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_q31.c@2:da51fb522205, 2013-05-30 (annotated)

Committer:
emilmont
Date:
Thu May 30 17:10:11 2013 +0100
Revision:
2:da51fb522205
Parent:
1:fdd22bb7aa52
Child:
3:7a284390b0ce
Keep "cmsis-dsp" module in synch with its source

Who changed what in which revision?

UserRevisionLine numberNew contents of line
emilmont 1:fdd22bb7aa52 1 /* ----------------------------------------------------------------------
emilmont 1:fdd22bb7aa52 2 * Copyright (C) 2010 ARM Limited. All rights reserved.
emilmont 1:fdd22bb7aa52 3 *
emilmont 1:fdd22bb7aa52 4 * $Date: 15. February 2012
emilmont 2:da51fb522205 5 * $Revision: V1.1.0
emilmont 1:fdd22bb7aa52 6 *
emilmont 2:da51fb522205 7 * Project: CMSIS DSP Library
emilmont 2:da51fb522205 8 * Title: arm_fir_q31.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 2:da51fb522205 10 * Description: Q31 FIR filter processing function.
emilmont 1:fdd22bb7aa52 11 *
emilmont 1:fdd22bb7aa52 12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
emilmont 1:fdd22bb7aa52 13 *
emilmont 1:fdd22bb7aa52 14 * Version 1.1.0 2012/02/15
emilmont 1:fdd22bb7aa52 15 * Updated with more optimizations, bug fixes and minor API changes.
emilmont 1:fdd22bb7aa52 16 *
emilmont 1:fdd22bb7aa52 17 * Version 1.0.10 2011/7/15
emilmont 1:fdd22bb7aa52 18 * Big Endian support added and Merged M0 and M3/M4 Source code.
emilmont 1:fdd22bb7aa52 19 *
emilmont 1:fdd22bb7aa52 20 * Version 1.0.3 2010/11/29
emilmont 1:fdd22bb7aa52 21 * Re-organized the CMSIS folders and updated documentation.
emilmont 1:fdd22bb7aa52 22 *
emilmont 1:fdd22bb7aa52 23 * Version 1.0.2 2010/11/11
emilmont 1:fdd22bb7aa52 24 * Documentation updated.
emilmont 1:fdd22bb7aa52 25 *
emilmont 1:fdd22bb7aa52 26 * Version 1.0.1 2010/10/05
emilmont 1:fdd22bb7aa52 27 * Production release and review comments incorporated.
emilmont 1:fdd22bb7aa52 28 *
emilmont 1:fdd22bb7aa52 29 * Version 1.0.0 2010/09/20
emilmont 1:fdd22bb7aa52 30 * Production release and review comments incorporated.
emilmont 1:fdd22bb7aa52 31 *
emilmont 1:fdd22bb7aa52 32 * Version 0.0.5 2010/04/26
emilmont 2:da51fb522205 33 * incorporated review comments and updated with latest CMSIS layer
emilmont 1:fdd22bb7aa52 34 *
emilmont 1:fdd22bb7aa52 35 * Version 0.0.3 2010/03/10
emilmont 1:fdd22bb7aa52 36 * Initial version
emilmont 1:fdd22bb7aa52 37 * -------------------------------------------------------------------- */
emilmont 1:fdd22bb7aa52 38
emilmont 1:fdd22bb7aa52 39 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 40
emilmont 1:fdd22bb7aa52 41 /**
emilmont 1:fdd22bb7aa52 42 * @ingroup groupFilters
emilmont 1:fdd22bb7aa52 43 */
emilmont 1:fdd22bb7aa52 44
emilmont 1:fdd22bb7aa52 45 /**
emilmont 1:fdd22bb7aa52 46 * @addtogroup FIR
emilmont 1:fdd22bb7aa52 47 * @{
emilmont 1:fdd22bb7aa52 48 */
emilmont 1:fdd22bb7aa52 49
emilmont 1:fdd22bb7aa52 50 /**
emilmont 1:fdd22bb7aa52 51 * @param[in] *S points to an instance of the Q31 FIR filter structure.
emilmont 1:fdd22bb7aa52 52 * @param[in] *pSrc points to the block of input data.
emilmont 1:fdd22bb7aa52 53 * @param[out] *pDst points to the block of output data.
