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_lms_q15.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 2:da51fb522205 8 * Title: arm_lms_q15.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 2:da51fb522205 10 * Description: Processing function for the Q15 LMS filter.
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 * @ingroup groupFilters
emilmont 1:fdd22bb7aa52 44 */
emilmont 1:fdd22bb7aa52 45
emilmont 1:fdd22bb7aa52 46 /**
emilmont 1:fdd22bb7aa52 47 * @addtogroup LMS
emilmont 1:fdd22bb7aa52 48 * @{
emilmont 1:fdd22bb7aa52 49 */
emilmont 1:fdd22bb7aa52 50
emilmont 1:fdd22bb7aa52 51 /**
emilmont 1:fdd22bb7aa52 52 * @brief Processing function for Q15 LMS filter.
emilmont 1:fdd22bb7aa52 53 * @param[in] *S points to an instance of the Q15 LMS filter structure.
emilmont 1:fdd22bb7aa52 54 * @param[in] *pSrc points to the block of input data.
emilmont 1:fdd22bb7aa52 55 * @param[in] *pRef points to the block of reference data.
emilmont 1:fdd22bb7aa52 56 * @param[out] *pOut points to the block of output data.
emilmont 1:fdd22bb7aa52 57 * @param[out] *pErr points to the block of error data.
emilmont 1:fdd22bb7aa52 58 * @param[in] blockSize number of samples to process.
emilmont 1:fdd22bb7aa52 59 * @return none.
emilmont 1:fdd22bb7aa52 60 *
emilmont 1:fdd22bb7aa52 61 * \par Scaling and Overflow Behavior:
emilmont 1:fdd22bb7aa52 62 * The function is implemented using a 64-bit internal accumulator.
emilmont 1:fdd22bb7aa52 63 * Both coefficients and state variables are represented in 1.15 format and multiplications yield a 2.30 result.
emilmont 1:fdd22bb7aa52 64 * The 2.30 intermediate results are accumulated in a 64-bit accumulator in 34.30 format.
emilmont 1:fdd22bb7aa52 65 * There is no risk of internal overflow with this approach and the full precision of intermediate multiplications is preserved.
emilmont 1:fdd22bb7aa52 66 * After all additions have been performed, the accumulator is truncated to 34.15 format by discarding low 15 bits.
emilmont 1:fdd22bb7aa52 67 * Lastly, the accumulator is saturated to yield a result in 1.15 format.
emilmont 1:fdd22bb7aa52 68 *
emilmont 1:fdd22bb7aa52 69 * \par
emilmont 2:da51fb522205 70 * In this filter, filter coefficients are updated for each sample and the updation of filter cofficients are saturted.
emilmont 1:fdd22bb7aa52 71 *
emilmont 1:fdd22bb7aa52 72 */
emilmont 1:fdd22bb7aa52 73
emilmont 1:fdd22bb7aa52 74 void arm_lms_q15(
emilmont 1:fdd22bb7aa52 75 const arm_lms_instance_q15 * S,
emilmont 1:fdd22bb7aa52 76 q15_t * pSrc,
emilmont 1:fdd22bb7aa52 77 q15_t * pRef,
emilmont 1:fdd22bb7aa52 78 q15_t * pOut,
emilmont 1:fdd22bb7aa52 79 q15_t * pErr,
emilmont 1:fdd22bb7aa52 80 uint32_t blockSize)
emilmont 1:fdd22bb7aa52 81 {
emilmont 1:fdd22bb7aa52 82 q15_t *pState = S->pState; /* State pointer */
emilmont 1:fdd22bb7aa52 83 uint32_t numTaps = S->numTaps; /* Number of filter coefficients in the filter */
emilmont 1:fdd22bb7aa52 84 q15_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emilmont 1:fdd22bb7aa52 85 q15_t *pStateCurnt; /* Points to the current sample of the state */
