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_q15.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_q15.c |
emilmont | 1:fdd22bb7aa52 | 9 | * |
emilmont | 2:da51fb522205 | 10 | * Description: Q15 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 Q15 FIR interpolator. |
emilmont | 1:fdd22bb7aa52 | 49 | * @param[in] *S points to an instance of the Q15 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 a 64-bit internal accumulator. |
emilmont | 1:fdd22bb7aa52 | 58 | * Both coefficients and state variables are represented in 1.15 format and multiplications yield a 2.30 result. |
emilmont | 1:fdd22bb7aa52 | 59 | * The 2.30 intermediate results are accumulated in a 64-bit accumulator in 34.30 format. |
emilmont | 1:fdd22bb7aa52 | 60 | * There is no risk of internal overflow with this approach and the full precision of intermediate multiplications is preserved. |
emilmont | 1:fdd22bb7aa52 | 61 | * After all additions have been performed, the accumulator is truncated to 34.15 format by discarding low 15 bits. |
emilmont | 1:fdd22bb7aa52 | 62 | * Lastly, the accumulator is saturated to yield a result in 1.15 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_q15( |
emilmont | 1:fdd22bb7aa52 | 70 | const arm_fir_interpolate_instance_q15 * S, |
emilmont | 1:fdd22bb7aa52 | 71 | q15_t * pSrc, |
emilmont | 1:fdd22bb7aa52 | 72 | q15_t * pDst, |
emilmont | 1:fdd22bb7aa52 | 73 | uint32_t blockSize) |
emilmont | 1:fdd22bb7aa52 | 74 | { |
emilmont | 1:fdd22bb7aa52 | 75 | q15_t *pState = S->pState; /* State pointer */ |
emilmont | 1:fdd22bb7aa52 | 76 | q15_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */ |
emilmont | 1:fdd22bb7aa52 | 77 | q15_t *pStateCurnt; /* Points to the current sample of the state */ |
emilmont | 1:fdd22bb7aa52 | 78 | q15_t *ptr1, *ptr2; /* Temporary pointers for state and coefficient buffers */ |
emilmont | 1:fdd22bb7aa52 | 79 | q63_t sum0; /* Accumulators */ |
emilmont | 1:fdd22bb7aa52 | 80 | q15_t x0, c0; /* Temporary variables to hold state and coefficient values */ |
emilmont | 1:fdd22bb7aa52 | 81 | uint32_t i, blkCnt, j, tapCnt; /* Loop counters */ |
emilmont | 1:fdd22bb7aa52 | 82 | uint16_t phaseLen = S->phaseLength; /* Length of each polyphase filter component */ |
emilmont | 1:fdd22bb7aa52 | 83 | uint32_t blkCntN2; |
emilmont | 1:fdd22bb7aa52 | 84 | q63_t acc0, acc1; |
emilmont | 1:fdd22bb7aa52 | 85 | q15_t x1; |
emilmont | 1:fdd22bb7aa52 | 86 | |
emilmont | 1:fdd22bb7aa52 | 87 | /* S->pState buffer contains previous frame (phaseLen - 1) samples */ |
emilmont | 1:fdd22bb7aa52 | 88 | /* pStateCurnt points to the location where the new input data should be written */ |
emilmont | 1:fdd22bb7aa52 | 89 | pStateCurnt = S->pState + ((q31_t) phaseLen - 1); |
emilmont | 1:fdd22bb7aa52 | 90 | |
emilmont | 1:fdd22bb7aa52 | 91 | /* Initialise blkCnt */ |
emilmont | 1:fdd22bb7aa52 | 92 | blkCnt = blockSize / 2; |
emilmont | 1:fdd22bb7aa52 | 93 | blkCntN2 = blockSize - (2 * blkCnt); |
emilmont | 1:fdd22bb7aa52 | 94 | |
emilmont | 1:fdd22bb7aa52 | 95 | /* Samples loop unrolled by 2 */ |
emilmont | 1:fdd22bb7aa52 | 96 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 97 | { |
