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_interpolate_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?
User | Revision | Line number | New 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_interpolate_q31.c |
emilmont | 1:fdd22bb7aa52 | 9 | * |
emilmont | 2:da51fb522205 | 10 | * Description: Q31 FIR interpolation. |
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 | |
emilmont | 1:fdd22bb7aa52 | 36 | #include "arm_math.h" |
emilmont | 1:fdd22bb7aa52 | 37 | |
emilmont | 1:fdd22bb7aa52 | 38 | /** |
emilmont | 1:fdd22bb7aa52 | 39 | * @ingroup groupFilters |
emilmont | 1:fdd22bb7aa52 | 40 | */ |
emilmont | 1:fdd22bb7aa52 | 41 | |
emilmont | 1:fdd22bb7aa52 | 42 | /** |
emilmont | 1:fdd22bb7aa52 | 43 | * @addtogroup FIR_Interpolate |
emilmont | 1:fdd22bb7aa52 | 44 | * @{ |
emilmont | 1:fdd22bb7aa52 | 45 | */ |
emilmont | 1:fdd22bb7aa52 | 46 | |
emilmont | 1:fdd22bb7aa52 | 47 | /** |
emilmont | 1:fdd22bb7aa52 | 48 | * @brief Processing function for the Q31 FIR interpolator. |
emilmont | 1:fdd22bb7aa52 | 49 | * @param[in] *S points to an instance of the Q31 FIR interpolator structure. |
emilmont | 1:fdd22bb7aa52 | 50 | * @param[in] *pSrc points to the block of input data. |
emilmont | 1:fdd22bb7aa52 | 51 | * @param[out] *pDst points to the block of output data. |
emilmont | 1:fdd22bb7aa52 | 52 | * @param[in] blockSize number of input samples to process per call. |
emilmont | 1:fdd22bb7aa52 | 53 | * @return none. |
emilmont | 1:fdd22bb7aa52 | 54 | * |
emilmont | 1:fdd22bb7aa52 | 55 | * <b>Scaling and Overflow Behavior:</b> |
emilmont | 1:fdd22bb7aa52 | 56 | * \par |
emilmont | 1:fdd22bb7aa52 | 57 | * The function is implemented using an internal 64-bit accumulator. |
emilmont | 1:fdd22bb7aa52 | 58 | * 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 | 59 | * Thus, if the accumulator result overflows it wraps around rather than clip. |
emilmont | 1:fdd22bb7aa52 | 60 | * In order to avoid overflows completely the input signal must be scaled down by <code>1/(numTaps/L)</code>. |
emilmont | 1:fdd22bb7aa52 | 61 | * since <code>numTaps/L</code> additions occur per output sample. |
emilmont | 1:fdd22bb7aa52 | 62 | * After all multiply-accumulates are performed, the 2.62 accumulator is truncated to 1.32 format and then saturated to 1.31 format. |
emilmont | 1:fdd22bb7aa52 | 63 | */ |
emilmont | 1:fdd22bb7aa52 | 64 | |
emilmont | 1:fdd22bb7aa52 | 65 | #ifndef ARM_MATH_CM0 |
emilmont | 1:fdd22bb7aa52 | 66 | |
emilmont | 1:fdd22bb7aa52 | 67 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
emilmont | 1:fdd22bb7aa52 | 68 | |
emilmont | 1:fdd22bb7aa52 | 69 | void arm_fir_interpolate_q31( |
emilmont | 1:fdd22bb7aa52 | 70 | const arm_fir_interpolate_instance_q31 * S, |
emilmont | 1:fdd22bb7aa52 | 71 | q31_t * pSrc, |
emilmont | 1:fdd22bb7aa52 | 72 | q31_t * pDst, |
emilmont | 1:fdd22bb7aa52 | 73 | uint32_t blockSize) |
emilmont | 1:fdd22bb7aa52 | 74 | { |
emilmont | 1:fdd22bb7aa52 | 75 | q31_t *pState = S->pState; /* State pointer */ |
emilmont | 1:fdd22bb7aa52 | 76 | q31_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */ |
emilmont | 1:fdd22bb7aa52 | 77 | q31_t *pStateCurnt; /* Points to the current sample of the state */ |
emilmont | 1:fdd22bb7aa52 | 78 | q31_t *ptr1, *ptr2; /* Temporary pointers for state and coefficient buffers */ |
