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_sparse_f32.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_sparse_f32.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 2:da51fb522205 10 * Description: Floating-point sparse 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 0.0.7 2010/06/10
emilmont 1:fdd22bb7aa52 33 * Misra-C changes done
emilmont 1:fdd22bb7aa52 34 * ------------------------------------------------------------------- */
emilmont 1:fdd22bb7aa52 35 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 36
emilmont 1:fdd22bb7aa52 37 /**
emilmont 1:fdd22bb7aa52 38 * @ingroup groupFilters
emilmont 1:fdd22bb7aa52 39 */
emilmont 1:fdd22bb7aa52 40
emilmont 1:fdd22bb7aa52 41 /**
emilmont 1:fdd22bb7aa52 42 * @defgroup FIR_Sparse Finite Impulse Response (FIR) Sparse Filters
emilmont 1:fdd22bb7aa52 43 *
emilmont 1:fdd22bb7aa52 44 * This group of functions implements sparse FIR filters.
emilmont 1:fdd22bb7aa52 45 * Sparse FIR filters are equivalent to standard FIR filters except that most of the coefficients are equal to zero.
emilmont 1:fdd22bb7aa52 46 * Sparse filters are used for simulating reflections in communications and audio applications.
emilmont 1:fdd22bb7aa52 47 *
emilmont 1:fdd22bb7aa52 48 * There are separate functions for Q7, Q15, Q31, and floating-point data types.
emilmont 1:fdd22bb7aa52 49 * The functions operate on blocks of input and output data and each call to the function processes
emilmont 1:fdd22bb7aa52 50 * <code>blockSize</code> samples through the filter. <code>pSrc</code> and
emilmont 1:fdd22bb7aa52 51 * <code>pDst</code> points to input and output arrays respectively containing <code>blockSize</code> values.
emilmont 1:fdd22bb7aa52 52 *
emilmont 1:fdd22bb7aa52 53 * \par Algorithm:
emilmont 1:fdd22bb7aa52 54 * The sparse filter instant structure contains an array of tap indices <code>pTapDelay</code> which specifies the locations of the non-zero coefficients.
emilmont 1:fdd22bb7aa52 55 * This is in addition to the coefficient array <code>b</code>.
emilmont 1:fdd22bb7aa52 56 * The implementation essentially skips the multiplications by zero and leads to an efficient realization.
emilmont 1:fdd22bb7aa52 57 * <pre>
emilmont 1:fdd22bb7aa52 58 * y[n] = b[0] * x[n-pTapDelay[0]] + b[1] * x[n-pTapDelay[1]] + b[2] * x[n-pTapDelay[2]] + ...+ b[numTaps-1] * x[n-pTapDelay[numTaps-1]]
emilmont 1:fdd22bb7aa52 59 * </pre>
emilmont 1:fdd22bb7aa52 60 * \par
emilmont 1:fdd22bb7aa52 61 * \image html FIRSparse.gif "Sparse FIR filter. b[n] represents the filter coefficients"
emilmont 1:fdd22bb7aa52 62 * \par
emilmont 1:fdd22bb7aa52 63 * <code>pCoeffs</code> points to a coefficient array of size <code>numTaps</code>;
emilmont 1:fdd22bb7aa52 64 * <code>pTapDelay</code> points to an array of nonzero indices and is also of size <code>numTaps</code>;
emilmont 1:fdd22bb7aa52 65 * <code>pState</code> points to a state array of size <code>maxDelay + blockSize</code>, where
emilmont 1:fdd22bb7aa52 66 * <code>maxDelay</code> is the largest offset value that is ever used in the <code>pTapDelay</code> array.
emilmont 1:fdd22bb7aa52 67 * Some of the processing functions also require temporary working buffers.
emilmont 1:fdd22bb7aa52 68 *
emilmont 1:fdd22bb7aa52 69 * \par Instance Structure
emilmont 1:fdd22bb7aa52 70 * The coefficients and state variables for a filter are stored together in an instance data structure.
emilmont 1:fdd22bb7aa52 71 * A separate instance structure must be defined for each filter.
emilmont 1:fdd22bb7aa52 72 * Coefficient and offset arrays may be shared among several instances while state variable arrays cannot be shared.
emilmont 1:fdd22bb7aa52 73 * There are separate instance structure declarations for each of the 4 supported data types.