emilmont 1:fdd22bb7aa52 86 q15_t mu = S->mu; /* Adaptive factor */
emilmont 1:fdd22bb7aa52 87 q15_t *px; /* Temporary pointer for state */
emilmont 1:fdd22bb7aa52 88 q15_t *pb; /* Temporary pointer for coefficient buffer */
emilmont 1:fdd22bb7aa52 89 uint32_t tapCnt, blkCnt; /* Loop counters */
emilmont 1:fdd22bb7aa52 90 q63_t acc; /* Accumulator */
emilmont 1:fdd22bb7aa52 91 q15_t e = 0; /* error of data sample */
emilmont 1:fdd22bb7aa52 92 q15_t alpha; /* Intermediate constant for taps update */
mbed_official 5:3762170b6d4d 93 q31_t coef; /* Teporary variable for coefficient */
emilmont 1:fdd22bb7aa52 94 q31_t acc_l, acc_h;
emilmont 1:fdd22bb7aa52 95 int32_t lShift = (15 - (int32_t) S->postShift); /* Post shift */
emilmont 1:fdd22bb7aa52 96 int32_t uShift = (32 - lShift);
emilmont 1:fdd22bb7aa52 97
emilmont 1:fdd22bb7aa52 98
mbed_official 3:7a284390b0ce 99 #ifndef ARM_MATH_CM0_FAMILY
emilmont 1:fdd22bb7aa52 100
emilmont 1:fdd22bb7aa52 101 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 102
emilmont 1:fdd22bb7aa52 103
emilmont 1:fdd22bb7aa52 104 /* S->pState points to buffer which contains previous frame (numTaps - 1) samples */
emilmont 1:fdd22bb7aa52 105 /* pStateCurnt points to the location where the new input data should be written */
emilmont 1:fdd22bb7aa52 106 pStateCurnt = &(S->pState[(numTaps - 1u)]);
emilmont 1:fdd22bb7aa52 107
emilmont 1:fdd22bb7aa52 108 /* Initializing blkCnt with blockSize */
emilmont 1:fdd22bb7aa52 109 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 110
emilmont 1:fdd22bb7aa52 111 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 112 {
emilmont 1:fdd22bb7aa52 113 /* Copy the new input sample into the state buffer */
emilmont 1:fdd22bb7aa52 114 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 115
emilmont 1:fdd22bb7aa52 116 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 117 px = pState;
emilmont 1:fdd22bb7aa52 118
emilmont 1:fdd22bb7aa52 119 /* Initialize coefficient pointer */
emilmont 1:fdd22bb7aa52 120 pb = pCoeffs;
emilmont 1:fdd22bb7aa52 121
emilmont 1:fdd22bb7aa52 122 /* Set the accumulator to zero */
emilmont 1:fdd22bb7aa52 123 acc = 0;
emilmont 1:fdd22bb7aa52 124
emilmont 1:fdd22bb7aa52 125 /* Loop unrolling. Process 4 taps at a time. */
emilmont 1:fdd22bb7aa52 126 tapCnt = numTaps >> 2u;
emilmont 1:fdd22bb7aa52 127
emilmont 1:fdd22bb7aa52 128 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 129 {
emilmont 1:fdd22bb7aa52 130 /* acc += b[N] * x[n-N] + b[N-1] * x[n-N-1] */
emilmont 1:fdd22bb7aa52 131 /* Perform the multiply-accumulate */
emilmont 1:fdd22bb7aa52 132 #ifndef UNALIGNED_SUPPORT_DISABLE
emilmont 1:fdd22bb7aa52 133
emilmont 1:fdd22bb7aa52 134 acc = __SMLALD(*__SIMD32(px)++, (*__SIMD32(pb)++), acc);
emilmont 1:fdd22bb7aa52 135 acc = __SMLALD(*__SIMD32(px)++, (*__SIMD32(pb)++), acc);
emilmont 1:fdd22bb7aa52 136
emilmont 1:fdd22bb7aa52 137 #else
emilmont 1:fdd22bb7aa52 138
emilmont 1:fdd22bb7aa52 139 acc += (q63_t) (((q31_t) (*px++) * (*pb++)));
