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
|
/* ----------------------------------------------------------------------
* Project: CMSIS DSP Library
* Title: arm_cfft_radix8_f32.c
* Description: Radix-8 Decimation in Frequency CFFT & CIFFT Floating point processing function
*
* $Date: 18. March 2019
* $Revision: V1.6.0
*
* Target Processor: Cortex-M cores
* -------------------------------------------------------------------- */
/*
* Copyright (C) 2010-2019 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 "arm_math.h"
/* ----------------------------------------------------------------------
* Internal helper function used by the FFTs
* -------------------------------------------------------------------- */
/**
brief Core function for the floating-point CFFT butterfly process.
param[in,out] pSrc points to the in-place buffer of floating-point 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_radix8_butterfly_f32(
float32_t * pSrc,
uint16_t fftLen,
const float32_t * pCoef,
uint16_t twidCoefModifier)
{
uint32_t ia1, ia2, ia3, ia4, ia5, ia6, ia7;
uint32_t i1, i2, i3, i4, i5, i6, i7, i8;
uint32_t id;
uint32_t n1, n2, j;
float32_t r1, r2, r3, r4, r5, r6, r7, r8;
float32_t t1, t2;
float32_t s1, s2, s3, s4, s5, s6, s7, s8;
float32_t p1, p2, p3, p4;
float32_t co2, co3, co4, co5, co6, co7, co8;
float32_t si2, si3, si4, si5, si6, si7, si8;
const float32_t C81 = 0.70710678118f;
n2 = fftLen;
do
{
n1 = n2;
n2 = n2 >> 3;
i1 = 0;
do
{
i2 = i1 + n2;
i3 = i2 + n2;
i4 = i3 + n2;
i5 = i4 + n2;
i6 = i5 + n2;
i7 = i6 + n2;
i8 = i7 + n2;
r1 = pSrc[2 * i1] + pSrc[2 * i5];
r5 = pSrc[2 * i1] - pSrc[2 * i5];
r2 = pSrc[2 * i2] + pSrc[2 * i6];
r6 = pSrc[2 * i2] - pSrc[2 * i6];
r3 = pSrc[2 * i3] + pSrc[2 * i7];
r7 = pSrc[2 * i3] - pSrc[2 * i7];
r4 = pSrc[2 * i4] + pSrc[2 * i8];
r8 = pSrc[2 * i4] - pSrc[2 * i8];
t1 = r1 - r3;
r1 = r1 + r3;
r3 = r2 - r4;
r2 = r2 + r4;
pSrc[2 * i1] = r1 + r2;
pSrc[2 * i5] = r1 - r2;
r1 = pSrc[2 * i1 + 1] + pSrc[2 * i5 + 1];
s5 = pSrc[2 * i1 + 1] - pSrc[2 * i5 + 1];
r2 = pSrc[2 * i2 + 1] + pSrc[2 * i6 + 1];
s6 = pSrc[2 * i2 + 1] - pSrc[2 * i6 + 1];
s3 = pSrc[2 * i3 + 1] + pSrc[2 * i7 + 1];
s7 = pSrc[2 * i3 + 1] - pSrc[2 * i7 + 1];
r4 = pSrc[2 * i4 + 1] + pSrc[2 * i8 + 1];
s8 = pSrc[2 * i4 + 1] - pSrc[2 * i8 + 1];
t2 = r1 - s3;
r1 = r1 + s3;
s3 = r2 - r4;
r2 = r2 + r4;
pSrc[2 * i1 + 1] = r1 + r2;
pSrc[2 * i5 + 1] = r1 - r2;
pSrc[2 * i3] = t1 + s3;
pSrc[2 * i7] = t1 - s3;
pSrc[2 * i3 + 1] = t2 - r3;
pSrc[2 * i7 + 1] = t2 + r3;
r1 = (r6 - r8) * C81;
r6 = (r6 + r8) * C81;
r2 = (s6 - s8) * C81;
s6 = (s6 + s8) * C81;
t1 = r5 - r1;
r5 = r5 + r1;
r8 = r7 - r6;
r7 = r7 + r6;
t2 = s5 - r2;
s5 = s5 + r2;
s8 = s7 - s6;
s7 = s7 + s6;
pSrc[2 * i2] = r5 + s7;
pSrc[2 * i8] = r5 - s7;
pSrc[2 * i6] = t1 + s8;
pSrc[2 * i4] = t1 - s8;
pSrc[2 * i2 + 1] = s5 - r7;
pSrc[2 * i8 + 1] = s5 + r7;
