CMSIS DSP library
Dependents: KL25Z_FFT_Demo Hat_Board_v5_1 KL25Z_FFT_Demo_tony KL25Z_FFT_Demo_tony ... more
Fork of mbed-dsp by
cmsis_dsp/FilteringFunctions/arm_biquad_cascade_df1_q31.c@1:fdd22bb7aa52, 2012-11-28 (annotated)
- Committer:
- emilmont
- Date:
- Wed Nov 28 12:30:09 2012 +0000
- Revision:
- 1:fdd22bb7aa52
- Child:
- 2:da51fb522205
DSP library code
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 | 1:fdd22bb7aa52 | 5 | * $Revision: V1.1.0 |
emilmont | 1:fdd22bb7aa52 | 6 | * |
emilmont | 1:fdd22bb7aa52 | 7 | * Project: CMSIS DSP Library |
emilmont | 1:fdd22bb7aa52 | 8 | * Title: arm_biquad_cascade_df1_q31.c |
emilmont | 1:fdd22bb7aa52 | 9 | * |
emilmont | 1:fdd22bb7aa52 | 10 | * Description: Processing function for the |
emilmont | 1:fdd22bb7aa52 | 11 | * Q31 Biquad cascade filter |
emilmont | 1:fdd22bb7aa52 | 12 | * |
emilmont | 1:fdd22bb7aa52 | 13 | * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0 |
emilmont | 1:fdd22bb7aa52 | 14 | * |
emilmont | 1:fdd22bb7aa52 | 15 | * Version 1.1.0 2012/02/15 |
emilmont | 1:fdd22bb7aa52 | 16 | * Updated with more optimizations, bug fixes and minor API changes. |
emilmont | 1:fdd22bb7aa52 | 17 | * |
emilmont | 1:fdd22bb7aa52 | 18 | * Version 1.0.10 2011/7/15 |
emilmont | 1:fdd22bb7aa52 | 19 | * Big Endian support added and Merged M0 and M3/M4 Source code. |
emilmont | 1:fdd22bb7aa52 | 20 | * |
emilmont | 1:fdd22bb7aa52 | 21 | * Version 1.0.3 2010/11/29 |
emilmont | 1:fdd22bb7aa52 | 22 | * Re-organized the CMSIS folders and updated documentation. |
emilmont | 1:fdd22bb7aa52 | 23 | * |
emilmont | 1:fdd22bb7aa52 | 24 | * Version 1.0.2 2010/11/11 |
emilmont | 1:fdd22bb7aa52 | 25 | * Documentation updated. |
emilmont | 1:fdd22bb7aa52 | 26 | * |
emilmont | 1:fdd22bb7aa52 | 27 | * Version 1.0.1 2010/10/05 |
emilmont | 1:fdd22bb7aa52 | 28 | * Production release and review comments incorporated. |
emilmont | 1:fdd22bb7aa52 | 29 | * |
emilmont | 1:fdd22bb7aa52 | 30 | * Version 1.0.0 2010/09/20 |
emilmont | 1:fdd22bb7aa52 | 31 | * Production release and review comments incorporated. |
emilmont | 1:fdd22bb7aa52 | 32 | * |
emilmont | 1:fdd22bb7aa52 | 33 | * Version 0.0.5 2010/04/26 |
emilmont | 1:fdd22bb7aa52 | 34 | * incorporated review comments and updated with latest CMSIS layer |
emilmont | 1:fdd22bb7aa52 | 35 | * |
emilmont | 1:fdd22bb7aa52 | 36 | * Version 0.0.3 2010/03/10 |
emilmont | 1:fdd22bb7aa52 | 37 | * Initial version |
emilmont | 1:fdd22bb7aa52 | 38 | * -------------------------------------------------------------------- */ |
emilmont | 1:fdd22bb7aa52 | 39 | |
emilmont | 1:fdd22bb7aa52 | 40 | #include "arm_math.h" |
emilmont | 1:fdd22bb7aa52 | 41 | |
emilmont | 1:fdd22bb7aa52 | 42 | /** |