emilmont 1:fdd22bb7aa52 140 acc += (q63_t) (((q31_t) (*px++) * (*pb++)));
emilmont 1:fdd22bb7aa52 141 acc += (q63_t) (((q31_t) (*px++) * (*pb++)));
emilmont 1:fdd22bb7aa52 142 acc += (q63_t) (((q31_t) (*px++) * (*pb++)));
emilmont 1:fdd22bb7aa52 143
emilmont 1:fdd22bb7aa52 144
emilmont 2:da51fb522205 145 #endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */
emilmont 1:fdd22bb7aa52 146
emilmont 1:fdd22bb7aa52 147 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 148 tapCnt--;
emilmont 1:fdd22bb7aa52 149 }
emilmont 1:fdd22bb7aa52 150
emilmont 1:fdd22bb7aa52 151 /* If the filter length is not a multiple of 4, compute the remaining filter taps */
emilmont 1:fdd22bb7aa52 152 tapCnt = numTaps % 0x4u;
emilmont 1:fdd22bb7aa52 153
emilmont 1:fdd22bb7aa52 154 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 155 {
emilmont 1:fdd22bb7aa52 156 /* Perform the multiply-accumulate */
emilmont 1:fdd22bb7aa52 157 acc += (q63_t) (((q31_t) (*px++) * (*pb++)));
emilmont 1:fdd22bb7aa52 158
emilmont 1:fdd22bb7aa52 159 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 160 tapCnt--;
emilmont 1:fdd22bb7aa52 161 }
emilmont 1:fdd22bb7aa52 162
emilmont 1:fdd22bb7aa52 163 /* Calc lower part of acc */
emilmont 1:fdd22bb7aa52 164 acc_l = acc & 0xffffffff;
emilmont 1:fdd22bb7aa52 165
emilmont 1:fdd22bb7aa52 166 /* Calc upper part of acc */
emilmont 1:fdd22bb7aa52 167 acc_h = (acc >> 32) & 0xffffffff;
emilmont 1:fdd22bb7aa52 168
emilmont 1:fdd22bb7aa52 169 /* Apply shift for lower part of acc and upper part of acc */
emilmont 1:fdd22bb7aa52 170 acc = (uint32_t) acc_l >> lShift | acc_h << uShift;
emilmont 1:fdd22bb7aa52 171
emilmont 1:fdd22bb7aa52 172 /* Converting the result to 1.15 format and saturate the output */
emilmont 1:fdd22bb7aa52 173 acc = __SSAT(acc, 16);
emilmont 1:fdd22bb7aa52 174
emilmont 1:fdd22bb7aa52 175 /* Store the result from accumulator into the destination buffer. */
emilmont 1:fdd22bb7aa52 176 *pOut++ = (q15_t) acc;
emilmont 1:fdd22bb7aa52 177
emilmont 1:fdd22bb7aa52 178 /* Compute and store error */
emilmont 1:fdd22bb7aa52 179 e = *pRef++ - (q15_t) acc;
emilmont 1:fdd22bb7aa52 180
emilmont 1:fdd22bb7aa52 181 *pErr++ = (q15_t) e;
emilmont 1:fdd22bb7aa52 182
emilmont 1:fdd22bb7aa52 183 /* Compute alpha i.e. intermediate constant for taps update */
emilmont 1:fdd22bb7aa52 184 alpha = (q15_t) (((q31_t) e * (mu)) >> 15);
emilmont 1:fdd22bb7aa52 185
emilmont 1:fdd22bb7aa52 186 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 187 /* Advance state pointer by 1 for the next sample */
emilmont 1:fdd22bb7aa52 188 px = pState++;
emilmont 1:fdd22bb7aa52 189
emilmont 1:fdd22bb7aa52 190 /* Initialize coefficient pointer */
emilmont 1:fdd22bb7aa52 191 pb = pCoeffs;
emilmont 1:fdd22bb7aa52 192
emilmont 1:fdd22bb7aa52 193 /* Loop unrolling. Process 4 taps at a time. */
emilmont 1:fdd22bb7aa52 194 tapCnt = numTaps >> 2u;
emilmont 1:fdd22bb7aa52 195
emilmont 1:fdd22bb7aa52 196 /* Update filter coefficients */
emilmont 1:fdd22bb7aa52 197 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 198 {