emilmont | 1:fdd22bb7aa52 | 98 | /* Copy new input sample into the state buffer */ |
emilmont | 1:fdd22bb7aa52 | 99 | *pStateCurnt++ = *pSrc++; |
emilmont | 1:fdd22bb7aa52 | 100 | *pStateCurnt++ = *pSrc++; |
emilmont | 1:fdd22bb7aa52 | 101 | |
emilmont | 1:fdd22bb7aa52 | 102 | /* Address modifier index of coefficient buffer */ |
emilmont | 1:fdd22bb7aa52 | 103 | j = 1u; |
emilmont | 1:fdd22bb7aa52 | 104 | |
emilmont | 1:fdd22bb7aa52 | 105 | /* Loop over the Interpolation factor. */ |
emilmont | 1:fdd22bb7aa52 | 106 | i = (S->L); |
emilmont | 1:fdd22bb7aa52 | 107 | |
emilmont | 1:fdd22bb7aa52 | 108 | while(i > 0u) |
emilmont | 1:fdd22bb7aa52 | 109 | { |
emilmont | 1:fdd22bb7aa52 | 110 | /* Set accumulator to zero */ |
emilmont | 1:fdd22bb7aa52 | 111 | acc0 = 0; |
emilmont | 1:fdd22bb7aa52 | 112 | acc1 = 0; |
emilmont | 1:fdd22bb7aa52 | 113 | |
emilmont | 1:fdd22bb7aa52 | 114 | /* Initialize state pointer */ |
emilmont | 1:fdd22bb7aa52 | 115 | ptr1 = pState; |
emilmont | 1:fdd22bb7aa52 | 116 | |
emilmont | 1:fdd22bb7aa52 | 117 | /* Initialize coefficient pointer */ |
emilmont | 1:fdd22bb7aa52 | 118 | ptr2 = pCoeffs + (S->L - j); |
emilmont | 1:fdd22bb7aa52 | 119 | |
emilmont | 1:fdd22bb7aa52 | 120 | /* Loop over the polyPhase length. Unroll by a factor of 4. |
emilmont | 1:fdd22bb7aa52 | 121 | ** Repeat until we've computed numTaps-(4*S->L) coefficients. */ |
emilmont | 1:fdd22bb7aa52 | 122 | tapCnt = phaseLen >> 2u; |
emilmont | 1:fdd22bb7aa52 | 123 | |
emilmont | 1:fdd22bb7aa52 | 124 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 125 | |
emilmont | 1:fdd22bb7aa52 | 126 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 127 | { |
emilmont | 1:fdd22bb7aa52 | 128 | |
emilmont | 1:fdd22bb7aa52 | 129 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 130 | x1 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 131 | |
emilmont | 1:fdd22bb7aa52 | 132 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 133 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 134 | |
emilmont | 1:fdd22bb7aa52 | 135 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 136 | acc0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 137 | acc1 += (q63_t) x1 *c0; |
emilmont | 1:fdd22bb7aa52 | 138 | |
emilmont | 1:fdd22bb7aa52 | 139 | |
emilmont | 1:fdd22bb7aa52 | 140 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 141 | c0 = *(ptr2 + S->L); |
emilmont | 1:fdd22bb7aa52 | 142 | |
emilmont | 1:fdd22bb7aa52 | 143 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 144 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 145 | |
emilmont | 1:fdd22bb7aa52 | 146 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 147 | acc0 += (q63_t) x1 *c0; |
emilmont | 1:fdd22bb7aa52 | 148 | acc1 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 149 | |
emilmont | 1:fdd22bb7aa52 | 150 | |
emilmont | 1:fdd22bb7aa52 | 151 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 152 | c0 = *(ptr2 + S->L * 2); |
emilmont | 1:fdd22bb7aa52 | 153 | |
emilmont | 1:fdd22bb7aa52 | 154 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 155 | x1 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 156 | |