emilmont 1:fdd22bb7aa52 74 *
emilmont 1:fdd22bb7aa52 75 * \par Initialization Functions
emilmont 1:fdd22bb7aa52 76 * There is also an associated initialization function for each data type.
emilmont 1:fdd22bb7aa52 77 * The initialization function performs the following operations:
emilmont 1:fdd22bb7aa52 78 * - Sets the values of the internal structure fields.
emilmont 1:fdd22bb7aa52 79 * - Zeros out the values in the state buffer.
emilmont 1:fdd22bb7aa52 80 *
emilmont 1:fdd22bb7aa52 81 * \par
emilmont 1:fdd22bb7aa52 82 * Use of the initialization function is optional.
emilmont 1:fdd22bb7aa52 83 * However, if the initialization function is used, then the instance structure cannot be placed into a const data section.
emilmont 1:fdd22bb7aa52 84 * To place an instance structure into a const data section, the instance structure must be manually initialized.
emilmont 1:fdd22bb7aa52 85 * Set the values in the state buffer to zeros before static initialization.
emilmont 1:fdd22bb7aa52 86 * The code below statically initializes each of the 4 different data type filter instance structures
emilmont 1:fdd22bb7aa52 87 * <pre>
emilmont 1:fdd22bb7aa52 88 *arm_fir_sparse_instance_f32 S = {numTaps, 0, pState, pCoeffs, maxDelay, pTapDelay};
emilmont 1:fdd22bb7aa52 89 *arm_fir_sparse_instance_q31 S = {numTaps, 0, pState, pCoeffs, maxDelay, pTapDelay};
emilmont 1:fdd22bb7aa52 90 *arm_fir_sparse_instance_q15 S = {numTaps, 0, pState, pCoeffs, maxDelay, pTapDelay};
emilmont 1:fdd22bb7aa52 91 *arm_fir_sparse_instance_q7 S = {numTaps, 0, pState, pCoeffs, maxDelay, pTapDelay};
emilmont 1:fdd22bb7aa52 92 * </pre>
emilmont 1:fdd22bb7aa52 93 * \par
emilmont 1:fdd22bb7aa52 94 *
emilmont 1:fdd22bb7aa52 95 * \par Fixed-Point Behavior
emilmont 1:fdd22bb7aa52 96 * Care must be taken when using the fixed-point versions of the sparse FIR filter functions.
emilmont 1:fdd22bb7aa52 97 * In particular, the overflow and saturation behavior of the accumulator used in each function must be considered.
emilmont 1:fdd22bb7aa52 98 * Refer to the function specific documentation below for usage guidelines.
emilmont 1:fdd22bb7aa52 99 */
emilmont 1:fdd22bb7aa52 100
emilmont 1:fdd22bb7aa52 101 /**
emilmont 1:fdd22bb7aa52 102 * @addtogroup FIR_Sparse
emilmont 1:fdd22bb7aa52 103 * @{
emilmont 1:fdd22bb7aa52 104 */
emilmont 1:fdd22bb7aa52 105
emilmont 1:fdd22bb7aa52 106 /**
emilmont 1:fdd22bb7aa52 107 * @brief Processing function for the floating-point sparse FIR filter.
emilmont 1:fdd22bb7aa52 108 * @param[in] *S points to an instance of the floating-point sparse FIR structure.
emilmont 1:fdd22bb7aa52 109 * @param[in] *pSrc points to the block of input data.
emilmont 1:fdd22bb7aa52 110 * @param[out] *pDst points to the block of output data
emilmont 1:fdd22bb7aa52 111 * @param[in] *pScratchIn points to a temporary buffer of size blockSize.
emilmont 1:fdd22bb7aa52 112 * @param[in] blockSize number of input samples to process per call.
emilmont 1:fdd22bb7aa52 113 * @return none.