emilmont 1:fdd22bb7aa52 54 * @param[in] blockSize number of samples to process per call.
emilmont 1:fdd22bb7aa52 55 * @return none.
emilmont 1:fdd22bb7aa52 56 *
emilmont 1:fdd22bb7aa52 57 * @details
emilmont 1:fdd22bb7aa52 58 * <b>Scaling and Overflow Behavior:</b>
emilmont 1:fdd22bb7aa52 59 * \par
emilmont 1:fdd22bb7aa52 60 * The function is implemented using an internal 64-bit accumulator.
emilmont 1:fdd22bb7aa52 61 * The accumulator has a 2.62 format and maintains full precision of the intermediate multiplication results but provides only a single guard bit.
emilmont 1:fdd22bb7aa52 62 * Thus, if the accumulator result overflows it wraps around rather than clip.
emilmont 1:fdd22bb7aa52 63 * In order to avoid overflows completely the input signal must be scaled down by log2(numTaps) bits.
emilmont 1:fdd22bb7aa52 64 * After all multiply-accumulates are performed, the 2.62 accumulator is right shifted by 31 bits and saturated to 1.31 format to yield the final result.
emilmont 1:fdd22bb7aa52 65 *
emilmont 1:fdd22bb7aa52 66 * \par
emilmont 1:fdd22bb7aa52 67 * Refer to the function <code>arm_fir_fast_q31()</code> for a faster but less precise implementation of this filter for Cortex-M3 and Cortex-M4.
emilmont 1:fdd22bb7aa52 68 */
emilmont 1:fdd22bb7aa52 69
emilmont 1:fdd22bb7aa52 70 void arm_fir_q31(
emilmont 1:fdd22bb7aa52 71 const arm_fir_instance_q31 * S,
emilmont 1:fdd22bb7aa52 72 q31_t * pSrc,
emilmont 1:fdd22bb7aa52 73 q31_t * pDst,
emilmont 1:fdd22bb7aa52 74 uint32_t blockSize)
emilmont 1:fdd22bb7aa52 75 {
emilmont 1:fdd22bb7aa52 76 q31_t *pState = S->pState; /* State pointer */
emilmont 1:fdd22bb7aa52 77 q31_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emilmont 1:fdd22bb7aa52 78 q31_t *pStateCurnt; /* Points to the current sample of the state */
emilmont 1:fdd22bb7aa52 79
emilmont 1:fdd22bb7aa52 80
emilmont 1:fdd22bb7aa52 81 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 82
emilmont 1:fdd22bb7aa52 83 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 84
emilmont 1:fdd22bb7aa52 85 q31_t x0, x1, x2; /* Temporary variables to hold state */
emilmont 1:fdd22bb7aa52 86 q31_t c0; /* Temporary variable to hold coefficient value */
emilmont 1:fdd22bb7aa52 87 q31_t *px; /* Temporary pointer for state */
emilmont 1:fdd22bb7aa52 88 q31_t *pb; /* Temporary pointer for coefficient buffer */
emilmont 1:fdd22bb7aa52 89 q63_t acc0, acc1, acc2; /* Accumulators */
emilmont 1:fdd22bb7aa52 90 uint32_t numTaps = S->numTaps; /* Number of filter coefficients in the filter */
emilmont 1:fdd22bb7aa52 91 uint32_t i, tapCnt, blkCnt, tapCntN3; /* Loop counters */
emilmont 1:fdd22bb7aa52 92
emilmont 1:fdd22bb7aa52 93 /* S->pState points to state array which contains previous frame (numTaps - 1) samples */
emilmont 1:fdd22bb7aa52 94 /* pStateCurnt points to the location where the new input data should be written */
emilmont 1:fdd22bb7aa52 95 pStateCurnt = &(S->pState[(numTaps - 1u)]);
emilmont 1:fdd22bb7aa52 96
emilmont 1:fdd22bb7aa52 97 /* Apply loop unrolling and compute 4 output values simultaneously.
