mbed official / mbed-dsp

CMSIS DSP library

Dependents:   performance_timer Surfboard_ gps2rtty Capstone ... more

Legacy Warning

This is an mbed 2 library. To learn more about mbed OS 5, visit the docs.

cmsis_dsp/FilteringFunctions/arm_fir_q7.c@5:3762170b6d4d, 2015-11-20 (annotated)

Committer:
mbed_official
Date:
Fri Nov 20 08:45:18 2015 +0000
Revision:
5:3762170b6d4d
Parent:
3:7a284390b0ce 
Synchronized with git revision 2eb940b9a73af188d3004a2575fdfbb05febe62b



Full URL: https://github.com/mbedmicro/mbed/commit/2eb940b9a73af188d3004a2575fdfbb05febe62b/



Added option to build rpc library. closes #1426

Who changed what in which revision?

UserRevisionLine numberNew contents of line
emilmont 1:fdd22bb7aa52 1 /* ----------------------------------------------------------------------
mbed_official 5:3762170b6d4d 2 * Copyright (C) 2010-2014 ARM Limited. All rights reserved.
emilmont 1:fdd22bb7aa52 3 *
mbed_official 5:3762170b6d4d 4 * $Date: 19. March 2015
mbed_official 5:3762170b6d4d 5 * $Revision: V.1.4.5
emilmont 1:fdd22bb7aa52 6 *
emilmont 2:da51fb522205 7 * Project: CMSIS DSP Library
emilmont 1:fdd22bb7aa52 8 * Title: arm_fir_q7.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 1:fdd22bb7aa52 10 * Description: Q7 FIR filter processing function.
emilmont 1:fdd22bb7aa52 11 *
emilmont 1:fdd22bb7aa52 12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
emilmont 1:fdd22bb7aa52 13 *
mbed_official 3:7a284390b0ce 14 * Redistribution and use in source and binary forms, with or without
mbed_official 3:7a284390b0ce 15 * modification, are permitted provided that the following conditions
mbed_official 3:7a284390b0ce 16 * are met:
mbed_official 3:7a284390b0ce 17 * - Redistributions of source code must retain the above copyright
mbed_official 3:7a284390b0ce 18 * notice, this list of conditions and the following disclaimer.
mbed_official 3:7a284390b0ce 19 * - Redistributions in binary form must reproduce the above copyright
mbed_official 3:7a284390b0ce 20 * notice, this list of conditions and the following disclaimer in
mbed_official 3:7a284390b0ce 21 * the documentation and/or other materials provided with the
mbed_official 3:7a284390b0ce 22 * distribution.
mbed_official 3:7a284390b0ce 23 * - Neither the name of ARM LIMITED nor the names of its contributors
mbed_official 3:7a284390b0ce 24 * may be used to endorse or promote products derived from this
mbed_official 3:7a284390b0ce 25 * software without specific prior written permission.
mbed_official 3:7a284390b0ce 26 *
mbed_official 3:7a284390b0ce 27 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
mbed_official 3:7a284390b0ce 28 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
mbed_official 3:7a284390b0ce 29 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
mbed_official 3:7a284390b0ce 30 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
mbed_official 3:7a284390b0ce 31 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
mbed_official 3:7a284390b0ce 32 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
mbed_official 3:7a284390b0ce 33 * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
mbed_official 3:7a284390b0ce 34 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
mbed_official 3:7a284390b0ce 35 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
mbed_official 3:7a284390b0ce 36 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
mbed_official 3:7a284390b0ce 37 * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
mbed_official 3:7a284390b0ce 38 * POSSIBILITY OF SUCH DAMAGE.
emilmont 1:fdd22bb7aa52 39 * -------------------------------------------------------------------- */
emilmont 1:fdd22bb7aa52 40
emilmont 1:fdd22bb7aa52 41 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 42
emilmont 1:fdd22bb7aa52 43 /**
emilmont 1:fdd22bb7aa52 44 * @ingroup groupFilters
emilmont 1:fdd22bb7aa52 45 */
emilmont 1:fdd22bb7aa52 46
emilmont 1:fdd22bb7aa52 47 /**
emilmont 1:fdd22bb7aa52 48 * @addtogroup FIR
emilmont 1:fdd22bb7aa52 49 * @{
emilmont 1:fdd22bb7aa52 50 */
emilmont 1:fdd22bb7aa52 51
emilmont 1:fdd22bb7aa52 52 /**
emilmont 1:fdd22bb7aa52 53 * @param[in] *S points to an instance of the Q7 FIR filter structure.