emilmont 1:fdd22bb7aa52 199 coef = (q31_t) * pb + (((q31_t) alpha * (*px++)) >> 15);
emilmont 1:fdd22bb7aa52 200 *pb++ = (q15_t) __SSAT((coef), 16);
emilmont 1:fdd22bb7aa52 201 coef = (q31_t) * pb + (((q31_t) alpha * (*px++)) >> 15);
emilmont 1:fdd22bb7aa52 202 *pb++ = (q15_t) __SSAT((coef), 16);
emilmont 1:fdd22bb7aa52 203 coef = (q31_t) * pb + (((q31_t) alpha * (*px++)) >> 15);
emilmont 1:fdd22bb7aa52 204 *pb++ = (q15_t) __SSAT((coef), 16);
emilmont 1:fdd22bb7aa52 205 coef = (q31_t) * pb + (((q31_t) alpha * (*px++)) >> 15);
emilmont 1:fdd22bb7aa52 206 *pb++ = (q15_t) __SSAT((coef), 16);
emilmont 1:fdd22bb7aa52 207
emilmont 1:fdd22bb7aa52 208 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 209 tapCnt--;
emilmont 1:fdd22bb7aa52 210 }
emilmont 1:fdd22bb7aa52 211
emilmont 1:fdd22bb7aa52 212 /* If the filter length is not a multiple of 4, compute the remaining filter taps */
emilmont 1:fdd22bb7aa52 213 tapCnt = numTaps % 0x4u;
emilmont 1:fdd22bb7aa52 214
emilmont 1:fdd22bb7aa52 215 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 216 {
emilmont 1:fdd22bb7aa52 217 /* Perform the multiply-accumulate */
emilmont 1:fdd22bb7aa52 218 coef = (q31_t) * pb + (((q31_t) alpha * (*px++)) >> 15);
emilmont 1:fdd22bb7aa52 219 *pb++ = (q15_t) __SSAT((coef), 16);
emilmont 1:fdd22bb7aa52 220
emilmont 1:fdd22bb7aa52 221 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 222 tapCnt--;
emilmont 1:fdd22bb7aa52 223 }
emilmont 1:fdd22bb7aa52 224
emilmont 1:fdd22bb7aa52 225 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 226 blkCnt--;
emilmont 1:fdd22bb7aa52 227
emilmont 1:fdd22bb7aa52 228 }
emilmont 1:fdd22bb7aa52 229
emilmont 1:fdd22bb7aa52 230 /* Processing is complete. Now copy the last numTaps - 1 samples to the
emilmont 1:fdd22bb7aa52 231 satrt of the state buffer. This prepares the state buffer for the
emilmont 1:fdd22bb7aa52 232 next function call. */
emilmont 1:fdd22bb7aa52 233
emilmont 1:fdd22bb7aa52 234 /* Points to the start of the pState buffer */
emilmont 1:fdd22bb7aa52 235 pStateCurnt = S->pState;
emilmont 1:fdd22bb7aa52 236
emilmont 1:fdd22bb7aa52 237 /* Calculation of count for copying integer writes */
emilmont 1:fdd22bb7aa52 238 tapCnt = (numTaps - 1u) >> 2;
emilmont 1:fdd22bb7aa52 239
emilmont 1:fdd22bb7aa52 240 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 241 {
emilmont 1:fdd22bb7aa52 242
emilmont 1:fdd22bb7aa52 243 #ifndef UNALIGNED_SUPPORT_DISABLE
emilmont 1:fdd22bb7aa52 244
emilmont 1:fdd22bb7aa52 245 *__SIMD32(pStateCurnt)++ = *__SIMD32(pState)++;
emilmont 1:fdd22bb7aa52 246 *__SIMD32(pStateCurnt)++ = *__SIMD32(pState)++;
emilmont 1:fdd22bb7aa52 247 #else
emilmont 1:fdd22bb7aa52 248 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 249 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 250 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 251 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 252 #endif
emilmont 1:fdd22bb7aa52 253
emilmont 1:fdd22bb7aa52 254 tapCnt--;
emilmont 1:fdd22bb7aa52 255