emilmont | 1:fdd22bb7aa52 | 157 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 158 | acc0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 159 | acc1 += (q63_t) x1 *c0; |
emilmont | 1:fdd22bb7aa52 | 160 | |
emilmont | 1:fdd22bb7aa52 | 161 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 162 | c0 = *(ptr2 + S->L * 3); |
emilmont | 1:fdd22bb7aa52 | 163 | |
emilmont | 1:fdd22bb7aa52 | 164 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 165 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 166 | |
emilmont | 1:fdd22bb7aa52 | 167 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 168 | acc0 += (q63_t) x1 *c0; |
emilmont | 1:fdd22bb7aa52 | 169 | acc1 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 170 | |
emilmont | 1:fdd22bb7aa52 | 171 | |
emilmont | 1:fdd22bb7aa52 | 172 | /* Upsampling is done by stuffing L-1 zeros between each sample. |
emilmont | 1:fdd22bb7aa52 | 173 | * So instead of multiplying zeros with coefficients, |
emilmont | 1:fdd22bb7aa52 | 174 | * Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 175 | ptr2 += 4 * S->L; |
emilmont | 1:fdd22bb7aa52 | 176 | |
emilmont | 1:fdd22bb7aa52 | 177 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 178 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 179 | } |
emilmont | 1:fdd22bb7aa52 | 180 | |
emilmont | 1:fdd22bb7aa52 | 181 | /* If the polyPhase length is not a multiple of 4, compute the remaining filter taps */ |
emilmont | 1:fdd22bb7aa52 | 182 | tapCnt = phaseLen % 0x4u; |
emilmont | 1:fdd22bb7aa52 | 183 | |
emilmont | 1:fdd22bb7aa52 | 184 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 185 | { |
emilmont | 1:fdd22bb7aa52 | 186 | |
emilmont | 1:fdd22bb7aa52 | 187 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 188 | x1 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 189 | |
emilmont | 1:fdd22bb7aa52 | 190 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 191 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 192 | |
emilmont | 1:fdd22bb7aa52 | 193 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 194 | acc0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 195 | acc1 += (q63_t) x1 *c0; |
emilmont | 1:fdd22bb7aa52 | 196 | |
emilmont | 1:fdd22bb7aa52 | 197 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 198 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 199 | |
emilmont | 1:fdd22bb7aa52 | 200 | /* update states for next sample processing */ |
emilmont | 1:fdd22bb7aa52 | 201 | x0 = x1; |
emilmont | 1:fdd22bb7aa52 | 202 | |
emilmont | 1:fdd22bb7aa52 | 203 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 204 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 205 | } |
emilmont | 1:fdd22bb7aa52 | 206 | |
emilmont | 1:fdd22bb7aa52 | 207 | /* The result is in the accumulator, store in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 208 | *pDst = (q15_t) (__SSAT((acc0 >> 15), 16)); |
emilmont | 1:fdd22bb7aa52 | 209 | *(pDst + S->L) = (q15_t) (__SSAT((acc1 >> 15), 16)); |
emilmont | 1:fdd22bb7aa52 | 210 | |
emilmont | 1:fdd22bb7aa52 | 211 | pDst++; |
emilmont | 1:fdd22bb7aa52 | 212 | |
emilmont | 1:fdd22bb7aa52 | 213 | /* Increment the address modifier index of coefficient buffer */ |
