,
const float64_t * pCoef)
{
uint32_t i, l;
uint32_t n2, ia;
float64_t xt, yt, cosVal, sinVal;
float64_t p0, p1,p2,p3,a0,a1;
n2 = fftLen >> 1;
ia = 0;
for (i = 0; i < n2; i++)
{
cosVal = pCoef[2*ia];
sinVal = pCoef[2*ia + 1];
ia++;
l = i + n2;
/* Butterfly implementation */
a0 = pSrc[2 * i] + pSrc[2 * l];
xt = pSrc[2 * i] - pSrc[2 * l];
yt = pSrc[2 * i + 1] - pSrc[2 * l + 1];
a1 = pSrc[2 * l + 1] + pSrc[2 * i + 1];
p0 = xt * cosVal;
p1 = yt * sinVal;
p2 = yt * cosVal;
p3 = xt * sinVal;
pSrc[2 * i] = a0;
pSrc[2 * i + 1] = a1;
pSrc[2 * l] = p0 + p1;
pSrc[2 * l + 1] = p2 - p3;
}
// first col
arm_radix4_butterfly_f64( pSrc, n2, (float64_t*)pCoef, 2U);
// second col
arm_radix4_butterfly_f64( pSrc + fftLen, n2, (float64_t*)pCoef, 2U);
}
/**
@addtogroup ComplexFFT
@{
*/
/**
@brief Processing function for the Double Precision floating-point complex FFT.
@param[in] S points to an instance of the Double Precision floating-point CFFT structure
@param[in,out] p1 points to the complex data buffer of size <code>2*fftLen</code>. Processing occurs in-place
@param[in] ifftFlag flag that selects transform direction
- value = 0: forward transform
- value = 1: inverse transform
@param[in] bitReverseFlag flag that enables / disables bit reversal of output
- value = 0: disables bit reversal of output
- value = 1: enables bit reversal of output
@return none
*/
void arm_cfft_f64(
const arm_cfft_instance_f64 * S,
float64_t * p1,
uint8_t ifftFlag,
uint8_t bitReverseFlag)
{
uint32_t L = S->fftLen, l;
float64_t invL, * pSrc;
if (ifftFlag == 1U)
{
/* Conjugate input data */
pSrc = p1 + 1;
for(l=0; l<L; l++)
{
*pSrc = -*pSrc;
pSrc += 2;
}
}
switch (L)
{
case 16:
case 64:
case 256:
case 1024:
case 4096:
arm_radix4_butterfly_f64 (p1, L, (float64_t*)S->pTwiddle, 1U);
break;
case 32:
case 128:
case 512:
case 2048:
arm_cfft_radix4by2_f64 ( p1, L, (float64_t*)S->pTwiddle);
break;
}
if ( bitReverseFlag )
arm_bitreversal_64((uint64_t*)p1, S->bitRevLength,S->pBitRevTable);
if (ifftFlag == 1U)
{
invL = 1.0 / (float64_t)L;
/* Conjugate and scale output data */
pSrc = p1;
for(l=0; l<L; l++)
{
*pSrc++ *= invL ;
*pSrc = -(*pSrc) * invL;
pSrc++;
}
}
}
/**
@} end of ComplexFFT group
*/
|