emilmont | 1:fdd22bb7aa52 | 79 | q63_t sum0; /* Accumulators */ |
emilmont | 1:fdd22bb7aa52 | 80 | q31_t x0, c0; /* Temporary variables to hold state and coefficient values */ |
emilmont | 1:fdd22bb7aa52 | 81 | uint32_t i, blkCnt, j; /* Loop counters */ |
emilmont | 1:fdd22bb7aa52 | 82 | uint16_t phaseLen = S->phaseLength, tapCnt; /* Length of each polyphase filter component */ |
emilmont | 1:fdd22bb7aa52 | 83 | |
emilmont | 1:fdd22bb7aa52 | 84 | uint32_t blkCntN2; |
emilmont | 1:fdd22bb7aa52 | 85 | q63_t acc0, acc1; |
emilmont | 1:fdd22bb7aa52 | 86 | q31_t x1; |
emilmont | 1:fdd22bb7aa52 | 87 | |
emilmont | 1:fdd22bb7aa52 | 88 | /* S->pState buffer contains previous frame (phaseLen - 1) samples */ |
emilmont | 1:fdd22bb7aa52 | 89 | /* pStateCurnt points to the location where the new input data should be written */ |
emilmont | 1:fdd22bb7aa52 | 90 | pStateCurnt = S->pState + ((q31_t) phaseLen - 1); |
emilmont | 1:fdd22bb7aa52 | 91 | |
emilmont | 1:fdd22bb7aa52 | 92 | /* Initialise blkCnt */ |
emilmont | 1:fdd22bb7aa52 | 93 | blkCnt = blockSize / 2; |
emilmont | 1:fdd22bb7aa52 | 94 | blkCntN2 = blockSize - (2 * blkCnt); |
emilmont | 1:fdd22bb7aa52 | 95 | |
emilmont | 1:fdd22bb7aa52 | 96 | /* Samples loop unrolled by 2 */ |
emilmont | 1:fdd22bb7aa52 | 97 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 98 | { |
emilmont | 1:fdd22bb7aa52 | 99 | /* Copy new input sample into the state buffer */ |
emilmont | 1:fdd22bb7aa52 | 100 | *pStateCurnt++ = *pSrc++; |
emilmont | 1:fdd22bb7aa52 | 101 | *pStateCurnt++ = *pSrc++; |
emilmont | 1:fdd22bb7aa52 | 102 | |
emilmont | 1:fdd22bb7aa52 | 103 | /* Address modifier index of coefficient buffer */ |
emilmont | 1:fdd22bb7aa52 | 104 | j = 1u; |
emilmont | 1:fdd22bb7aa52 | 105 | |
emilmont | 1:fdd22bb7aa52 | 106 | /* Loop over the Interpolation factor. */ |
emilmont | 1:fdd22bb7aa52 | 107 | i = (S->L); |
emilmont | 1:fdd22bb7aa52 | 108 | |
emilmont | 1:fdd22bb7aa52 | 109 | while(i > 0u) |
emilmont | 1:fdd22bb7aa52 | 110 | { |
emilmont | 1:fdd22bb7aa52 | 111 | /* Set accumulator to zero */ |
emilmont | 1:fdd22bb7aa52 | 112 | acc0 = 0; |
emilmont | 1:fdd22bb7aa52 | 113 | acc1 = 0; |
emilmont | 1:fdd22bb7aa52 | 114 | |
emilmont | 1:fdd22bb7aa52 | 115 | /* Initialize state pointer */ |
emilmont | 1:fdd22bb7aa52 | 116 | ptr1 = pState; |
emilmont | 1:fdd22bb7aa52 | 117 | |
emilmont | 1:fdd22bb7aa52 | 118 | /* Initialize coefficient pointer */ |
emilmont | 1:fdd22bb7aa52 | 119 | ptr2 = pCoeffs + (S->L - j); |
emilmont | 1:fdd22bb7aa52 | 120 | |
emilmont | 1:fdd22bb7aa52 | 121 | /* Loop over the polyPhase length. Unroll by a factor of 4. |
emilmont | 1:fdd22bb7aa52 | 122 | ** Repeat until we've computed numTaps-(4*S->L) coefficients. */ |
emilmont | 1:fdd22bb7aa52 | 123 | tapCnt = phaseLen >> 2u; |
emilmont | 1:fdd22bb7aa52 | 124 | |
emilmont | 1:fdd22bb7aa52 | 125 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 126 | |
emilmont | 1:fdd22bb7aa52 | 127 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 128 | { |
emilmont | 1:fdd22bb7aa52 | 129 | |
emilmont | 1:fdd22bb7aa52 | 130 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 131 | x1 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 132 | |
emilmont | 1:fdd22bb7aa52 | 133 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 134 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 135 | |