emilmont 1:fdd22bb7aa52 114 */
emilmont 1:fdd22bb7aa52 115
emilmont 1:fdd22bb7aa52 116 void arm_fir_sparse_f32(
emilmont 1:fdd22bb7aa52 117 arm_fir_sparse_instance_f32 * S,
emilmont 1:fdd22bb7aa52 118 float32_t * pSrc,
emilmont 1:fdd22bb7aa52 119 float32_t * pDst,
emilmont 1:fdd22bb7aa52 120 float32_t * pScratchIn,
emilmont 1:fdd22bb7aa52 121 uint32_t blockSize)
emilmont 1:fdd22bb7aa52 122 {
emilmont 1:fdd22bb7aa52 123
emilmont 1:fdd22bb7aa52 124 float32_t *pState = S->pState; /* State pointer */
emilmont 1:fdd22bb7aa52 125 float32_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emilmont 1:fdd22bb7aa52 126 float32_t *px; /* Scratch buffer pointer */
emilmont 1:fdd22bb7aa52 127 float32_t *py = pState; /* Temporary pointers for state buffer */
emilmont 1:fdd22bb7aa52 128 float32_t *pb = pScratchIn; /* Temporary pointers for scratch buffer */
emilmont 1:fdd22bb7aa52 129 float32_t *pOut; /* Destination pointer */
emilmont 1:fdd22bb7aa52 130 int32_t *pTapDelay = S->pTapDelay; /* Pointer to the array containing offset of the non-zero tap values. */
emilmont 1:fdd22bb7aa52 131 uint32_t delaySize = S->maxDelay + blockSize; /* state length */
emilmont 1:fdd22bb7aa52 132 uint16_t numTaps = S->numTaps; /* Number of filter coefficients in the filter */
emilmont 1:fdd22bb7aa52 133 int32_t readIndex; /* Read index of the state buffer */
emilmont 1:fdd22bb7aa52 134 uint32_t tapCnt, blkCnt; /* loop counters */
emilmont 1:fdd22bb7aa52 135 float32_t coeff = *pCoeffs++; /* Read the first coefficient value */
emilmont 1:fdd22bb7aa52 136
emilmont 1:fdd22bb7aa52 137
emilmont 1:fdd22bb7aa52 138
emilmont 1:fdd22bb7aa52 139 /* BlockSize of Input samples are copied into the state buffer */
emilmont 1:fdd22bb7aa52 140 /* StateIndex points to the starting position to write in the state buffer */
emilmont 1:fdd22bb7aa52 141 arm_circularWrite_f32((int32_t *) py, delaySize, &S->stateIndex, 1,
emilmont 1:fdd22bb7aa52 142 (int32_t *) pSrc, 1, blockSize);
emilmont 1:fdd22bb7aa52 143
emilmont 1:fdd22bb7aa52 144
emilmont 1:fdd22bb7aa52 145 /* Read Index, from where the state buffer should be read, is calculated. */
emilmont 1:fdd22bb7aa52 146 readIndex = ((int32_t) S->stateIndex - (int32_t) blockSize) - *pTapDelay++;
emilmont 1:fdd22bb7aa52 147
emilmont 1:fdd22bb7aa52 148 /* Wraparound of readIndex */
emilmont 1:fdd22bb7aa52 149 if(readIndex < 0)
emilmont 1:fdd22bb7aa52 150 {
emilmont 1:fdd22bb7aa52 151 readIndex += (int32_t) delaySize;
emilmont 1:fdd22bb7aa52 152 }
emilmont 1:fdd22bb7aa52 153
emilmont 1:fdd22bb7aa52 154 /* Working pointer for state buffer is updated */
emilmont 1:fdd22bb7aa52 155 py = pState;
emilmont 1:fdd22bb7aa52 156
emilmont 1:fdd22bb7aa52 157 /* blockSize samples are read from the state buffer */
emilmont 1:fdd22bb7aa52 158 arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
emilmont 1:fdd22bb7aa52 159 (int32_t *) pb, (int32_t *) pb, blockSize, 1,
emilmont 1:fdd22bb7aa52 160 blockSize);
emilmont 1:fdd22bb7aa52 161
emilmont 1:fdd22bb7aa52 162 /* Working pointer for the scratch buffer */
emilmont 1:fdd22bb7aa52 163 px = pb;
emilmont 1:fdd22bb7aa52 164
emilmont 1:fdd22bb7aa52 165 /* Working pointer for destination buffer */
emilmont 1:fdd22bb7aa52 166 pOut = pDst;
emilmont 1:fdd22bb7aa52 167
emilmont 1:fdd22bb7aa52 168
emilmont 1:fdd22bb7aa52 169 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 170
emilmont 1:fdd22bb7aa52 171 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 172
emilmont 1:fdd22bb7aa52 173 /* Loop over the blockSize. Unroll by a factor of 4.