emilmont 1:fdd22bb7aa52 98 * The variables acc0 ... acc3 hold output values that are being computed:
emilmont 1:fdd22bb7aa52 99 *
emilmont 1:fdd22bb7aa52 100 * 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 101 * 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 102 * 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 103 * 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 104 */
emilmont 1:fdd22bb7aa52 105 blkCnt = blockSize / 3;
emilmont 1:fdd22bb7aa52 106 blockSize = blockSize - (3 * blkCnt);
emilmont 1:fdd22bb7aa52 107
emilmont 1:fdd22bb7aa52 108 tapCnt = numTaps / 3;
emilmont 1:fdd22bb7aa52 109 tapCntN3 = numTaps - (3 * tapCnt);
emilmont 1:fdd22bb7aa52 110
emilmont 1:fdd22bb7aa52 111 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
emilmont 1:fdd22bb7aa52 112 ** a second loop below computes the remaining 1 to 3 samples. */
emilmont 1:fdd22bb7aa52 113 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 114 {
emilmont 1:fdd22bb7aa52 115 /* Copy three new input samples into the state buffer */
emilmont 1:fdd22bb7aa52 116 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 117 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 118 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 119
emilmont 1:fdd22bb7aa52 120 /* Set all accumulators to zero */
emilmont 1:fdd22bb7aa52 121 acc0 = 0;
emilmont 1:fdd22bb7aa52 122 acc1 = 0;
emilmont 1:fdd22bb7aa52 123 acc2 = 0;
emilmont 1:fdd22bb7aa52 124
emilmont 1:fdd22bb7aa52 125 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 126 px = pState;
emilmont 1:fdd22bb7aa52 127
emilmont 1:fdd22bb7aa52 128 /* Initialize coefficient pointer */
emilmont 1:fdd22bb7aa52 129 pb = pCoeffs;
emilmont 1:fdd22bb7aa52 130
emilmont 1:fdd22bb7aa52 131 /* Read the first two samples from the state buffer:
emilmont 1:fdd22bb7aa52 132 * x[n-numTaps], x[n-numTaps-1] */
emilmont 1:fdd22bb7aa52 133 x0 = *(px++);
emilmont 1:fdd22bb7aa52 134 x1 = *(px++);
emilmont 1:fdd22bb7aa52 135
emilmont 1:fdd22bb7aa52 136 /* Loop unrolling. Process 3 taps at a time. */
emilmont 1:fdd22bb7aa52 137 i = tapCnt;
emilmont 1:fdd22bb7aa52 138
emilmont 1:fdd22bb7aa52 139 while(i > 0u)
emilmont 1:fdd22bb7aa52 140 {
emilmont 1:fdd22bb7aa52 141 /* Read the b[numTaps] coefficient */
emilmont 1:fdd22bb7aa52 142 c0 = *pb;
emilmont 1:fdd22bb7aa52 143
emilmont 1:fdd22bb7aa52 144 /* Read x[n-numTaps-2] sample */
emilmont 1:fdd22bb7aa52 145 x2 = *(px++);
emilmont 1:fdd22bb7aa52 146
emilmont 1:fdd22bb7aa52 147 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 148 acc0 += ((q63_t) x0 * c0);
emilmont 1:fdd22bb7aa52 149 acc1 += ((q63_t) x1 * c0);
emilmont 1:fdd22bb7aa52 150 acc2 += ((q63_t) x2 * c0);
emilmont 1:fdd22bb7aa52 151
emilmont 1:fdd22bb7aa52 152 /* Read the coefficient and state */
emilmont 1:fdd22bb7aa52 153 c0 = *(pb + 1u);
emilmont 1:fdd22bb7aa52 154 x0 = *(px++);
emilmont 1:fdd22bb7aa52 155