pSrc[2 * i6 + 1] = t2 - r8;
pSrc[2 * i4 + 1] = t2 + r8;
i1 += n1;
} while (i1 < fftLen);
if (n2 < 8)
break;
ia1 = 0;
j = 1;
do
{
/* index calculation for the coefficients */
id = ia1 + twidCoefModifier;
ia1 = id;
ia2 = ia1 + id;
ia3 = ia2 + id;
ia4 = ia3 + id;
ia5 = ia4 + id;
ia6 = ia5 + id;
ia7 = ia6 + id;
co2 = pCoef[2 * ia1];
co3 = pCoef[2 * ia2];
co4 = pCoef[2 * ia3];
co5 = pCoef[2 * ia4];
co6 = pCoef[2 * ia5];
co7 = pCoef[2 * ia6];
co8 = pCoef[2 * ia7];
si2 = pCoef[2 * ia1 + 1];
si3 = pCoef[2 * ia2 + 1];
si4 = pCoef[2 * ia3 + 1];
si5 = pCoef[2 * ia4 + 1];
si6 = pCoef[2 * ia5 + 1];
si7 = pCoef[2 * ia6 + 1];
si8 = pCoef[2 * ia7 + 1];
i1 = j;
do
{
/* index calculation for the input */
i2 = i1 + n2;
i3 = i2 + n2;
i4 = i3 + n2;
i5 = i4 + n2;
i6 = i5 + n2;
i7 = i6 + n2;
i8 = i7 + n2;
r1 = pSrc[2 * i1] + pSrc[2 * i5];
r5 = pSrc[2 * i1] - pSrc[2 * i5];
r2 = pSrc[2 * i2] + pSrc[2 * i6];
r6 = pSrc[2 * i2] - pSrc[2 * i6];
r3 = pSrc[2 * i3] + pSrc[2 * i7];
r7 = pSrc[2 * i3] - pSrc[2 * i7];
r4 = pSrc[2 * i4] + pSrc[2 * i8];
r8 = pSrc[2 * i4] - pSrc[2 * i8];
t1 = r1 - r3;
r1 = r1 + r3;
r3 = r2 - r4;
r2 = r2 + r4;
pSrc[2 * i1] = r1 + r2;
r2 = r1 - r2;
s1 = pSrc[2 * i1 + 1] + pSrc[2 * i5 + 1];
s5 = pSrc[2 * i1 + 1] - pSrc[2 * i5 + 1];
s2 = pSrc[2 * i2 + 1] + pSrc[2 * i6 + 1];
s6 = pSrc[2 * i2 + 1] - pSrc[2 * i6 + 1];
s3 = pSrc[2 * i3 + 1] + pSrc[2 * i7 + 1];
s7 = pSrc[2 * i3 + 1] - pSrc[2 * i7 + 1];
s4 = pSrc[2 * i4 + 1] + pSrc[2 * i8 + 1];
s8 = pSrc[2 * i4 + 1] - pSrc[2 * i8 + 1];
t2 = s1 - s3;
s1 = s1 + s3;
s3 = s2 - s4;
s2 = s2 + s4;
r1 = t1 + s3;
t1 = t1 - s3;
pSrc[2 * i1 + 1] = s1 + s2;
s2 = s1 - s2;
s1 = t2 - r3;
t2 = t2 + r3;
p1 = co5 * r2;
p2 = si5 * s2;
p3 = co5 * s2;
p4 = si5 * r2;
pSrc[2 * i5] = p1 + p2;
pSrc[2 * i5 + 1] = p3 - p4;
p1 = co3 * r1;
p2 = si3 * s1;
p3 = co3 * s1;
p4 = si3 * r1;
pSrc[2 * i3] = p1 + p2;
pSrc[2 * i3 + 1] = p3 - p4;
p1 = co7 * t1;
p2 = si7 * t2;
p3 = co7 * t2;
p4 = si7 * t1;
pSrc[2 * i7] = p1 + p2;
pSrc[2 * i7 + 1] = p3 - p4;
r1 = (r6 - r8) * C81;
r6 = (r6 + r8) * C81;
s1 = (s6 - s8) * C81;
s6 = (s6 + s8) * C81;
t1 = r5 - r1;
r5 = r5 + r1;
r8 = r7 - r6;
r7 = r7 + r6;
t2 = s5 - s1;
s5 = s5 + s1;
s8 = s7 - s6;
s7 = s7 + s6;
r1 = r5 + s7;
r5 = r5 - s7;
r6 = t1 + s8;
t1 = t1 - s8;
s1 = s5 - r7;
s5 = s5 + r7;
s6 = t2 - r8;
t2 = t2 + r8;
p1 = co2 * r1;
p2 = si2 * s1;
p3 = co2 * s1;
p4 = si2 * r1;
pSrc[2 * i2] = p1 + p2;
pSrc[2 * i2 + 1] = p3 - p4;
p1 = co8 * r5;
p2 = si8 * s5;
p3 = co8 * s5;
p4 = si8 * r5;
pSrc[2 * i8] = p1 + p2;
pSrc[2 * i8 + 1] = p3 - p4;
p1 = co6 * r6;
p2 = si6 * s6;
p3 = co6 * s6;
p4 = si6 * r6;
pSrc[2 * i6] = p1 + p2;
pSrc[2 * i6 + 1] = p3 - p4;
p1 = co4 * t1;
p2 = si4 * t2;
p3 = co4 * t2;
p4 = si4 * t1;
pSrc[2 * i4] = p1 + p2;
pSrc[2 * i4 + 1] = p3 - p4;
i1 += n1;
} while (i1 < fftLen);
j++;
} while (j < n2);
twidCoefModifier <<= 3;
} while (n2 > 7);
}
|