emilmont 1:fdd22bb7aa52 54 * @param[in] *pSrc points to the block of input data.
emilmont 1:fdd22bb7aa52 55 * @param[out] *pDst points to the block of output data.
emilmont 1:fdd22bb7aa52 56 * @param[in] blockSize number of samples to process per call.
emilmont 2:da51fb522205 57 * @return none.
emilmont 1:fdd22bb7aa52 58 *
emilmont 1:fdd22bb7aa52 59 * <b>Scaling and Overflow Behavior:</b>
emilmont 1:fdd22bb7aa52 60 * \par
emilmont 1:fdd22bb7aa52 61 * The function is implemented using a 32-bit internal accumulator.
emilmont 1:fdd22bb7aa52 62 * Both coefficients and state variables are represented in 1.7 format and multiplications yield a 2.14 result.
emilmont 1:fdd22bb7aa52 63 * The 2.14 intermediate results are accumulated in a 32-bit accumulator in 18.14 format.
emilmont 1:fdd22bb7aa52 64 * There is no risk of internal overflow with this approach and the full precision of intermediate multiplications is preserved.
emilmont 1:fdd22bb7aa52 65 * The accumulator is converted to 18.7 format by discarding the low 7 bits.
emilmont 1:fdd22bb7aa52 66 * Finally, the result is truncated to 1.7 format.
emilmont 1:fdd22bb7aa52 67 */
emilmont 1:fdd22bb7aa52 68
emilmont 1:fdd22bb7aa52 69 void arm_fir_q7(
emilmont 1:fdd22bb7aa52 70 const arm_fir_instance_q7 * S,
emilmont 1:fdd22bb7aa52 71 q7_t * pSrc,
emilmont 1:fdd22bb7aa52 72 q7_t * pDst,
emilmont 1:fdd22bb7aa52 73 uint32_t blockSize)
emilmont 1:fdd22bb7aa52 74 {
emilmont 1:fdd22bb7aa52 75
mbed_official 3:7a284390b0ce 76 #ifndef ARM_MATH_CM0_FAMILY
emilmont 1:fdd22bb7aa52 77
emilmont 1:fdd22bb7aa52 78 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 79
emilmont 1:fdd22bb7aa52 80 q7_t *pState = S->pState; /* State pointer */
emilmont 1:fdd22bb7aa52 81 q7_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emilmont 1:fdd22bb7aa52 82 q7_t *pStateCurnt; /* Points to the current sample of the state */
emilmont 1:fdd22bb7aa52 83 q7_t x0, x1, x2, x3; /* Temporary variables to hold state */
emilmont 1:fdd22bb7aa52 84 q7_t c0; /* Temporary variable to hold coefficient value */
emilmont 1:fdd22bb7aa52 85 q7_t *px; /* Temporary pointer for state */
emilmont 1:fdd22bb7aa52 86 q7_t *pb; /* Temporary pointer for coefficient buffer */
emilmont 1:fdd22bb7aa52 87 q31_t acc0, acc1, acc2, acc3; /* Accumulators */
emilmont 1:fdd22bb7aa52 88 uint32_t numTaps = S->numTaps; /* Number of filter coefficients in the filter */
emilmont 1:fdd22bb7aa52 89 uint32_t i, tapCnt, blkCnt; /* Loop counters */
emilmont 1:fdd22bb7aa52 90
emilmont 1:fdd22bb7aa52 91 /* S->pState points to state array which contains previous frame (numTaps - 1) samples */
emilmont 1:fdd22bb7aa52 92 /* pStateCurnt points to the location where the new input data should be written */
emilmont 1:fdd22bb7aa52 93 pStateCurnt = &(S->pState[(numTaps - 1u)]);
emilmont 1:fdd22bb7aa52 94
emilmont 1:fdd22bb7aa52 95 /* Apply loop unrolling and compute 4 output values simultaneously.