emilmont 1:fdd22bb7aa52 156 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 157 acc0 += ((q63_t) x1 * c0);
emilmont 1:fdd22bb7aa52 158 acc1 += ((q63_t) x2 * c0);
emilmont 1:fdd22bb7aa52 159 acc2 += ((q63_t) x0 * c0);
emilmont 1:fdd22bb7aa52 160
emilmont 1:fdd22bb7aa52 161 /* Read the coefficient and state */
emilmont 1:fdd22bb7aa52 162 c0 = *(pb + 2u);
emilmont 1:fdd22bb7aa52 163 x1 = *(px++);
emilmont 1:fdd22bb7aa52 164
emilmont 1:fdd22bb7aa52 165 /* update coefficient pointer */
emilmont 1:fdd22bb7aa52 166 pb += 3u;
emilmont 1:fdd22bb7aa52 167
emilmont 1:fdd22bb7aa52 168 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 169 acc0 += ((q63_t) x2 * c0);
emilmont 1:fdd22bb7aa52 170 acc1 += ((q63_t) x0 * c0);
emilmont 1:fdd22bb7aa52 171 acc2 += ((q63_t) x1 * c0);
emilmont 1:fdd22bb7aa52 172
emilmont 1:fdd22bb7aa52 173 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 174 i--;
emilmont 1:fdd22bb7aa52 175 }
emilmont 1:fdd22bb7aa52 176
emilmont 1:fdd22bb7aa52 177 /* If the filter length is not a multiple of 3, compute the remaining filter taps */
emilmont 1:fdd22bb7aa52 178
emilmont 1:fdd22bb7aa52 179 i = tapCntN3;
emilmont 1:fdd22bb7aa52 180
emilmont 1:fdd22bb7aa52 181 while(i > 0u)
emilmont 1:fdd22bb7aa52 182 {
emilmont 1:fdd22bb7aa52 183 /* Read coefficients */
emilmont 1:fdd22bb7aa52 184 c0 = *(pb++);
emilmont 1:fdd22bb7aa52 185
emilmont 1:fdd22bb7aa52 186 /* Fetch 1 state variable */
emilmont 1:fdd22bb7aa52 187 x2 = *(px++);
emilmont 1:fdd22bb7aa52 188
emilmont 1:fdd22bb7aa52 189 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 190 acc0 += ((q63_t) x0 * c0);
emilmont 1:fdd22bb7aa52 191 acc1 += ((q63_t) x1 * c0);
emilmont 1:fdd22bb7aa52 192 acc2 += ((q63_t) x2 * c0);
emilmont 1:fdd22bb7aa52 193
emilmont 1:fdd22bb7aa52 194 /* Reuse the present sample states for next sample */
emilmont 1:fdd22bb7aa52 195 x0 = x1;
emilmont 1:fdd22bb7aa52 196 x1 = x2;
emilmont 1:fdd22bb7aa52 197
emilmont 1:fdd22bb7aa52 198 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 199 i--;
emilmont 1:fdd22bb7aa52 200 }
emilmont 1:fdd22bb7aa52 201
emilmont 1:fdd22bb7aa52 202 /* Advance the state pointer by 3 to process the next group of 3 samples */
emilmont 1:fdd22bb7aa52 203 pState = pState + 3;
emilmont 1:fdd22bb7aa52 204
emilmont 1:fdd22bb7aa52 205 /* The results in the 3 accumulators are in 2.30 format. Convert to 1.31
emilmont 1:fdd22bb7aa52 206 ** Then store the 3 outputs in the destination buffer. */
emilmont 1:fdd22bb7aa52 207 *pDst++ = (q31_t) (acc0 >> 31u);
emilmont 1:fdd22bb7aa52 208 *pDst++ = (q31_t) (acc1 >> 31u);
emilmont 1:fdd22bb7aa52 209 *pDst++ = (q31_t) (acc2 >> 31u);
emilmont 1:fdd22bb7aa52 210
emilmont 1:fdd22bb7aa52 211 /* Decrement the samples loop counter */
emilmont 1:fdd22bb7aa52 212 blkCnt--;
emilmont 1:fdd22bb7aa52 213 }
emilmont 1:fdd22bb7aa52 214
emilmont 1:fdd22bb7aa52 215 /* If the blockSize is not a multiple of 3, compute any remaining output samples here.