emilmont 1:fdd22bb7aa52 174 * Compute 4 Multiplications at a time. */
emilmont 1:fdd22bb7aa52 175 blkCnt = blockSize >> 2u;
emilmont 1:fdd22bb7aa52 176
emilmont 1:fdd22bb7aa52 177 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 178 {
emilmont 1:fdd22bb7aa52 179 /* Perform Multiplications and store in destination buffer */
emilmont 1:fdd22bb7aa52 180 *pOut++ = *px++ * coeff;
emilmont 1:fdd22bb7aa52 181 *pOut++ = *px++ * coeff;
emilmont 1:fdd22bb7aa52 182 *pOut++ = *px++ * coeff;
emilmont 1:fdd22bb7aa52 183 *pOut++ = *px++ * coeff;
emilmont 1:fdd22bb7aa52 184
emilmont 1:fdd22bb7aa52 185 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 186 blkCnt--;
emilmont 1:fdd22bb7aa52 187 }
emilmont 1:fdd22bb7aa52 188
emilmont 1:fdd22bb7aa52 189 /* If the blockSize is not a multiple of 4,
emilmont 1:fdd22bb7aa52 190 * compute the remaining samples */
emilmont 1:fdd22bb7aa52 191 blkCnt = blockSize % 0x4u;
emilmont 1:fdd22bb7aa52 192
emilmont 1:fdd22bb7aa52 193 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 194 {
emilmont 1:fdd22bb7aa52 195 /* Perform Multiplications and store in destination buffer */
emilmont 1:fdd22bb7aa52 196 *pOut++ = *px++ * coeff;
emilmont 1:fdd22bb7aa52 197
emilmont 1:fdd22bb7aa52 198 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 199 blkCnt--;
emilmont 1:fdd22bb7aa52 200 }
emilmont 1:fdd22bb7aa52 201
emilmont 1:fdd22bb7aa52 202 /* Load the coefficient value and
emilmont 1:fdd22bb7aa52 203 * increment the coefficient buffer for the next set of state values */
emilmont 1:fdd22bb7aa52 204 coeff = *pCoeffs++;
emilmont 1:fdd22bb7aa52 205
emilmont 1:fdd22bb7aa52 206 /* Read Index, from where the state buffer should be read, is calculated. */
emilmont 1:fdd22bb7aa52 207 readIndex = ((int32_t) S->stateIndex - (int32_t) blockSize) - *pTapDelay++;
emilmont 1:fdd22bb7aa52 208
emilmont 1:fdd22bb7aa52 209 /* Wraparound of readIndex */
emilmont 1:fdd22bb7aa52 210 if(readIndex < 0)
emilmont 1:fdd22bb7aa52 211 {
emilmont 1:fdd22bb7aa52 212 readIndex += (int32_t) delaySize;
emilmont 1:fdd22bb7aa52 213 }
emilmont 1:fdd22bb7aa52 214
emilmont 1:fdd22bb7aa52 215 /* Loop over the number of taps. */
emilmont 1:fdd22bb7aa52 216 tapCnt = (uint32_t) numTaps - 1u;
emilmont 1:fdd22bb7aa52 217
emilmont 1:fdd22bb7aa52 218 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 219 {
emilmont 1:fdd22bb7aa52 220
emilmont 1:fdd22bb7aa52 221 /* Working pointer for state buffer is updated */
emilmont 1:fdd22bb7aa52 222 py = pState;
emilmont 1:fdd22bb7aa52 223
emilmont 1:fdd22bb7aa52 224 /* blockSize samples are read from the state buffer */
emilmont 1:fdd22bb7aa52 225 arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
emilmont 1:fdd22bb7aa52 226 (int32_t *) pb, (int32_t *) pb, blockSize, 1,
emilmont 1:fdd22bb7aa52 227 blockSize);
emilmont 1:fdd22bb7aa52 228
emilmont 1:fdd22bb7aa52 229 /* Working pointer for the scratch buffer */
emilmont 1:fdd22bb7aa52 230 px = pb;
emilmont 1:fdd22bb7aa52 231
emilmont 1:fdd22bb7aa52 232 /* Working pointer for destination buffer */
emilmont 1:fdd22bb7aa52 233 pOut = pDst;
emilmont 1:fdd22bb7aa52 234
emilmont 1:fdd22bb7aa52 235 /* Loop over the blockSize. Unroll by a factor of 4.