emilmont | 1:fdd22bb7aa52 | 136 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 137 | acc0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 138 | acc1 += (q63_t) x1 *c0; |
emilmont | 1:fdd22bb7aa52 | 139 | |
emilmont | 1:fdd22bb7aa52 | 140 | |
emilmont | 1:fdd22bb7aa52 | 141 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 142 | c0 = *(ptr2 + S->L); |
emilmont | 1:fdd22bb7aa52 | 143 | |
emilmont | 1:fdd22bb7aa52 | 144 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 145 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 146 | |
emilmont | 1:fdd22bb7aa52 | 147 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 148 | acc0 += (q63_t) x1 *c0; |
emilmont | 1:fdd22bb7aa52 | 149 | acc1 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 150 | |
emilmont | 1:fdd22bb7aa52 | 151 | |
emilmont | 1:fdd22bb7aa52 | 152 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 153 | c0 = *(ptr2 + S->L * 2); |
emilmont | 1:fdd22bb7aa52 | 154 | |
emilmont | 1:fdd22bb7aa52 | 155 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 156 | x1 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 157 | |
emilmont | 1:fdd22bb7aa52 | 158 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 159 | acc0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 160 | acc1 += (q63_t) x1 *c0; |
emilmont | 1:fdd22bb7aa52 | 161 | |
emilmont | 1:fdd22bb7aa52 | 162 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 163 | c0 = *(ptr2 + S->L * 3); |
emilmont | 1:fdd22bb7aa52 | 164 | |
emilmont | 1:fdd22bb7aa52 | 165 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 166 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 167 | |
emilmont | 1:fdd22bb7aa52 | 168 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 169 | acc0 += (q63_t) x1 *c0; |
emilmont | 1:fdd22bb7aa52 | 170 | acc1 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 171 | |
emilmont | 1:fdd22bb7aa52 | 172 | |
emilmont | 1:fdd22bb7aa52 | 173 | /* Upsampling is done by stuffing L-1 zeros between each sample. |
emilmont | 1:fdd22bb7aa52 | 174 | * So instead of multiplying zeros with coefficients, |
emilmont | 1:fdd22bb7aa52 | 175 | * Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 176 | ptr2 += 4 * S->L; |
emilmont | 1:fdd22bb7aa52 | 177 | |
emilmont | 1:fdd22bb7aa52 | 178 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 179 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 180 | } |
emilmont | 1:fdd22bb7aa52 | 181 | |
emilmont | 1:fdd22bb7aa52 | 182 | /* If the polyPhase length is not a multiple of 4, compute the remaining filter taps */ |
emilmont | 1:fdd22bb7aa52 | 183 | tapCnt = phaseLen % 0x4u; |
emilmont | 1:fdd22bb7aa52 | 184 | |
emilmont | 1:fdd22bb7aa52 | 185 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 186 | { |
emilmont | 1:fdd22bb7aa52 | 187 | |
emilmont | 1:fdd22bb7aa52 | 188 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 189 | x1 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 190 | |
emilmont | 1:fdd22bb7aa52 | 191 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 192 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 193 | |
emilmont | 1:fdd22bb7aa52 | 194 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 195 | acc0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 196 | acc1 += (q63_t) x1 *c0; |
emilmont | 1:fdd22bb7aa52 | 197 | |