emilmont | 1:fdd22bb7aa52 | 214 | j++; |
emilmont | 1:fdd22bb7aa52 | 215 | |
emilmont | 1:fdd22bb7aa52 | 216 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 217 | i--; |
emilmont | 1:fdd22bb7aa52 | 218 | } |
emilmont | 1:fdd22bb7aa52 | 219 | |
emilmont | 1:fdd22bb7aa52 | 220 | /* Advance the state pointer by 1 |
emilmont | 1:fdd22bb7aa52 | 221 | * to process the next group of interpolation factor number samples */ |
emilmont | 1:fdd22bb7aa52 | 222 | pState = pState + 2; |
emilmont | 1:fdd22bb7aa52 | 223 | |
emilmont | 1:fdd22bb7aa52 | 224 | pDst += S->L; |
emilmont | 1:fdd22bb7aa52 | 225 | |
emilmont | 1:fdd22bb7aa52 | 226 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 227 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 228 | } |
emilmont | 1:fdd22bb7aa52 | 229 | |
emilmont | 1:fdd22bb7aa52 | 230 | /* If the blockSize is not a multiple of 2, compute any remaining output samples here. |
emilmont | 1:fdd22bb7aa52 | 231 | ** No loop unrolling is used. */ |
emilmont | 1:fdd22bb7aa52 | 232 | blkCnt = blkCntN2; |
emilmont | 1:fdd22bb7aa52 | 233 | |
emilmont | 1:fdd22bb7aa52 | 234 | /* Loop over the blockSize. */ |
emilmont | 1:fdd22bb7aa52 | 235 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 236 | { |
emilmont | 1:fdd22bb7aa52 | 237 | /* Copy new input sample into the state buffer */ |
emilmont | 1:fdd22bb7aa52 | 238 | *pStateCurnt++ = *pSrc++; |
emilmont | 1:fdd22bb7aa52 | 239 | |
emilmont | 1:fdd22bb7aa52 | 240 | /* Address modifier index of coefficient buffer */ |
emilmont | 1:fdd22bb7aa52 | 241 | j = 1u; |
emilmont | 1:fdd22bb7aa52 | 242 | |
emilmont | 1:fdd22bb7aa52 | 243 | /* Loop over the Interpolation factor. */ |
emilmont | 1:fdd22bb7aa52 | 244 | i = S->L; |
emilmont | 1:fdd22bb7aa52 | 245 | while(i > 0u) |
emilmont | 1:fdd22bb7aa52 | 246 | { |
emilmont | 1:fdd22bb7aa52 | 247 | /* Set accumulator to zero */ |
emilmont | 1:fdd22bb7aa52 | 248 | sum0 = 0; |
emilmont | 1:fdd22bb7aa52 | 249 | |
emilmont | 1:fdd22bb7aa52 | 250 | /* Initialize state pointer */ |
emilmont | 1:fdd22bb7aa52 | 251 | ptr1 = pState; |
emilmont | 1:fdd22bb7aa52 | 252 | |
emilmont | 1:fdd22bb7aa52 | 253 | /* Initialize coefficient pointer */ |
emilmont | 1:fdd22bb7aa52 | 254 | ptr2 = pCoeffs + (S->L - j); |
emilmont | 1:fdd22bb7aa52 | 255 | |
emilmont | 1:fdd22bb7aa52 | 256 | /* Loop over the polyPhase length. Unroll by a factor of 4. |
emilmont | 1:fdd22bb7aa52 | 257 | ** Repeat until we've computed numTaps-(4*S->L) coefficients. */ |
emilmont | 1:fdd22bb7aa52 | 258 | tapCnt = phaseLen >> 2; |
emilmont | 1:fdd22bb7aa52 | 259 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 260 | { |
emilmont | 1:fdd22bb7aa52 | 261 | |
emilmont | 1:fdd22bb7aa52 | 262 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 263 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 264 | |
emilmont | 1:fdd22bb7aa52 | 265 | /* Upsampling is done by stuffing L-1 zeros between each sample. |
emilmont | 1:fdd22bb7aa52 | 266 | * So instead of multiplying zeros with coefficients, |
emilmont | 1:fdd22bb7aa52 | 267 | * Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 268 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 269 | |