emilmont 1:fdd22bb7aa52 236 * Compute 4 MACS at a time. */
emilmont 1:fdd22bb7aa52 237 blkCnt = blockSize >> 2u;
emilmont 1:fdd22bb7aa52 238
emilmont 1:fdd22bb7aa52 239 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 240 {
emilmont 1:fdd22bb7aa52 241 /* Perform Multiply-Accumulate */
emilmont 1:fdd22bb7aa52 242 *pOut++ += *px++ * coeff;
emilmont 1:fdd22bb7aa52 243 *pOut++ += *px++ * coeff;
emilmont 1:fdd22bb7aa52 244 *pOut++ += *px++ * coeff;
emilmont 1:fdd22bb7aa52 245 *pOut++ += *px++ * coeff;
emilmont 1:fdd22bb7aa52 246
emilmont 1:fdd22bb7aa52 247 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 248 blkCnt--;
emilmont 1:fdd22bb7aa52 249 }
emilmont 1:fdd22bb7aa52 250
emilmont 1:fdd22bb7aa52 251 /* If the blockSize is not a multiple of 4,
emilmont 1:fdd22bb7aa52 252 * compute the remaining samples */
emilmont 1:fdd22bb7aa52 253 blkCnt = blockSize % 0x4u;
emilmont 1:fdd22bb7aa52 254
emilmont 1:fdd22bb7aa52 255 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 256 {
emilmont 1:fdd22bb7aa52 257 /* Perform Multiply-Accumulate */
emilmont 1:fdd22bb7aa52 258 *pOut++ += *px++ * coeff;
emilmont 1:fdd22bb7aa52 259
emilmont 1:fdd22bb7aa52 260 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 261 blkCnt--;
emilmont 1:fdd22bb7aa52 262 }
emilmont 1:fdd22bb7aa52 263
emilmont 1:fdd22bb7aa52 264 /* Load the coefficient value and
emilmont 1:fdd22bb7aa52 265 * increment the coefficient buffer for the next set of state values */
emilmont 1:fdd22bb7aa52 266 coeff = *pCoeffs++;
emilmont 1:fdd22bb7aa52 267
emilmont 1:fdd22bb7aa52 268 /* Read Index, from where the state buffer should be read, is calculated. */
emilmont 1:fdd22bb7aa52 269 readIndex = ((int32_t) S->stateIndex -
emilmont 1:fdd22bb7aa52 270 (int32_t) blockSize) - *pTapDelay++;
emilmont 1:fdd22bb7aa52 271
emilmont 1:fdd22bb7aa52 272 /* Wraparound of readIndex */
emilmont 1:fdd22bb7aa52 273 if(readIndex < 0)
emilmont 1:fdd22bb7aa52 274 {
emilmont 1:fdd22bb7aa52 275 readIndex += (int32_t) delaySize;
emilmont 1:fdd22bb7aa52 276 }
emilmont 1:fdd22bb7aa52 277
emilmont 1:fdd22bb7aa52 278 /* Decrement the tap loop counter */
emilmont 1:fdd22bb7aa52 279 tapCnt--;
emilmont 1:fdd22bb7aa52 280 }
emilmont 1:fdd22bb7aa52 281
emilmont 1:fdd22bb7aa52 282 #else
emilmont 1:fdd22bb7aa52 283
emilmont 1:fdd22bb7aa52 284 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 285
emilmont 1:fdd22bb7aa52 286 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 287
emilmont 1:fdd22bb7aa52 288 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 289 {
emilmont 1:fdd22bb7aa52 290 /* Perform Multiplications and store in destination buffer */
emilmont 1:fdd22bb7aa52 291 *pOut++ = *px++ * coeff;
emilmont 1:fdd22bb7aa52 292
emilmont 1:fdd22bb7aa52 293 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 294 blkCnt--;
emilmont 1:fdd22bb7aa52 295 }
emilmont 1:fdd22bb7aa52 296
emilmont 1:fdd22bb7aa52 297 /* Load the coefficient value and
emilmont 1:fdd22bb7aa52 298 * increment the coefficient buffer for the next set of state values */
emilmont 1:fdd22bb7aa52 299 coeff = *pCoeffs++;
emilmont 1:fdd22bb7aa52 300