emilmont | 1:fdd22bb7aa52 | 198 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 199 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 200 | |
emilmont | 1:fdd22bb7aa52 | 201 | /* update states for next sample processing */ |
emilmont | 1:fdd22bb7aa52 | 202 | x0 = x1; |
emilmont | 1:fdd22bb7aa52 | 203 | |
emilmont | 1:fdd22bb7aa52 | 204 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 205 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 206 | } |
emilmont | 1:fdd22bb7aa52 | 207 | |
emilmont | 1:fdd22bb7aa52 | 208 | /* The result is in the accumulator, store in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 209 | *pDst = (q31_t) (acc0 >> 31); |
emilmont | 1:fdd22bb7aa52 | 210 | *(pDst + S->L) = (q31_t) (acc1 >> 31); |
emilmont | 1:fdd22bb7aa52 | 211 | |
emilmont | 1:fdd22bb7aa52 | 212 | |
emilmont | 1:fdd22bb7aa52 | 213 | pDst++; |
emilmont | 1:fdd22bb7aa52 | 214 | |
emilmont | 1:fdd22bb7aa52 | 215 | /* Increment the address modifier index of coefficient buffer */ |
emilmont | 1:fdd22bb7aa52 | 216 | j++; |
emilmont | 1:fdd22bb7aa52 | 217 | |
emilmont | 1:fdd22bb7aa52 | 218 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 219 | i--; |
emilmont | 1:fdd22bb7aa52 | 220 | } |
emilmont | 1:fdd22bb7aa52 | 221 | |
emilmont | 1:fdd22bb7aa52 | 222 | /* Advance the state pointer by 1 |
emilmont | 1:fdd22bb7aa52 | 223 | * to process the next group of interpolation factor number samples */ |
emilmont | 1:fdd22bb7aa52 | 224 | pState = pState + 2; |
emilmont | 1:fdd22bb7aa52 | 225 | |
emilmont | 1:fdd22bb7aa52 | 226 | pDst += S->L; |
emilmont | 1:fdd22bb7aa52 | 227 | |
emilmont | 1:fdd22bb7aa52 | 228 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 229 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 230 | } |
emilmont | 1:fdd22bb7aa52 | 231 | |
emilmont | 1:fdd22bb7aa52 | 232 | /* If the blockSize is not a multiple of 2, compute any remaining output samples here. |
emilmont | 1:fdd22bb7aa52 | 233 | ** No loop unrolling is used. */ |
emilmont | 1:fdd22bb7aa52 | 234 | blkCnt = blkCntN2; |
emilmont | 1:fdd22bb7aa52 | 235 | |
emilmont | 1:fdd22bb7aa52 | 236 | /* Loop over the blockSize. */ |
emilmont | 1:fdd22bb7aa52 | 237 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 238 | { |
emilmont | 1:fdd22bb7aa52 | 239 | /* Copy new input sample into the state buffer */ |
emilmont | 1:fdd22bb7aa52 | 240 | *pStateCurnt++ = *pSrc++; |
emilmont | 1:fdd22bb7aa52 | 241 | |
emilmont | 1:fdd22bb7aa52 | 242 | /* Address modifier index of coefficient buffer */ |
emilmont | 1:fdd22bb7aa52 | 243 | j = 1u; |
emilmont | 1:fdd22bb7aa52 | 244 | |
emilmont | 1:fdd22bb7aa52 | 245 | /* Loop over the Interpolation factor. */ |
emilmont | 1:fdd22bb7aa52 | 246 | i = S->L; |
emilmont | 1:fdd22bb7aa52 | 247 | while(i > 0u) |
emilmont | 1:fdd22bb7aa52 | 248 | { |
emilmont | 1:fdd22bb7aa52 | 249 | /* Set accumulator to zero */ |
emilmont | 1:fdd22bb7aa52 | 250 | sum0 = 0; |
emilmont | 1:fdd22bb7aa52 | 251 | |
emilmont | 1:fdd22bb7aa52 | 252 | /* Initialize state pointer */ |
emilmont | 1:fdd22bb7aa52 | 253 | ptr1 = pState; |
emilmont | 1:fdd22bb7aa52 | 254 | |
emilmont | 1:fdd22bb7aa52 | 255 | /* Initialize coefficient pointer */ |
emilmont | 1:fdd22bb7aa52 | 256 | ptr2 = pCoeffs + (S->L - j); |
emilmont | 1:fdd22bb7aa52 | 257 | |
emilmont | 1:fdd22bb7aa52 | 258 | /* Loop over the polyPhase length. Unroll by a factor of 4. |