emilmont 1:fdd22bb7aa52 216 ** No loop unrolling is used. */
emilmont 1:fdd22bb7aa52 217
emilmont 1:fdd22bb7aa52 218 while(blockSize > 0u)
emilmont 1:fdd22bb7aa52 219 {
emilmont 1:fdd22bb7aa52 220 /* Copy one sample at a time into state buffer */
emilmont 1:fdd22bb7aa52 221 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 222
emilmont 1:fdd22bb7aa52 223 /* Set the accumulator to zero */
emilmont 1:fdd22bb7aa52 224 acc0 = 0;
emilmont 1:fdd22bb7aa52 225
emilmont 1:fdd22bb7aa52 226 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 227 px = pState;
emilmont 1:fdd22bb7aa52 228
emilmont 1:fdd22bb7aa52 229 /* Initialize Coefficient pointer */
emilmont 1:fdd22bb7aa52 230 pb = (pCoeffs);
emilmont 1:fdd22bb7aa52 231
emilmont 1:fdd22bb7aa52 232 i = numTaps;
emilmont 1:fdd22bb7aa52 233
emilmont 1:fdd22bb7aa52 234 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 235 do
emilmont 1:fdd22bb7aa52 236 {
emilmont 1:fdd22bb7aa52 237 acc0 += (q63_t) * (px++) * (*(pb++));
emilmont 1:fdd22bb7aa52 238 i--;
emilmont 1:fdd22bb7aa52 239 } while(i > 0u);
emilmont 1:fdd22bb7aa52 240
emilmont 1:fdd22bb7aa52 241 /* The result is in 2.62 format. Convert to 1.31
emilmont 1:fdd22bb7aa52 242 ** Then store the output in the destination buffer. */
emilmont 1:fdd22bb7aa52 243 *pDst++ = (q31_t) (acc0 >> 31u);
emilmont 1:fdd22bb7aa52 244
emilmont 1:fdd22bb7aa52 245 /* Advance state pointer by 1 for the next sample */
emilmont 1:fdd22bb7aa52 246 pState = pState + 1;
emilmont 1:fdd22bb7aa52 247
emilmont 1:fdd22bb7aa52 248 /* Decrement the samples loop counter */
emilmont 1:fdd22bb7aa52 249 blockSize--;
emilmont 1:fdd22bb7aa52 250 }
emilmont 1:fdd22bb7aa52 251
emilmont 1:fdd22bb7aa52 252 /* Processing is complete.
emilmont 1:fdd22bb7aa52 253 ** Now copy the last numTaps - 1 samples to the satrt of the state buffer.
emilmont 1:fdd22bb7aa52 254 ** This prepares the state buffer for the next function call. */
emilmont 1:fdd22bb7aa52 255
emilmont 1:fdd22bb7aa52 256 /* Points to the start of the state buffer */
emilmont 1:fdd22bb7aa52 257 pStateCurnt = S->pState;
emilmont 1:fdd22bb7aa52 258
emilmont 1:fdd22bb7aa52 259 tapCnt = (numTaps - 1u) >> 2u;
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 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 266 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 267 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 268
emilmont 1:fdd22bb7aa52 269 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 270 tapCnt--;
emilmont 1:fdd22bb7aa52 271 }
emilmont 1:fdd22bb7aa52 272
emilmont 1:fdd22bb7aa52 273 /* Calculate remaining number of copies */
emilmont 1:fdd22bb7aa52 274 tapCnt = (numTaps - 1u) % 0x4u;
emilmont 1:fdd22bb7aa52 275
emilmont 1:fdd22bb7aa52 276 /* Copy the remaining q31_t data */
emilmont 1:fdd22bb7aa52 277 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 278 {
emilmont 1:fdd22bb7aa52 279 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 280
emilmont 1:fdd22bb7aa52 281 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 282 tapCnt--;
emilmont 1:fdd22bb7aa52 283 }
emilmont 1:fdd22bb7aa52 284
emilmont 1:fdd22bb7aa52 285 #else
emilmont 1:fdd22bb7aa52 286
emilmont 1:fdd22bb7aa52 287 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 288
emilmont 1:fdd22bb7aa52 289 q31_t *px; /* Temporary pointer for state */
emilmont 1:fdd22bb7aa52 290 q31_t *pb; /* Temporary pointer for coefficient buffer */
emilmont 1:fdd22bb7aa52 291 q63_t acc; /* Accumulator */