emilmont 1:fdd22bb7aa52 301 /* Read Index, from where the state buffer should be read, is calculated. */
emilmont 1:fdd22bb7aa52 302 readIndex = ((int32_t) S->stateIndex - (int32_t) blockSize) - *pTapDelay++;
emilmont 1:fdd22bb7aa52 303
emilmont 1:fdd22bb7aa52 304 /* Wraparound of readIndex */
emilmont 1:fdd22bb7aa52 305 if(readIndex < 0)
emilmont 1:fdd22bb7aa52 306 {
emilmont 1:fdd22bb7aa52 307 readIndex += (int32_t) delaySize;
emilmont 1:fdd22bb7aa52 308 }
emilmont 1:fdd22bb7aa52 309
emilmont 1:fdd22bb7aa52 310 /* Loop over the number of taps. */
emilmont 1:fdd22bb7aa52 311 tapCnt = (uint32_t) numTaps - 1u;
emilmont 1:fdd22bb7aa52 312
emilmont 1:fdd22bb7aa52 313 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 314 {
emilmont 1:fdd22bb7aa52 315
emilmont 1:fdd22bb7aa52 316 /* Working pointer for state buffer is updated */
emilmont 1:fdd22bb7aa52 317 py = pState;
emilmont 1:fdd22bb7aa52 318
emilmont 1:fdd22bb7aa52 319 /* blockSize samples are read from the state buffer */
emilmont 1:fdd22bb7aa52 320 arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
emilmont 1:fdd22bb7aa52 321 (int32_t *) pb, (int32_t *) pb, blockSize, 1,
emilmont 1:fdd22bb7aa52 322 blockSize);
emilmont 1:fdd22bb7aa52 323
emilmont 1:fdd22bb7aa52 324 /* Working pointer for the scratch buffer */
emilmont 1:fdd22bb7aa52 325 px = pb;
emilmont 1:fdd22bb7aa52 326
emilmont 1:fdd22bb7aa52 327 /* Working pointer for destination buffer */
emilmont 1:fdd22bb7aa52 328 pOut = pDst;
emilmont 1:fdd22bb7aa52 329
emilmont 1:fdd22bb7aa52 330 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 331
emilmont 1:fdd22bb7aa52 332 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 333 {
emilmont 1:fdd22bb7aa52 334 /* Perform Multiply-Accumulate */
emilmont 1:fdd22bb7aa52 335 *pOut++ += *px++ * coeff;
emilmont 1:fdd22bb7aa52 336
emilmont 1:fdd22bb7aa52 337 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 338 blkCnt--;
emilmont 1:fdd22bb7aa52 339 }
emilmont 1:fdd22bb7aa52 340
emilmont 1:fdd22bb7aa52 341 /* Load the coefficient value and
emilmont 1:fdd22bb7aa52 342 * increment the coefficient buffer for the next set of state values */
emilmont 1:fdd22bb7aa52 343 coeff = *pCoeffs++;
emilmont 1:fdd22bb7aa52 344
emilmont 1:fdd22bb7aa52 345 /* Read Index, from where the state buffer should be read, is calculated. */
emilmont 1:fdd22bb7aa52 346 readIndex =
emilmont 1:fdd22bb7aa52 347 ((int32_t) S->stateIndex - (int32_t) blockSize) - *pTapDelay++;
emilmont 1:fdd22bb7aa52 348
emilmont 1:fdd22bb7aa52 349 /* Wraparound of readIndex */
emilmont 1:fdd22bb7aa52 350 if(readIndex < 0)
emilmont 1:fdd22bb7aa52 351 {
emilmont 1:fdd22bb7aa52 352 readIndex += (int32_t) delaySize;
emilmont 1:fdd22bb7aa52 353 }
emilmont 1:fdd22bb7aa52 354
emilmont 1:fdd22bb7aa52 355 /* Decrement the tap loop counter */
emilmont 1:fdd22bb7aa52 356 tapCnt--;
emilmont 1:fdd22bb7aa52 357 }
emilmont 1:fdd22bb7aa52 358
emilmont 1:fdd22bb7aa52 359 #endif /* #ifndef ARM_MATH_CM0 */
emilmont 1:fdd22bb7aa52 360
emilmont 1:fdd22bb7aa52 361 }
emilmont 1:fdd22bb7aa52 362
emilmont 1:fdd22bb7aa52 363 /**
emilmont 1:fdd22bb7aa52 364 * @} end of FIR_Sparse group
emilmont 1:fdd22bb7aa52 365 */