emilmont | 1:fdd22bb7aa52 | 270 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 271 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 272 | |
emilmont | 1:fdd22bb7aa52 | 273 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 274 | sum0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 275 | |
emilmont | 1:fdd22bb7aa52 | 276 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 277 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 278 | |
emilmont | 1:fdd22bb7aa52 | 279 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 280 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 281 | |
emilmont | 1:fdd22bb7aa52 | 282 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 283 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 284 | |
emilmont | 1:fdd22bb7aa52 | 285 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 286 | sum0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 287 | |
emilmont | 1:fdd22bb7aa52 | 288 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 289 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 290 | |
emilmont | 1:fdd22bb7aa52 | 291 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 292 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 293 | |
emilmont | 1:fdd22bb7aa52 | 294 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 295 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 296 | |
emilmont | 1:fdd22bb7aa52 | 297 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 298 | sum0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 299 | |
emilmont | 1:fdd22bb7aa52 | 300 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 301 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 302 | |
emilmont | 1:fdd22bb7aa52 | 303 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 304 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 305 | |
emilmont | 1:fdd22bb7aa52 | 306 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 307 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 308 | |
emilmont | 1:fdd22bb7aa52 | 309 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 310 | sum0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 311 | |
emilmont | 1:fdd22bb7aa52 | 312 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 313 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 314 | } |
emilmont | 1:fdd22bb7aa52 | 315 | |
emilmont | 1:fdd22bb7aa52 | 316 | /* If the polyPhase length is not a multiple of 4, compute the remaining filter taps */ |
emilmont | 1:fdd22bb7aa52 | 317 | tapCnt = phaseLen & 0x3u; |
emilmont | 1:fdd22bb7aa52 | 318 | |
emilmont | 1:fdd22bb7aa52 | 319 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 320 | { |
emilmont | 1:fdd22bb7aa52 | 321 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 322 | c0 = *(ptr2); |
emilmont | 1:fdd22bb7aa52 | 323 | |
emilmont | 1:fdd22bb7aa52 | 324 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 325 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 326 | |
emilmont | 1:fdd22bb7aa52 | 327 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 328 | x0 = *(ptr1++); |
emilmont | 1:fdd22bb7aa52 | 329 | |