emilmont 1:fdd22bb7aa52 256 }
emilmont 1:fdd22bb7aa52 257
emilmont 1:fdd22bb7aa52 258 /* Calculation of count for remaining q15_t data */
emilmont 1:fdd22bb7aa52 259 tapCnt = (numTaps - 1u) % 0x4u;
emilmont 1:fdd22bb7aa52 260
emilmont 1:fdd22bb7aa52 261 /* copy data */
emilmont 1:fdd22bb7aa52 262 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 263 {
emilmont 1:fdd22bb7aa52 264 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 265
emilmont 1:fdd22bb7aa52 266 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 267 tapCnt--;
emilmont 1:fdd22bb7aa52 268 }
emilmont 1:fdd22bb7aa52 269
emilmont 1:fdd22bb7aa52 270 #else
emilmont 1:fdd22bb7aa52 271
emilmont 1:fdd22bb7aa52 272 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 273
emilmont 1:fdd22bb7aa52 274 /* S->pState points to buffer which contains previous frame (numTaps - 1) samples */
emilmont 1:fdd22bb7aa52 275 /* pStateCurnt points to the location where the new input data should be written */
emilmont 1:fdd22bb7aa52 276 pStateCurnt = &(S->pState[(numTaps - 1u)]);
emilmont 1:fdd22bb7aa52 277
emilmont 1:fdd22bb7aa52 278 /* Loop over blockSize number of values */
emilmont 1:fdd22bb7aa52 279 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 280
emilmont 1:fdd22bb7aa52 281 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 282 {
emilmont 1:fdd22bb7aa52 283 /* Copy the new input sample into the state buffer */
emilmont 1:fdd22bb7aa52 284 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 285
emilmont 1:fdd22bb7aa52 286 /* Initialize pState pointer */
emilmont 1:fdd22bb7aa52 287 px = pState;
emilmont 1:fdd22bb7aa52 288
emilmont 1:fdd22bb7aa52 289 /* Initialize pCoeffs pointer */
emilmont 1:fdd22bb7aa52 290 pb = pCoeffs;
emilmont 1:fdd22bb7aa52 291
emilmont 1:fdd22bb7aa52 292 /* Set the accumulator to zero */
emilmont 1:fdd22bb7aa52 293 acc = 0;
emilmont 1:fdd22bb7aa52 294
emilmont 1:fdd22bb7aa52 295 /* Loop over numTaps number of values */
emilmont 1:fdd22bb7aa52 296 tapCnt = numTaps;
emilmont 1:fdd22bb7aa52 297
emilmont 1:fdd22bb7aa52 298 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 299 {
emilmont 1:fdd22bb7aa52 300 /* Perform the multiply-accumulate */
emilmont 1:fdd22bb7aa52 301 acc += (q63_t) ((q31_t) (*px++) * (*pb++));
emilmont 1:fdd22bb7aa52 302
emilmont 1:fdd22bb7aa52 303 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 304 tapCnt--;
emilmont 1:fdd22bb7aa52 305 }
emilmont 1:fdd22bb7aa52 306
emilmont 1:fdd22bb7aa52 307 /* Calc lower part of acc */
emilmont 1:fdd22bb7aa52 308 acc_l = acc & 0xffffffff;
emilmont 1:fdd22bb7aa52 309
emilmont 1:fdd22bb7aa52 310 /* Calc upper part of acc */
emilmont 1:fdd22bb7aa52 311 acc_h = (acc >> 32) & 0xffffffff;
emilmont 1:fdd22bb7aa52 312
emilmont 1:fdd22bb7aa52 313 /* Apply shift for lower part of acc and upper part of acc */
emilmont 1:fdd22bb7aa52 314 acc = (uint32_t) acc_l >> lShift | acc_h << uShift;
emilmont 1:fdd22bb7aa52 315
emilmont 1:fdd22bb7aa52 316 /* Converting the result to 1.15 format and saturate the output */
emilmont 1:fdd22bb7aa52 317 acc = __SSAT(acc, 16);