emilmont 1:fdd22bb7aa52 96 * The variables acc0 ... acc3 hold output values that are being computed:
emilmont 1:fdd22bb7aa52 97 *
emilmont 1:fdd22bb7aa52 98 * acc0 = b[numTaps-1] * x[n-numTaps-1] + b[numTaps-2] * x[n-numTaps-2] + b[numTaps-3] * x[n-numTaps-3] +...+ b[0] * x[0]
emilmont 1:fdd22bb7aa52 99 * acc1 = b[numTaps-1] * x[n-numTaps] + b[numTaps-2] * x[n-numTaps-1] + b[numTaps-3] * x[n-numTaps-2] +...+ b[0] * x[1]
emilmont 1:fdd22bb7aa52 100 * acc2 = b[numTaps-1] * x[n-numTaps+1] + b[numTaps-2] * x[n-numTaps] + b[numTaps-3] * x[n-numTaps-1] +...+ b[0] * x[2]
emilmont 1:fdd22bb7aa52 101 * acc3 = b[numTaps-1] * x[n-numTaps+2] + b[numTaps-2] * x[n-numTaps+1] + b[numTaps-3] * x[n-numTaps] +...+ b[0] * x[3]
emilmont 1:fdd22bb7aa52 102 */
emilmont 1:fdd22bb7aa52 103 blkCnt = blockSize >> 2;
emilmont 1:fdd22bb7aa52 104
emilmont 1:fdd22bb7aa52 105 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
emilmont 1:fdd22bb7aa52 106 ** a second loop below computes the remaining 1 to 3 samples. */
emilmont 1:fdd22bb7aa52 107 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 108 {
emilmont 1:fdd22bb7aa52 109 /* Copy four new input samples into the state buffer */
emilmont 1:fdd22bb7aa52 110 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 111 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 112 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 113 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 114
emilmont 1:fdd22bb7aa52 115 /* Set all accumulators to zero */
emilmont 1:fdd22bb7aa52 116 acc0 = 0;
emilmont 1:fdd22bb7aa52 117 acc1 = 0;
emilmont 1:fdd22bb7aa52 118 acc2 = 0;
emilmont 1:fdd22bb7aa52 119 acc3 = 0;
emilmont 1:fdd22bb7aa52 120
emilmont 1:fdd22bb7aa52 121 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 122 px = pState;
emilmont 1:fdd22bb7aa52 123
emilmont 1:fdd22bb7aa52 124 /* Initialize coefficient pointer */
emilmont 1:fdd22bb7aa52 125 pb = pCoeffs;
emilmont 1:fdd22bb7aa52 126
emilmont 1:fdd22bb7aa52 127 /* Read the first three samples from the state buffer:
emilmont 1:fdd22bb7aa52 128 * x[n-numTaps], x[n-numTaps-1], x[n-numTaps-2] */
emilmont 1:fdd22bb7aa52 129 x0 = *(px++);
emilmont 1:fdd22bb7aa52 130 x1 = *(px++);
emilmont 1:fdd22bb7aa52 131 x2 = *(px++);
emilmont 1:fdd22bb7aa52 132
emilmont 1:fdd22bb7aa52 133 /* Loop unrolling. Process 4 taps at a time. */
emilmont 1:fdd22bb7aa52 134 tapCnt = numTaps >> 2;
emilmont 1:fdd22bb7aa52 135 i = tapCnt;
emilmont 1:fdd22bb7aa52 136
emilmont 1:fdd22bb7aa52 137 while(i > 0u)
emilmont 1:fdd22bb7aa52 138 {
emilmont 1:fdd22bb7aa52 139 /* Read the b[numTaps] coefficient */
mbed_official 5:3762170b6d4d 140 c0 = *pb;
emilmont 1:fdd22bb7aa52 141
emilmont 1:fdd22bb7aa52 142 /* Read x[n-numTaps-3] sample */
mbed_official 5:3762170b6d4d 143 x3 = *px;
mbed_official 5:3762170b6d4d 144
emilmont 1:fdd22bb7aa52 145 /* acc0 += b[numTaps] * x[n-numTaps] */
emilmont 1:fdd22bb7aa52 146 acc0 += ((q15_t) x0 * c0);
emilmont 1:fdd22bb7aa52 147
emilmont 1:fdd22bb7aa52 148 /* acc1 += b[numTaps] * x[n-numTaps-1] */
emilmont 1:fdd22bb7aa52 149 acc1 += ((q15_t) x1 * c0);
emilmont 1:fdd22bb7aa52 150
emilmont 1:fdd22bb7aa52 151 /* acc2 += b[numTaps] * x[n-numTaps-2] */