emilmont | 1:fdd22bb7aa52 | 259 | ** Repeat until we've computed numTaps-(4*S->L) coefficients. */ |
emilmont | 1:fdd22bb7aa52 | 260 | tapCnt = phaseLen >> 2; |
emilmont | 1:fdd22bb7aa52 | 261 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 262 | { |
emilmont | 1:fdd22bb7aa52 | 263 | |
emilmont | 1:fdd22bb7aa52 | 264 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 265 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 266 | |
emilmont | 1:fdd22bb7aa52 | 267 | /* Upsampling is done by stuffing L-1 zeros between each sample. |
emilmont | 1:fdd22bb7aa52 | 268 | * So instead of multiplying zeros with coefficients, |
emilmont | 1:fdd22bb7aa52 | 269 | * Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 270 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 271 | |
emilmont | 1:fdd22bb7aa52 | 272 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 273 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 274 | |
emilmont | 1:fdd22bb7aa52 | 275 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 276 | sum0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 277 | |
emilmont | 1:fdd22bb7aa52 | 278 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 279 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 280 | |
emilmont | 1:fdd22bb7aa52 | 281 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 282 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 283 | |
emilmont | 1:fdd22bb7aa52 | 284 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 285 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 286 | |
emilmont | 1:fdd22bb7aa52 | 287 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 288 | sum0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 289 | |
emilmont | 1:fdd22bb7aa52 | 290 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 291 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 292 | |
emilmont | 1:fdd22bb7aa52 | 293 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 294 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 295 | |
emilmont | 1:fdd22bb7aa52 | 296 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 297 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 298 | |
emilmont | 1:fdd22bb7aa52 | 299 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 300 | sum0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 301 | |
emilmont | 1:fdd22bb7aa52 | 302 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 303 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 304 | |
emilmont | 1:fdd22bb7aa52 | 305 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 306 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 307 | |
emilmont | 1:fdd22bb7aa52 | 308 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 309 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 310 | |
emilmont | 1:fdd22bb7aa52 | 311 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 312 | sum0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 313 | |
emilmont | 1:fdd22bb7aa52 | 314 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 315 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 316 | } |
emilmont | 1:fdd22bb7aa52 | 317 | |
emilmont | 1:fdd22bb7aa52 | 318 | /* If the polyPhase length is not a multiple of 4, compute the remaining filter taps */ |
emilmont | 1:fdd22bb7aa52 | 319 | tapCnt = phaseLen & 0x3u; |