emilmont | 1:fdd22bb7aa52 | 330 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 331 | sum0 += (q63_t) x0 *c0; |
emilmont | 1:fdd22bb7aa52 | 332 | |
emilmont | 1:fdd22bb7aa52 | 333 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 334 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 335 | } |
emilmont | 1:fdd22bb7aa52 | 336 | |
emilmont | 1:fdd22bb7aa52 | 337 | /* The result is in the accumulator, store in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 338 | *pDst++ = (q15_t) (__SSAT((sum0 >> 15), 16)); |
emilmont | 1:fdd22bb7aa52 | 339 | |
emilmont | 1:fdd22bb7aa52 | 340 | j++; |
emilmont | 1:fdd22bb7aa52 | 341 | |
emilmont | 1:fdd22bb7aa52 | 342 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 343 | i--; |
emilmont | 1:fdd22bb7aa52 | 344 | } |
emilmont | 1:fdd22bb7aa52 | 345 | |
emilmont | 1:fdd22bb7aa52 | 346 | /* Advance the state pointer by 1 |
emilmont | 1:fdd22bb7aa52 | 347 | * to process the next group of interpolation factor number samples */ |
emilmont | 1:fdd22bb7aa52 | 348 | pState = pState + 1; |
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. |
emilmont | 1:fdd22bb7aa52 | 356 | ** Now copy the last phaseLen - 1 samples to the satrt of the state buffer. |
emilmont | 1:fdd22bb7aa52 | 357 | ** This prepares the state buffer for the next function call. */ |
emilmont | 1:fdd22bb7aa52 | 358 | |
emilmont | 1:fdd22bb7aa52 | 359 | /* Points to the start of the state buffer */ |
emilmont | 1:fdd22bb7aa52 | 360 | pStateCurnt = S->pState; |
emilmont | 1:fdd22bb7aa52 | 361 | |
emilmont | 1:fdd22bb7aa52 | 362 | i = ((uint32_t) phaseLen - 1u) >> 2u; |
emilmont | 1:fdd22bb7aa52 | 363 | |
emilmont | 1:fdd22bb7aa52 | 364 | /* copy data */ |
emilmont | 1:fdd22bb7aa52 | 365 | while(i > 0u) |
emilmont | 1:fdd22bb7aa52 | 366 | { |
emilmont | 1:fdd22bb7aa52 | 367 | #ifndef UNALIGNED_SUPPORT_DISABLE |
emilmont | 1:fdd22bb7aa52 | 368 | |
emilmont | 1:fdd22bb7aa52 | 369 | *__SIMD32(pStateCurnt)++ = *__SIMD32(pState)++; |
emilmont | 1:fdd22bb7aa52 | 370 | *__SIMD32(pStateCurnt)++ = *__SIMD32(pState)++; |
emilmont | 1:fdd22bb7aa52 | 371 | |
emilmont | 1:fdd22bb7aa52 | 372 | #else |
emilmont | 1:fdd22bb7aa52 | 373 | |
emilmont | 1:fdd22bb7aa52 | 374 | *pStateCurnt++ = *pState++; |
emilmont | 2:da51fb522205 | 375 | *pStateCurnt++ = *pState++; |
emilmont | 2:da51fb522205 | 376 | *pStateCurnt++ = *pState++; |
emilmont | 2:da51fb522205 | 377 | *pStateCurnt++ = *pState++; |
emilmont | 2:da51fb522205 | 378 | |
emilmont | 2:da51fb522205 | 379 | #endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */ |
emilmont | 2:da51fb522205 | 380 | |
emilmont | 2:da51fb522205 | 381 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 382 | i--; |
emilmont | 1:fdd22bb7aa52 | 383 | } |
emilmont | 1:fdd22bb7aa52 | 384 | |
emilmont | 1:fdd22bb7aa52 | 385 | i = ((uint32_t) phaseLen - 1u) % 0x04u; |
emilmont | 1:fdd22bb7aa52 | 386 | |
emilmont | 1:fdd22bb7aa52 | 387 | while(i > 0u) |
emilmont | 1:fdd22bb7aa52 | 388 | { |
emilmont | 1:fdd22bb7aa52 | 389 | *pStateCurnt++ = *pState++; |
emilmont | 1:fdd22bb7aa52 | 390 | |
emilmont | 1:fdd22bb7aa52 | 391 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 392 | i--; |
emilmont | 1:fdd22bb7aa52 | 393 | } |