emilmont 1:fdd22bb7aa52 292 uint32_t numTaps = S->numTaps; /* Length of the filter */
emilmont 1:fdd22bb7aa52 293 uint32_t i, tapCnt, blkCnt; /* Loop counters */
emilmont 1:fdd22bb7aa52 294
emilmont 1:fdd22bb7aa52 295 /* S->pState buffer contains previous frame (numTaps - 1) samples */
emilmont 1:fdd22bb7aa52 296 /* pStateCurnt points to the location where the new input data should be written */
emilmont 1:fdd22bb7aa52 297 pStateCurnt = &(S->pState[(numTaps - 1u)]);
emilmont 1:fdd22bb7aa52 298
emilmont 1:fdd22bb7aa52 299 /* Initialize blkCnt with blockSize */
emilmont 1:fdd22bb7aa52 300 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 301
emilmont 1:fdd22bb7aa52 302 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 303 {
emilmont 1:fdd22bb7aa52 304 /* Copy one sample at a time into state buffer */
emilmont 1:fdd22bb7aa52 305 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 306
emilmont 1:fdd22bb7aa52 307 /* Set the accumulator to zero */
emilmont 1:fdd22bb7aa52 308 acc = 0;
emilmont 1:fdd22bb7aa52 309
emilmont 1:fdd22bb7aa52 310 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 311 px = pState;
emilmont 1:fdd22bb7aa52 312
emilmont 1:fdd22bb7aa52 313 /* Initialize Coefficient pointer */
emilmont 1:fdd22bb7aa52 314 pb = pCoeffs;
emilmont 1:fdd22bb7aa52 315
emilmont 1:fdd22bb7aa52 316 i = numTaps;
emilmont 1:fdd22bb7aa52 317
emilmont 1:fdd22bb7aa52 318 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 319 do
emilmont 1:fdd22bb7aa52 320 {
emilmont 1:fdd22bb7aa52 321 /* 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 322 acc += (q63_t) * px++ * *pb++;
emilmont 1:fdd22bb7aa52 323 i--;
emilmont 1:fdd22bb7aa52 324 } while(i > 0u);
emilmont 1:fdd22bb7aa52 325
emilmont 1:fdd22bb7aa52 326 /* The result is in 2.62 format. Convert to 1.31
emilmont 1:fdd22bb7aa52 327 ** Then store the output in the destination buffer. */
emilmont 1:fdd22bb7aa52 328 *pDst++ = (q31_t) (acc >> 31u);
emilmont 1:fdd22bb7aa52 329
emilmont 1:fdd22bb7aa52 330 /* Advance state pointer by 1 for the next sample */
emilmont 1:fdd22bb7aa52 331 pState = pState + 1;
emilmont 1:fdd22bb7aa52 332
emilmont 1:fdd22bb7aa52 333 /* Decrement the samples loop counter */
emilmont 1:fdd22bb7aa52 334 blkCnt--;
emilmont 1:fdd22bb7aa52 335 }
emilmont 1:fdd22bb7aa52 336
emilmont 1:fdd22bb7aa52 337 /* Processing is complete.
emilmont 1:fdd22bb7aa52 338 ** Now copy the last numTaps - 1 samples to the starting of the state buffer.
emilmont 1:fdd22bb7aa52 339 ** This prepares the state buffer for the next function call. */
emilmont 1:fdd22bb7aa52 340
emilmont 1:fdd22bb7aa52 341 /* Points to the start of the state buffer */
emilmont 1:fdd22bb7aa52 342 pStateCurnt = S->pState;
emilmont 1:fdd22bb7aa52 343
emilmont 1:fdd22bb7aa52 344 /* Copy numTaps number of values */
emilmont 1:fdd22bb7aa52 345 tapCnt = numTaps - 1u;
emilmont 1:fdd22bb7aa52 346
emilmont 1:fdd22bb7aa52 347 /* Copy the data */
emilmont 1:fdd22bb7aa52 348 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 349 {
emilmont 1:fdd22bb7aa52 350 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 351
emilmont 1:fdd22bb7aa52 352 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 353 tapCnt--;
emilmont 1:fdd22bb7aa52 354 }
emilmont 1:fdd22bb7aa52 355
emilmont 1:fdd22bb7aa52 356
emilmont 1:fdd22bb7aa52 357 #endif /* #ifndef ARM_MATH_CM0 */
emilmont 1:fdd22bb7aa52 358
emilmont 1:fdd22bb7aa52 359 }
emilmont 1:fdd22bb7aa52 360
emilmont 1:fdd22bb7aa52 361 /**
emilmont 1:fdd22bb7aa52 362 * @} end of FIR group
emilmont 1:fdd22bb7aa52 363 */