emilmont | 1:fdd22bb7aa52 | 320 | |
emilmont | 1:fdd22bb7aa52 | 321 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 322 | { |
emilmont | 1:fdd22bb7aa52 | 323 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 324 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 325 | |
emilmont | 1:fdd22bb7aa52 | 326 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 327 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 328 | |
emilmont | 1:fdd22bb7aa52 | 329 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 330 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 331 | |
emilmont | 1:fdd22bb7aa52 | 332 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 333 | sum0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 334 | |
emilmont | 1:fdd22bb7aa52 | 335 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 336 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 337 | } |
emilmont | 1:fdd22bb7aa52 | 338 | |
emilmont | 1:fdd22bb7aa52 | 339 | /* The result is in the accumulator, store in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 340 | *pDst++ = (q31_t) (sum0 >> 31); |
emilmont | 1:fdd22bb7aa52 | 341 | |
emilmont | 1:fdd22bb7aa52 | 342 | /* Increment the address modifier index of coefficient buffer */ |
emilmont | 1:fdd22bb7aa52 | 343 | j++; |
emilmont | 1:fdd22bb7aa52 | 344 | |
emilmont | 1:fdd22bb7aa52 | 345 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 346 | i--; |
emilmont | 1:fdd22bb7aa52 | 347 | } |
emilmont | 1:fdd22bb7aa52 | 348 | |
emilmont | 1:fdd22bb7aa52 | 349 | /* Advance the state pointer by 1 |
emilmont | 1:fdd22bb7aa52 | 350 | * to process the next group of interpolation factor number samples */ |
emilmont | 1:fdd22bb7aa52 | 351 | pState = pState + 1; |
emilmont | 1:fdd22bb7aa52 | 352 | |
emilmont | 1:fdd22bb7aa52 | 353 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 354 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 355 | } |
emilmont | 1:fdd22bb7aa52 | 356 | |
emilmont | 1:fdd22bb7aa52 | 357 | /* Processing is complete. |
emilmont | 1:fdd22bb7aa52 | 358 | ** Now copy the last phaseLen - 1 samples to the satrt of the state buffer. |
emilmont | 1:fdd22bb7aa52 | 359 | ** This prepares the state buffer for the next function call. */ |
emilmont | 1:fdd22bb7aa52 | 360 | |
emilmont | 1:fdd22bb7aa52 | 361 | /* Points to the start of the state buffer */ |
emilmont | 1:fdd22bb7aa52 | 362 | pStateCurnt = S->pState; |
emilmont | 1:fdd22bb7aa52 | 363 | |
emilmont | 1:fdd22bb7aa52 | 364 | tapCnt = (phaseLen - 1u) >> 2u; |
emilmont | 1:fdd22bb7aa52 | 365 | |
emilmont | 1:fdd22bb7aa52 | 366 | /* copy data */ |
emilmont | 1:fdd22bb7aa52 | 367 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 368 | { |
emilmont | 1:fdd22bb7aa52 | 369 | *pStateCurnt++ = *pState++; |
emilmont | 1:fdd22bb7aa52 | 370 | *pStateCurnt++ = *pState++; |
emilmont | 1:fdd22bb7aa52 | 371 | *pStateCurnt++ = *pState++; |
emilmont | 1:fdd22bb7aa52 | 372 | *pStateCurnt++ = *pState++; |
emilmont | 1:fdd22bb7aa52 | 373 | |
emilmont | 1:fdd22bb7aa52 | 374 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 375 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 376 | } |
emilmont | 1:fdd22bb7aa52 | 377 | |
emilmont | 1:fdd22bb7aa52 | 378 | tapCnt = (phaseLen - 1u) % 0x04u; |
emilmont | 1:fdd22bb7aa52 | 379 | |
emilmont | 1:fdd22bb7aa52 | 380 | /* copy data */ |