emilmont | 1:fdd22bb7aa52 | 394 | } |
emilmont | 1:fdd22bb7aa52 | 395 | |
emilmont | 1:fdd22bb7aa52 | 396 | #else |
emilmont | 1:fdd22bb7aa52 | 397 | |
emilmont | 1:fdd22bb7aa52 | 398 | /* Run the below code for Cortex-M0 */ |
emilmont | 1:fdd22bb7aa52 | 399 | |
emilmont | 1:fdd22bb7aa52 | 400 | void arm_fir_interpolate_q15( |
emilmont | 1:fdd22bb7aa52 | 401 | const arm_fir_interpolate_instance_q15 * S, |
emilmont | 1:fdd22bb7aa52 | 402 | q15_t * pSrc, |
emilmont | 1:fdd22bb7aa52 | 403 | q15_t * pDst, |
emilmont | 1:fdd22bb7aa52 | 404 | uint32_t blockSize) |
emilmont | 1:fdd22bb7aa52 | 405 | { |
emilmont | 1:fdd22bb7aa52 | 406 | q15_t *pState = S->pState; /* State pointer */ |
emilmont | 1:fdd22bb7aa52 | 407 | q15_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */ |
emilmont | 1:fdd22bb7aa52 | 408 | q15_t *pStateCurnt; /* Points to the current sample of the state */ |
emilmont | 1:fdd22bb7aa52 | 409 | q15_t *ptr1, *ptr2; /* Temporary pointers for state and coefficient buffers */ |
emilmont | 1:fdd22bb7aa52 | 410 | q63_t sum; /* Accumulator */ |
emilmont | 1:fdd22bb7aa52 | 411 | q15_t x0, c0; /* Temporary variables to hold state and coefficient values */ |
emilmont | 1:fdd22bb7aa52 | 412 | uint32_t i, blkCnt, tapCnt; /* Loop counters */ |
emilmont | 1:fdd22bb7aa52 | 413 | uint16_t phaseLen = S->phaseLength; /* Length of each polyphase filter component */ |
emilmont | 1:fdd22bb7aa52 | 414 | |
emilmont | 1:fdd22bb7aa52 | 415 | |
emilmont | 1:fdd22bb7aa52 | 416 | /* S->pState buffer contains previous frame (phaseLen - 1) samples */ |
emilmont | 1:fdd22bb7aa52 | 417 | /* pStateCurnt points to the location where the new input data should be written */ |
emilmont | 1:fdd22bb7aa52 | 418 | pStateCurnt = S->pState + (phaseLen - 1u); |
emilmont | 1:fdd22bb7aa52 | 419 | |
emilmont | 1:fdd22bb7aa52 | 420 | /* Total number of intput samples */ |
emilmont | 1:fdd22bb7aa52 | 421 | blkCnt = blockSize; |
emilmont | 1:fdd22bb7aa52 | 422 | |
emilmont | 1:fdd22bb7aa52 | 423 | /* Loop over the blockSize. */ |
emilmont | 1:fdd22bb7aa52 | 424 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 425 | { |
emilmont | 1:fdd22bb7aa52 | 426 | /* Copy new input sample into the state buffer */ |
emilmont | 1:fdd22bb7aa52 | 427 | *pStateCurnt++ = *pSrc++; |
emilmont | 1:fdd22bb7aa52 | 428 | |
emilmont | 1:fdd22bb7aa52 | 429 | /* Loop over the Interpolation factor. */ |
emilmont | 1:fdd22bb7aa52 | 430 | i = S->L; |
emilmont | 1:fdd22bb7aa52 | 431 | |
emilmont | 1:fdd22bb7aa52 | 432 | while(i > 0u) |
emilmont | 1:fdd22bb7aa52 | 433 | { |
emilmont | 1:fdd22bb7aa52 | 434 | /* Set accumulator to zero */ |
emilmont | 1:fdd22bb7aa52 | 435 | sum = 0; |
emilmont | 1:fdd22bb7aa52 | 436 | |
emilmont | 1:fdd22bb7aa52 | 437 | /* Initialize state pointer */ |
emilmont | 1:fdd22bb7aa52 | 438 | ptr1 = pState; |
emilmont | 1:fdd22bb7aa52 | 439 | |
emilmont | 1:fdd22bb7aa52 | 440 | /* Initialize coefficient pointer */ |
emilmont | 1:fdd22bb7aa52 | 441 | ptr2 = pCoeffs + (i - 1u); |
emilmont | 1:fdd22bb7aa52 | 442 | |
emilmont | 1:fdd22bb7aa52 | 443 | /* Loop over the polyPhase length */ |
emilmont | 1:fdd22bb7aa52 | 444 | tapCnt = (uint32_t) phaseLen; |