emilmont 1:fdd22bb7aa52 152 acc2 += ((q15_t) x2 * c0);
emilmont 1:fdd22bb7aa52 153
emilmont 1:fdd22bb7aa52 154 /* acc3 += b[numTaps] * x[n-numTaps-3] */
emilmont 1:fdd22bb7aa52 155 acc3 += ((q15_t) x3 * c0);
emilmont 1:fdd22bb7aa52 156
emilmont 1:fdd22bb7aa52 157 /* Read the b[numTaps-1] coefficient */
mbed_official 5:3762170b6d4d 158 c0 = *(pb + 1u);
emilmont 1:fdd22bb7aa52 159
emilmont 1:fdd22bb7aa52 160 /* Read x[n-numTaps-4] sample */
mbed_official 5:3762170b6d4d 161 x0 = *(px + 1u);
emilmont 1:fdd22bb7aa52 162
emilmont 1:fdd22bb7aa52 163 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 164 acc0 += ((q15_t) x1 * c0);
emilmont 1:fdd22bb7aa52 165 acc1 += ((q15_t) x2 * c0);
emilmont 1:fdd22bb7aa52 166 acc2 += ((q15_t) x3 * c0);
emilmont 1:fdd22bb7aa52 167 acc3 += ((q15_t) x0 * c0);
emilmont 1:fdd22bb7aa52 168
emilmont 1:fdd22bb7aa52 169 /* Read the b[numTaps-2] coefficient */
mbed_official 5:3762170b6d4d 170 c0 = *(pb + 2u);
emilmont 1:fdd22bb7aa52 171
emilmont 1:fdd22bb7aa52 172 /* Read x[n-numTaps-5] sample */
mbed_official 5:3762170b6d4d 173 x1 = *(px + 2u);
emilmont 1:fdd22bb7aa52 174
emilmont 1:fdd22bb7aa52 175 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 176 acc0 += ((q15_t) x2 * c0);
emilmont 1:fdd22bb7aa52 177 acc1 += ((q15_t) x3 * c0);
emilmont 1:fdd22bb7aa52 178 acc2 += ((q15_t) x0 * c0);
emilmont 1:fdd22bb7aa52 179 acc3 += ((q15_t) x1 * c0);
mbed_official 5:3762170b6d4d 180
emilmont 1:fdd22bb7aa52 181 /* Read the b[numTaps-3] coefficients */
mbed_official 5:3762170b6d4d 182 c0 = *(pb + 3u);
emilmont 1:fdd22bb7aa52 183
emilmont 1:fdd22bb7aa52 184 /* Read x[n-numTaps-6] sample */
mbed_official 5:3762170b6d4d 185 x2 = *(px + 3u);
mbed_official 5:3762170b6d4d 186
emilmont 1:fdd22bb7aa52 187 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 188 acc0 += ((q15_t) x3 * c0);
emilmont 1:fdd22bb7aa52 189 acc1 += ((q15_t) x0 * c0);
emilmont 1:fdd22bb7aa52 190 acc2 += ((q15_t) x1 * c0);
emilmont 1:fdd22bb7aa52 191 acc3 += ((q15_t) x2 * c0);
mbed_official 5:3762170b6d4d 192
mbed_official 5:3762170b6d4d 193 /* update coefficient pointer */
mbed_official 5:3762170b6d4d 194 pb += 4u;
mbed_official 5:3762170b6d4d 195 px += 4u;
mbed_official 5:3762170b6d4d 196
mbed_official 5:3762170b6d4d 197 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 198 i--;
emilmont 1:fdd22bb7aa52 199 }
emilmont 1:fdd22bb7aa52 200
emilmont 1:fdd22bb7aa52 201 /* If the filter length is not a multiple of 4, compute the remaining filter taps */
emilmont 1:fdd22bb7aa52 202
emilmont 1:fdd22bb7aa52 203 i = numTaps - (tapCnt * 4u);
emilmont 1:fdd22bb7aa52 204 while(i > 0u)
emilmont 1:fdd22bb7aa52 205 {
emilmont 1:fdd22bb7aa52 206 /* Read coefficients */
emilmont 1:fdd22bb7aa52 207 c0 = *(pb++);
emilmont 1:fdd22bb7aa52 208
emilmont 1:fdd22bb7aa52 209 /* Fetch 1 state variable */
emilmont 1:fdd22bb7aa52 210 x3 = *(px++);
emilmont 1:fdd22bb7aa52 211
emilmont 1:fdd22bb7aa52 212 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 213 acc0 += ((q15_t) x0 * c0);
emilmont 1:fdd22bb7aa52 214 acc1 += ((q15_t) x1 * c0);
emilmont 1:fdd22bb7aa52 215 acc2 += ((q15_t) x2 * c0);
emilmont 1:fdd22bb7aa52 216 acc3 += ((q15_t) x3 * c0);
emilmont 1:fdd22bb7aa52 217
