summaryrefslogtreecommitdiff
path: root/cdc-dials/Drivers/CMSIS/DSP/Source/TransformFunctions/arm_cfft_q31.c
blob: 934a3fc9fe8f4c32e2702afe342763e859307f34 (plain)
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
/* ----------------------------------------------------------------------
 * Project:      CMSIS DSP Library
 * Title:        arm_cfft_q31.c
 * Description:  Combined Radix Decimation in Frequency CFFT fixed point processing function
 *
 * $Date:        27. January 2017
 * $Revision:    V.1.5.1
 *
 * Target Processor: Cortex-M cores
 * -------------------------------------------------------------------- */
/*
 * Copyright (C) 2010-2017 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"

extern void arm_radix4_butterfly_q31(
    q31_t * pSrc,
    uint32_t fftLen,
    q31_t * pCoef,
    uint32_t twidCoefModifier);

extern void arm_radix4_butterfly_inverse_q31(
    q31_t * pSrc,
    uint32_t fftLen,
    q31_t * pCoef,
    uint32_t twidCoefModifier);

extern void arm_bitreversal_32(
    uint32_t * pSrc,
    const uint16_t bitRevLen,
    const uint16_t * pBitRevTable);

void arm_cfft_radix4by2_q31(
    q31_t * pSrc,
    uint32_t fftLen,
    const q31_t * pCoef);

void arm_cfft_radix4by2_inverse_q31(
    q31_t * pSrc,
    uint32_t fftLen,
    const q31_t * pCoef);

/**
* @ingroup groupTransforms
*/

/**
* @addtogroup ComplexFFT
* @{
*/

/**
* @details
* @brief       Processing function for the fixed-point complex FFT in Q31 format.
* @param[in]      *S    points to an instance of the fixed-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 forward (ifftFlag=0) or inverse (ifftFlag=1) transform.
* @param[in]     bitReverseFlag flag that enables (bitReverseFlag=1) or disables (bitReverseFlag=0) bit reversal of output.
* @return none.
*/

void arm_cfft_q31(
    const arm_cfft_instance_q31 * S,
    q31_t * p1,
    uint8_t ifftFlag,
    uint8_t bitReverseFlag)
{
    uint32_t L = S->fftLen;

    if (ifftFlag == 1U)
    {
        switch (L)
        {
        case 16:
        case 64:
        case 256:
        case 1024:
        case 4096:
            arm_radix4_butterfly_inverse_q31  ( p1, L, (q31_t*)S->pTwiddle, 1 );
            break;

        case 32:
        case 128:
        case 512:
        case 2048:
            arm_cfft_radix4by2_inverse_q31  ( p1, L, S->pTwiddle );
            break;
        }
    }
    else
    {
        switch (L)
        {
        case 16:
        case 64:
        case 256:
        case 1024:
        case 4096:
            arm_radix4_butterfly_q31  ( p1, L, (q31_t*)S->pTwiddle, 1 );
            break;

        case 32:
        case 128:
        case 512:
        case 2048:
            arm_cfft_radix4by2_q31  ( p1, L, S->pTwiddle );
            break;
        }
    }

    if ( bitReverseFlag )
        arm_bitreversal_32((uint32_t*)p1,S->bitRevLength,S->pBitRevTable);
}

/**
* @} end of ComplexFFT group
*/

void arm_cfft_radix4by2_q31(
    q31_t * pSrc,
    uint32_t fftLen,
    const q31_t * pCoef)
{
    uint32_t i, l;
    uint32_t n2, ia;
    q31_t xt, yt, cosVal, sinVal;
    q31_t p0, p1;

    n2 = fftLen >> 1;
    ia = 0;
    for (i = 0; i < n2; i++)
    {
        cosVal = pCoef[2*ia];
        sinVal = pCoef[2*ia + 1];
        ia++;

        l = i + n2;
        xt = (pSrc[2 * i] >> 2) - (pSrc[2 * l] >> 2);
        pSrc[2 * i] = (pSrc[2 * i] >> 2) + (pSrc[2 * l] >> 2);

        yt = (pSrc[2 * i + 1] >> 2) - (pSrc[2 * l + 1] >> 2);
        pSrc[2 * i + 1] = (pSrc[2 * l + 1] >> 2) + (pSrc[2 * i + 1] >> 2);

        mult_32x32_keep32_R(p0, xt, cosVal);
        mult_32x32_keep32_R(p1, yt, cosVal);
        multAcc_32x32_keep32_R(p0, yt, sinVal);
        multSub_32x32_keep32_R(p1, xt, sinVal);

        pSrc[2U * l] = p0 << 1;
        pSrc[2U * l + 1U] = p1 << 1;

    }

    // first col
    arm_radix4_butterfly_q31( pSrc, n2, (q31_t*)pCoef, 2U);
    // second col
    arm_radix4_butterfly_q31( pSrc + fftLen, n2, (q31_t*)pCoef, 2U);

    for (i = 0; i < fftLen >> 1; i++)
    {
        p0 = pSrc[4*i+0];
        p1 = pSrc[4*i+1];
        xt = pSrc[4*i+2];
        yt = pSrc[4*i+3];

        p0 <<= 1;
        p1 <<= 1;
        xt <<= 1;
        yt <<= 1;

        pSrc[4*i+0] = p0;
        pSrc[4*i+1] = p1;
        pSrc[4*i+2] = xt;
        pSrc[4*i+3] = yt;
    }

}

void arm_cfft_radix4by2_inverse_q31(
    q31_t * pSrc,
    uint32_t fftLen,
    const q31_t * pCoef)
{
    uint32_t i, l;
    uint32_t n2, ia;
    q31_t xt, yt, cosVal, sinVal;
    q31_t p0, p1;

    n2 = fftLen >> 1;
    ia = 0;
    for (i = 0; i < n2; i++)
    {
        cosVal = pCoef[2*ia];
        sinVal = pCoef[2*ia + 1];
        ia++;

        l = i + n2;
        xt = (pSrc[2 * i] >> 2) - (pSrc[2 * l] >> 2);
        pSrc[2 * i] = (pSrc[2 * i] >> 2) + (pSrc[2 * l] >> 2);

        yt = (pSrc[2 * i + 1] >> 2) - (pSrc[2 * l + 1] >> 2);
        pSrc[2 * i + 1] = (pSrc[2 * l + 1] >> 2) + (pSrc[2 * i + 1] >> 2);

        mult_32x32_keep32_R(p0, xt, cosVal);
        mult_32x32_keep32_R(p1, yt, cosVal);
        multSub_32x32_keep32_R(p0, yt, sinVal);
        multAcc_32x32_keep32_R(p1, xt, sinVal);

        pSrc[2U * l] = p0 << 1;
        pSrc[2U * l + 1U] = p1 << 1;

    }

    // first col
    arm_radix4_butterfly_inverse_q31( pSrc, n2, (q31_t*)pCoef, 2U);
    // second col
    arm_radix4_butterfly_inverse_q31( pSrc + fftLen, n2, (q31_t*)pCoef, 2U);

    for (i = 0; i < fftLen >> 1; i++)
    {
        p0 = pSrc[4*i+0];
        p1 = pSrc[4*i+1];
        xt = pSrc[4*i+2];
        yt = pSrc[4*i+3];

        p0 <<= 1;
        p1 <<= 1;
        xt <<= 1;
        yt <<= 1;

        pSrc[4*i+0] = p0;
        pSrc[4*i+1] = p1;
        pSrc[4*i+2] = xt;
        pSrc[4*i+3] = yt;
    }
}