emilmont | 1:fdd22bb7aa52 | 445 | |
emilmont | 1:fdd22bb7aa52 | 446 | while(tapCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 447 | { |
emilmont | 1:fdd22bb7aa52 | 448 | /* Read the coefficient */ |
emilmont | 1:fdd22bb7aa52 | 449 | c0 = *ptr2; |
emilmont | 1:fdd22bb7aa52 | 450 | |
emilmont | 1:fdd22bb7aa52 | 451 | /* Increment the coefficient pointer by interpolation factor times. */ |
emilmont | 1:fdd22bb7aa52 | 452 | ptr2 += S->L; |
emilmont | 1:fdd22bb7aa52 | 453 | |
emilmont | 1:fdd22bb7aa52 | 454 | /* Read the input sample */ |
emilmont | 1:fdd22bb7aa52 | 455 | x0 = *ptr1++; |
emilmont | 1:fdd22bb7aa52 | 456 | |
emilmont | 1:fdd22bb7aa52 | 457 | /* Perform the multiply-accumulate */ |
emilmont | 1:fdd22bb7aa52 | 458 | sum += ((q31_t) x0 * c0); |
emilmont | 1:fdd22bb7aa52 | 459 | |
emilmont | 1:fdd22bb7aa52 | 460 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 461 | tapCnt--; |
emilmont | 1:fdd22bb7aa52 | 462 | } |
emilmont | 1:fdd22bb7aa52 | 463 | |
emilmont | 1:fdd22bb7aa52 | 464 | /* Store the result after converting to 1.15 format in the destination buffer */ |
emilmont | 1:fdd22bb7aa52 | 465 | *pDst++ = (q15_t) (__SSAT((sum >> 15), 16)); |
emilmont | 1:fdd22bb7aa52 | 466 | |
emilmont | 1:fdd22bb7aa52 | 467 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 468 | i--; |
emilmont | 1:fdd22bb7aa52 | 469 | } |
emilmont | 1:fdd22bb7aa52 | 470 | |
emilmont | 1:fdd22bb7aa52 | 471 | /* Advance the state pointer by 1 |
emilmont | 1:fdd22bb7aa52 | 472 | * to process the next group of interpolation factor number samples */ |
emilmont | 1:fdd22bb7aa52 | 473 | pState = pState + 1; |
emilmont | 1:fdd22bb7aa52 | 474 | |
emilmont | 1:fdd22bb7aa52 | 475 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 476 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 477 | } |
emilmont | 1:fdd22bb7aa52 | 478 | |
emilmont | 1:fdd22bb7aa52 | 479 | /* Processing is complete. |
emilmont | 1:fdd22bb7aa52 | 480 | ** Now copy the last phaseLen - 1 samples to the start of the state buffer. |
emilmont | 1:fdd22bb7aa52 | 481 | ** This prepares the state buffer for the next function call. */ |
emilmont | 1:fdd22bb7aa52 | 482 | |
emilmont | 1:fdd22bb7aa52 | 483 | /* Points to the start of the state buffer */ |
emilmont | 1:fdd22bb7aa52 | 484 | pStateCurnt = S->pState; |
emilmont | 1:fdd22bb7aa52 | 485 | |
emilmont | 1:fdd22bb7aa52 | 486 | i = (uint32_t) phaseLen - 1u; |
emilmont | 1:fdd22bb7aa52 | 487 | |
emilmont | 1:fdd22bb7aa52 | 488 | while(i > 0u) |
emilmont | 1:fdd22bb7aa52 | 489 | { |
emilmont | 1:fdd22bb7aa52 | 490 | *pStateCurnt++ = *pState++; |
emilmont | 1:fdd22bb7aa52 | 491 | |
emilmont | 1:fdd22bb7aa52 | 492 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 493 | i--; |
emilmont | 1:fdd22bb7aa52 | 494 | } |
emilmont | 1:fdd22bb7aa52 | 495 | |
emilmont | 1:fdd22bb7aa52 | 496 | } |
emilmont | 1:fdd22bb7aa52 | 497 | |
emilmont | 1:fdd22bb7aa52 | 498 | #endif /* #ifndef ARM_MATH_CM0 */ |
emilmont | 1:fdd22bb7aa52 | 499 | |
emilmont | 1:fdd22bb7aa52 | 500 | |
emilmont | 1:fdd22bb7aa52 | 501 | /** |
emilmont | 1:fdd22bb7aa52 | 502 | * @} end of FIR_Interpolate group |
emilmont | 1:fdd22bb7aa52 | 503 | */ |