emilmont | 1:fdd22bb7aa52 | 381 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 382 | { |
emilmont | 1:fdd22bb7aa52 | 383 | *pStateCurnt++ = *pState++; |
emilmont | 1:fdd22bb7aa52 | 384 | |
emilmont | 1:fdd22bb7aa52 | 385 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 386 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 387 | } |
emilmont | 1:fdd22bb7aa52 | 388 | |
emilmont | 1:fdd22bb7aa52 | 389 | } |
emilmont | 1:fdd22bb7aa52 | 390 | |
emilmont | 1:fdd22bb7aa52 | 391 | |
emilmont | 1:fdd22bb7aa52 | 392 | #else |
emilmont | 1:fdd22bb7aa52 | 393 | |
emilmont | 1:fdd22bb7aa52 | 394 | void arm_fir_interpolate_q31( |
emilmont | 1:fdd22bb7aa52 | 395 | const arm_fir_interpolate_instance_q31 * S, |
emilmont | 1:fdd22bb7aa52 | 396 | q31_t * pSrc, |
emilmont | 1:fdd22bb7aa52 | 397 | q31_t * pDst, |
emilmont | 1:fdd22bb7aa52 | 398 | uint32_t blockSize) |
emilmont | 1:fdd22bb7aa52 | 399 | { |
emilmont | 1:fdd22bb7aa52 | 400 | q31_t *pState = S->pState; /* State pointer */ |
emilmont | 1:fdd22bb7aa52 | 401 | q31_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */ |
emilmont | 1:fdd22bb7aa52 | 402 | q31_t *pStateCurnt; /* Points to the current sample of the state */ |
emilmont | 1:fdd22bb7aa52 | 403 | q31_t *ptr1, *ptr2; /* Temporary pointers for state and coefficient buffers */ |
emilmont | 1:fdd22bb7aa52 | 404 | |
emilmont | 1:fdd22bb7aa52 | 405 | /* Run the below code for Cortex-M0 */ |
emilmont | 1:fdd22bb7aa52 | 406 | |
emilmont | 1:fdd22bb7aa52 | 407 | q63_t sum; /* Accumulator */ |
emilmont | 1:fdd22bb7aa52 | 408 | q31_t x0, c0; /* Temporary variables to hold state and coefficient values */ |
emilmont | 1:fdd22bb7aa52 | 409 | uint32_t i, blkCnt; /* Loop counters */ |
emilmont | 1:fdd22bb7aa52 | 410 | uint16_t phaseLen = S->phaseLength, tapCnt; /* Length of each polyphase filter component */ |
emilmont | 1:fdd22bb7aa52 | 411 | |
emilmont | 1:fdd22bb7aa52 | 412 | |
emilmont | 1:fdd22bb7aa52 | 413 | /* S->pState buffer contains previous frame (phaseLen - 1) samples */ |
emilmont | 1:fdd22bb7aa52 | 414 | /* pStateCurnt points to the location where the new input data should be written */ |
emilmont | 1:fdd22bb7aa52 | 415 | pStateCurnt = S->pState + ((q31_t) phaseLen - 1); |
emilmont | 1:fdd22bb7aa52 | 416 | |
emilmont | 1:fdd22bb7aa52 | 417 | /* Total number of intput samples */ |
emilmont | 1:fdd22bb7aa52 | 418 | blkCnt = blockSize; |
emilmont | 1:fdd22bb7aa52 | 419 | |
emilmont | 1:fdd22bb7aa52 | 420 | /* Loop over the blockSize. */ |
emilmont | 1:fdd22bb7aa52 | 421 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 422 | { |
emilmont | 1:fdd22bb7aa52 | 423 | /* Copy new input sample into the state buffer */ |
emilmont | 1:fdd22bb7aa52 | 424 | *pStateCurnt++ = *pSrc++; |
emilmont | 1:fdd22bb7aa52 | 425 | |
emilmont | 1:fdd22bb7aa52 | 426 | /* Loop over the Interpolation factor. */ |
emilmont | 1:fdd22bb7aa52 | 427 | i = S->L; |
emilmont | 1:fdd22bb7aa52 | 428 | |
emilmont | 1:fdd22bb7aa52 | 429 | while(i > 0u) |
emilmont | 1:fdd22bb7aa52 | 430 | { |
emilmont | 1:fdd22bb7aa52 | 431 | /* Set accumulator to zero */ |
emilmont | 1:fdd22bb7aa52 | 432 | sum = 0; |
emilmont | 1:fdd22bb7aa52 | 433 | |
emilmont | 1:fdd22bb7aa52 | 434 | /* Initialize state pointer */ |
emilmont | 1:fdd22bb7aa52 | 435 | ptr1 = pState; |
emilmont | 1:fdd22bb7aa52 | 436 | |