emilmont 1:fdd22bb7aa52 218 /* Reuse the present sample states for next sample */
emilmont 1:fdd22bb7aa52 219 x0 = x1;
emilmont 1:fdd22bb7aa52 220 x1 = x2;
emilmont 1:fdd22bb7aa52 221 x2 = x3;
emilmont 1:fdd22bb7aa52 222
emilmont 1:fdd22bb7aa52 223 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 224 i--;
emilmont 1:fdd22bb7aa52 225 }
emilmont 1:fdd22bb7aa52 226
emilmont 1:fdd22bb7aa52 227 /* Advance the state pointer by 4 to process the next group of 4 samples */
emilmont 1:fdd22bb7aa52 228 pState = pState + 4;
emilmont 1:fdd22bb7aa52 229
emilmont 1:fdd22bb7aa52 230 /* The results in the 4 accumulators are in 2.62 format. Convert to 1.31
emilmont 1:fdd22bb7aa52 231 ** Then store the 4 outputs in the destination buffer. */
emilmont 1:fdd22bb7aa52 232 acc0 = __SSAT((acc0 >> 7u), 8);
emilmont 1:fdd22bb7aa52 233 *pDst++ = acc0;
emilmont 1:fdd22bb7aa52 234 acc1 = __SSAT((acc1 >> 7u), 8);
emilmont 1:fdd22bb7aa52 235 *pDst++ = acc1;
emilmont 1:fdd22bb7aa52 236 acc2 = __SSAT((acc2 >> 7u), 8);
emilmont 1:fdd22bb7aa52 237 *pDst++ = acc2;
emilmont 1:fdd22bb7aa52 238 acc3 = __SSAT((acc3 >> 7u), 8);
emilmont 1:fdd22bb7aa52 239 *pDst++ = acc3;
emilmont 1:fdd22bb7aa52 240
emilmont 1:fdd22bb7aa52 241 /* Decrement the samples loop counter */
emilmont 1:fdd22bb7aa52 242 blkCnt--;
emilmont 1:fdd22bb7aa52 243 }
emilmont 1:fdd22bb7aa52 244
emilmont 1:fdd22bb7aa52 245
emilmont 1:fdd22bb7aa52 246 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
emilmont 1:fdd22bb7aa52 247 ** No loop unrolling is used. */
emilmont 1:fdd22bb7aa52 248 blkCnt = blockSize % 4u;
emilmont 1:fdd22bb7aa52 249
emilmont 1:fdd22bb7aa52 250 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 251 {
emilmont 1:fdd22bb7aa52 252 /* Copy one sample at a time into state buffer */
emilmont 1:fdd22bb7aa52 253 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 254
emilmont 1:fdd22bb7aa52 255 /* Set the accumulator to zero */
emilmont 1:fdd22bb7aa52 256 acc0 = 0;
emilmont 1:fdd22bb7aa52 257
emilmont 1:fdd22bb7aa52 258 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 259 px = pState;
emilmont 1:fdd22bb7aa52 260
emilmont 1:fdd22bb7aa52 261 /* Initialize Coefficient pointer */
emilmont 1:fdd22bb7aa52 262 pb = (pCoeffs);
emilmont 1:fdd22bb7aa52 263
emilmont 1:fdd22bb7aa52 264 i = numTaps;
emilmont 1:fdd22bb7aa52 265
emilmont 1:fdd22bb7aa52 266 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 267 do
emilmont 1:fdd22bb7aa52 268 {
emilmont 1:fdd22bb7aa52 269 acc0 += (q15_t) * (px++) * (*(pb++));
emilmont 1:fdd22bb7aa52 270 i--;
emilmont 1:fdd22bb7aa52 271 } while(i > 0u);
emilmont 1:fdd22bb7aa52 272
emilmont 1:fdd22bb7aa52 273 /* The result is in 2.14 format. Convert to 1.7
emilmont 1:fdd22bb7aa52 274 ** Then store the output in the destination buffer. */
emilmont 1:fdd22bb7aa52 275 *pDst++ = __SSAT((acc0 >> 7u), 8);
emilmont 1:fdd22bb7aa52 276
emilmont 1:fdd22bb7aa52 277 /* Advance state pointer by 1 for the next sample */
emilmont 1:fdd22bb7aa52 278 pState = pState + 1;
emilmont 1:fdd22bb7aa52 279
emilmont 1:fdd22bb7aa52 280 /* Decrement the samples loop counter */
emilmont 1:fdd22bb7aa52 281 blkCnt--;
emilmont 1:fdd22bb7aa52 282 }
emilmont 1:fdd22bb7aa52 283
emilmont 1:fdd22bb7aa52 284 /* Processing is complete.