emilmont 1:fdd22bb7aa52 318
emilmont 1:fdd22bb7aa52 319 /* Store the result from accumulator into the destination buffer. */
emilmont 1:fdd22bb7aa52 320 *pOut++ = (q15_t) acc;
emilmont 1:fdd22bb7aa52 321
emilmont 1:fdd22bb7aa52 322 /* Compute and store error */
emilmont 1:fdd22bb7aa52 323 e = *pRef++ - (q15_t) acc;
emilmont 1:fdd22bb7aa52 324
emilmont 1:fdd22bb7aa52 325 *pErr++ = (q15_t) e;
emilmont 1:fdd22bb7aa52 326
emilmont 1:fdd22bb7aa52 327 /* Compute alpha i.e. intermediate constant for taps update */
emilmont 1:fdd22bb7aa52 328 alpha = (q15_t) (((q31_t) e * (mu)) >> 15);
emilmont 1:fdd22bb7aa52 329
emilmont 1:fdd22bb7aa52 330 /* Initialize pState pointer */
emilmont 1:fdd22bb7aa52 331 /* Advance state pointer by 1 for the next sample */
emilmont 1:fdd22bb7aa52 332 px = pState++;
emilmont 1:fdd22bb7aa52 333
emilmont 1:fdd22bb7aa52 334 /* Initialize pCoeffs pointer */
emilmont 1:fdd22bb7aa52 335 pb = pCoeffs;
emilmont 1:fdd22bb7aa52 336
emilmont 1:fdd22bb7aa52 337 /* Loop over numTaps number of values */
emilmont 1:fdd22bb7aa52 338 tapCnt = numTaps;
emilmont 1:fdd22bb7aa52 339
emilmont 1:fdd22bb7aa52 340 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 341 {
emilmont 1:fdd22bb7aa52 342 /* Perform the multiply-accumulate */
mbed_official 5:3762170b6d4d 343 coef = (q31_t) * pb + (((q31_t) alpha * (*px++)) >> 15);
mbed_official 5:3762170b6d4d 344 *pb++ = (q15_t) __SSAT((coef), 16);
emilmont 1:fdd22bb7aa52 345
emilmont 1:fdd22bb7aa52 346 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 347 tapCnt--;
emilmont 1:fdd22bb7aa52 348 }
emilmont 1:fdd22bb7aa52 349
emilmont 1:fdd22bb7aa52 350 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 351 blkCnt--;
emilmont 1:fdd22bb7aa52 352
emilmont 1:fdd22bb7aa52 353 }
emilmont 1:fdd22bb7aa52 354
emilmont 1:fdd22bb7aa52 355 /* Processing is complete. Now copy the last numTaps - 1 samples to the
emilmont 1:fdd22bb7aa52 356 start of the state buffer. This prepares the state buffer for the
emilmont 1:fdd22bb7aa52 357 next function call. */
emilmont 1:fdd22bb7aa52 358
emilmont 1:fdd22bb7aa52 359 /* Points to the start of the pState buffer */
emilmont 1:fdd22bb7aa52 360 pStateCurnt = S->pState;
emilmont 1:fdd22bb7aa52 361
emilmont 1:fdd22bb7aa52 362 /* Copy (numTaps - 1u) samples */
emilmont 1:fdd22bb7aa52 363 tapCnt = (numTaps - 1u);
emilmont 1:fdd22bb7aa52 364
emilmont 1:fdd22bb7aa52 365 /* Copy the data */
emilmont 1:fdd22bb7aa52 366 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 367 {
emilmont 1:fdd22bb7aa52 368 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 369
emilmont 1:fdd22bb7aa52 370 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 371 tapCnt--;
emilmont 1:fdd22bb7aa52 372 }
emilmont 1:fdd22bb7aa52 373
mbed_official 3:7a284390b0ce 374 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
emilmont 1:fdd22bb7aa52 375
emilmont 1:fdd22bb7aa52 376 }
emilmont 1:fdd22bb7aa52 377
emilmont 1:fdd22bb7aa52 378 /**
emilmont 1:fdd22bb7aa52 379 * @} end of LMS group
emilmont 1:fdd22bb7aa52 380 */