emilmont | 1:fdd22bb7aa52 | 437 | /* Initialize coefficient pointer */ |
emilmont | 1:fdd22bb7aa52 | 438 | ptr2 = pCoeffs + (i - 1u); |
emilmont | 1:fdd22bb7aa52 | 439 | |
emilmont | 1:fdd22bb7aa52 | 440 | tapCnt = phaseLen; |
emilmont | 1:fdd22bb7aa52 | 441 | |
emilmont | 1:fdd22bb7aa52 | 442 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 443 | { |
emilmont | 1:fdd22bb7aa52 | 444 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 445 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 446 | |
emilmont | 1:fdd22bb7aa52 | 447 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 448 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 449 | |
emilmont | 1:fdd22bb7aa52 | 450 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 451 | x0 = *ptr1++; |
emilmont | 1:fdd22bb7aa52 | 452 | |
emilmont | 1:fdd22bb7aa52 | 453 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 454 | sum += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 455 | |
emilmont | 1:fdd22bb7aa52 | 456 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 457 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 458 | } |
emilmont | 1:fdd22bb7aa52 | 459 | |
emilmont | 1:fdd22bb7aa52 | 460 | /* The result is in the accumulator, store in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 461 | *pDst++ = (q31_t) (sum >> 31); |
emilmont | 1:fdd22bb7aa52 | 462 | |
emilmont | 1:fdd22bb7aa52 | 463 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 464 | i--; |
emilmont | 1:fdd22bb7aa52 | 465 | } |
emilmont | 1:fdd22bb7aa52 | 466 | |
emilmont | 1:fdd22bb7aa52 | 467 | /* Advance the state pointer by 1 |
emilmont | 1:fdd22bb7aa52 | 468 | * to process the next group of interpolation factor number samples */ |
emilmont | 1:fdd22bb7aa52 | 469 | pState = pState + 1; |
emilmont | 1:fdd22bb7aa52 | 470 | |
emilmont | 1:fdd22bb7aa52 | 471 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 472 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 473 | } |
emilmont | 1:fdd22bb7aa52 | 474 | |
emilmont | 1:fdd22bb7aa52 | 475 | /* Processing is complete. |
emilmont | 1:fdd22bb7aa52 | 476 | ** Now copy the last phaseLen - 1 samples to the satrt of the state buffer. |
emilmont | 1:fdd22bb7aa52 | 477 | ** This prepares the state buffer for the next function call. */ |
emilmont | 1:fdd22bb7aa52 | 478 | |
emilmont | 1:fdd22bb7aa52 | 479 | /* Points to the start of the state buffer */ |
emilmont | 1:fdd22bb7aa52 | 480 | pStateCurnt = S->pState; |
emilmont | 1:fdd22bb7aa52 | 481 | |
emilmont | 1:fdd22bb7aa52 | 482 | tapCnt = phaseLen - 1u; |
emilmont | 1:fdd22bb7aa52 | 483 | |
emilmont | 1:fdd22bb7aa52 | 484 | /* copy data */ |
emilmont | 1:fdd22bb7aa52 | 485 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 486 | { |
emilmont | 1:fdd22bb7aa52 | 487 | *pStateCurnt++ = *pState++; |
emilmont | 1:fdd22bb7aa52 | 488 | |
emilmont | 1:fdd22bb7aa52 | 489 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 490 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 491 | } |
emilmont | 1:fdd22bb7aa52 | 492 | |
emilmont | 1:fdd22bb7aa52 | 493 | } |
emilmont | 1:fdd22bb7aa52 | 494 | |
emilmont | 1:fdd22bb7aa52 | 495 | #endif /* #ifndef ARM_MATH_CM0 */ |
emilmont | 1:fdd22bb7aa52 | 496 | |
emilmont | 1:fdd22bb7aa52 | 497 | /** |
emilmont | 1:fdd22bb7aa52 | 498 | * @} end of FIR_Interpolate group |
emilmont | 1:fdd22bb7aa52 | 499 | */ |