emilmont 1:fdd22bb7aa52 285 ** Now copy the last numTaps - 1 samples to the satrt of the state buffer.
emilmont 1:fdd22bb7aa52 286 ** This prepares the state buffer for the next function call. */
emilmont 1:fdd22bb7aa52 287
emilmont 1:fdd22bb7aa52 288 /* Points to the start of the state buffer */
emilmont 1:fdd22bb7aa52 289 pStateCurnt = S->pState;
emilmont 1:fdd22bb7aa52 290
emilmont 1:fdd22bb7aa52 291 tapCnt = (numTaps - 1u) >> 2u;
emilmont 1:fdd22bb7aa52 292
emilmont 1:fdd22bb7aa52 293 /* copy data */
emilmont 1:fdd22bb7aa52 294 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 295 {
emilmont 1:fdd22bb7aa52 296 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 297 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 298 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 299 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 300
emilmont 1:fdd22bb7aa52 301 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 302 tapCnt--;
emilmont 1:fdd22bb7aa52 303 }
emilmont 1:fdd22bb7aa52 304
emilmont 1:fdd22bb7aa52 305 /* Calculate remaining number of copies */
emilmont 1:fdd22bb7aa52 306 tapCnt = (numTaps - 1u) % 0x4u;
emilmont 1:fdd22bb7aa52 307
emilmont 1:fdd22bb7aa52 308 /* Copy the remaining q31_t data */
emilmont 1:fdd22bb7aa52 309 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 310 {
emilmont 1:fdd22bb7aa52 311 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 312
emilmont 1:fdd22bb7aa52 313 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 314 tapCnt--;
emilmont 1:fdd22bb7aa52 315 }
emilmont 1:fdd22bb7aa52 316
emilmont 1:fdd22bb7aa52 317 #else
emilmont 1:fdd22bb7aa52 318
emilmont 1:fdd22bb7aa52 319 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 320
emilmont 1:fdd22bb7aa52 321 uint32_t numTaps = S->numTaps; /* Number of taps in the filter */
emilmont 1:fdd22bb7aa52 322 uint32_t i, blkCnt; /* Loop counters */
emilmont 1:fdd22bb7aa52 323 q7_t *pState = S->pState; /* State pointer */
emilmont 1:fdd22bb7aa52 324 q7_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emilmont 1:fdd22bb7aa52 325 q7_t *px, *pb; /* Temporary pointers to state and coeff */
emilmont 1:fdd22bb7aa52 326 q31_t acc = 0; /* Accumlator */
emilmont 1:fdd22bb7aa52 327 q7_t *pStateCurnt; /* Points to the current sample of the state */
emilmont 1:fdd22bb7aa52 328
emilmont 1:fdd22bb7aa52 329
emilmont 1:fdd22bb7aa52 330 /* S->pState points to state array which contains previous frame (numTaps - 1) samples */
emilmont 1:fdd22bb7aa52 331 /* pStateCurnt points to the location where the new input data should be written */
emilmont 1:fdd22bb7aa52 332 pStateCurnt = S->pState + (numTaps - 1u);
emilmont 1:fdd22bb7aa52 333
emilmont 1:fdd22bb7aa52 334 /* Initialize blkCnt with blockSize */
emilmont 1:fdd22bb7aa52 335 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 336
emilmont 1:fdd22bb7aa52 337 /* Perform filtering upto BlockSize - BlockSize%4 */
emilmont 1:fdd22bb7aa52 338 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 339 {
emilmont 1:fdd22bb7aa52 340 /* Copy one sample at a time into state buffer */
emilmont 1:fdd22bb7aa52 341 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 342
emilmont 1:fdd22bb7aa52 343 /* Set accumulator to zero */
emilmont 1:fdd22bb7aa52 344 acc = 0;
emilmont 1:fdd22bb7aa52 345
emilmont 1:fdd22bb7aa52 346 /* Initialize state pointer of type q7 */
emilmont 1:fdd22bb7aa52 347 px = pState;
emilmont 1:fdd22bb7aa52 348
emilmont 1:fdd22bb7aa52 349 /* Initialize coeff pointer of type q7 */
emilmont 1:fdd22bb7aa52 350 pb = pCoeffs;
emilmont 1:fdd22bb7aa52 351
emilmont 1:fdd22bb7aa52 352
emilmont 1:fdd22bb7aa52 353 i = numTaps;
emilmont 1:fdd22bb7aa52 354
emilmont 1:fdd22bb7aa52 355 while(i > 0u)
emilmont 1:fdd22bb7aa52 356 {
emilmont 1:fdd22bb7aa52 357 /* acc = b[numTaps-1] * x[n-numTaps-1] + b[numTaps-2] * x[n-numTaps-2] + b[numTaps-3] * x[n-numTaps-3] +...+ b[0] * x[0] */
emilmont 1:fdd22bb7aa52 358 acc += (q15_t) * px++ * *pb++;
emilmont 1:fdd22bb7aa52 359 i--;
emilmont 1:fdd22bb7aa52 360 }
emilmont 1:fdd22bb7aa52 361
emilmont 1:fdd22bb7aa52 362 /* Store the 1.7 format filter output in destination buffer */
emilmont 1:fdd22bb7aa52 363 *pDst++ = (q7_t) __SSAT((acc >> 7), 8);
emilmont 1:fdd22bb7aa52 364
emilmont 1:fdd22bb7aa52 365 /* Advance the state pointer by 1 to process the next sample */
emilmont 1:fdd22bb7aa52 366 pState = pState + 1;
emilmont 1:fdd22bb7aa52 367
emilmont 1:fdd22bb7aa52 368 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 369 blkCnt--;
emilmont 1:fdd22bb7aa52 370 }
emilmont 1:fdd22bb7aa52 371
emilmont 1:fdd22bb7aa52 372 /* Processing is complete.
emilmont 1:fdd22bb7aa52 373 ** Now copy the last numTaps - 1 samples to the satrt of the state buffer.
emilmont 1:fdd22bb7aa52 374 ** This prepares the state buffer for the next function call. */
emilmont 1:fdd22bb7aa52 375
emilmont 1:fdd22bb7aa52 376
emilmont 1:fdd22bb7aa52 377 /* Points to the start of the state buffer */
emilmont 1:fdd22bb7aa52 378 pStateCurnt = S->pState;
emilmont 1:fdd22bb7aa52 379
emilmont 1:fdd22bb7aa52 380
emilmont 1:fdd22bb7aa52 381 /* Copy numTaps number of values */
emilmont 1:fdd22bb7aa52 382 i = (numTaps - 1u);
emilmont 1:fdd22bb7aa52 383
emilmont 1:fdd22bb7aa52 384 /* Copy q7_t data */
emilmont 1:fdd22bb7aa52 385 while(i > 0u)
emilmont 1:fdd22bb7aa52 386 {
emilmont 1:fdd22bb7aa52 387 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 388 i--;
emilmont 1:fdd22bb7aa52 389 }
emilmont 1:fdd22bb7aa52 390
mbed_official 3:7a284390b0ce 391 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
emilmont 1:fdd22bb7aa52 392
emilmont 1:fdd22bb7aa52 393 }
emilmont 1:fdd22bb7aa52 394
emilmont 1:fdd22bb7aa52 395 /**
emilmont 1:fdd22bb7aa52 396 * @} end of FIR group
emilmont 1:fdd22bb7aa52 397 */