emilmont | 1:fdd22bb7aa52 | 43 | * @ingroup groupFilters |
emilmont | 1:fdd22bb7aa52 | 44 | */ |
emilmont | 1:fdd22bb7aa52 | 45 | |
emilmont | 1:fdd22bb7aa52 | 46 | /** |
emilmont | 1:fdd22bb7aa52 | 47 | * @addtogroup BiquadCascadeDF1 |
emilmont | 1:fdd22bb7aa52 | 48 | * @{ |
emilmont | 1:fdd22bb7aa52 | 49 | */ |
emilmont | 1:fdd22bb7aa52 | 50 | |
emilmont | 1:fdd22bb7aa52 | 51 | /** |
emilmont | 1:fdd22bb7aa52 | 52 | * @brief Processing function for the Q31 Biquad cascade filter. |
emilmont | 1:fdd22bb7aa52 | 53 | * @param[in] *S points to an instance of the Q31 Biquad cascade structure. |
emilmont | 1:fdd22bb7aa52 | 54 | * @param[in] *pSrc points to the block of input data. |
emilmont | 1:fdd22bb7aa52 | 55 | * @param[out] *pDst points to the block of output data. |
emilmont | 1:fdd22bb7aa52 | 56 | * @param[in] blockSize number of samples to process per call. |
emilmont | 1:fdd22bb7aa52 | 57 | * @return none. |
emilmont | 1:fdd22bb7aa52 | 58 | * |
emilmont | 1:fdd22bb7aa52 | 59 | * <b>Scaling and Overflow Behavior:</b> |
emilmont | 1:fdd22bb7aa52 | 60 | * \par |
emilmont | 1:fdd22bb7aa52 | 61 | * The function is implemented using an internal 64-bit accumulator. |
emilmont | 1:fdd22bb7aa52 | 62 | * 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 | 63 | * Thus, if the accumulator result overflows it wraps around rather than clip. |
emilmont | 1:fdd22bb7aa52 | 64 | * In order to avoid overflows completely the input signal must be scaled down by 2 bits and lie in the range [-0.25 +0.25). |
emilmont | 1:fdd22bb7aa52 | 65 | * After all 5 multiply-accumulates are performed, the 2.62 accumulator is shifted by <code>postShift</code> bits and the result truncated to |
emilmont | 1:fdd22bb7aa52 | 66 | * 1.31 format by discarding the low 32 bits. |
emilmont | 1:fdd22bb7aa52 | 67 | * |
emilmont | 1:fdd22bb7aa52 | 68 | * \par |
emilmont | 1:fdd22bb7aa52 | 69 | * Refer to the function <code>arm_biquad_cascade_df1_fast_q31()</code> for a faster but less precise implementation of this filter for Cortex-M3 and Cortex-M4. |
emilmont | 1:fdd22bb7aa52 | 70 | */ |
emilmont | 1:fdd22bb7aa52 | 71 | |
emilmont | 1:fdd22bb7aa52 | 72 | void arm_biquad_cascade_df1_q31( |
emilmont | 1:fdd22bb7aa52 | 73 | const arm_biquad_casd_df1_inst_q31 * S, |
emilmont | 1:fdd22bb7aa52 | 74 | q31_t * pSrc, |
emilmont | 1:fdd22bb7aa52 | 75 | q31_t * pDst, |
emilmont | 1:fdd22bb7aa52 | 76 | uint32_t blockSize) |
emilmont | 1:fdd22bb7aa52 | 77 | { |
emilmont | 1:fdd22bb7aa52 | 78 | q63_t acc; /* accumulator */ |
emilmont | 1:fdd22bb7aa52 | 79 | uint32_t uShift = ((uint32_t) S->postShift + 1u); |
emilmont | 1:fdd22bb7aa52 | 80 | uint32_t lShift = 32u - uShift; /* Shift to be applied to the output */ |
emilmont | 1:fdd22bb7aa52 | 81 | q31_t *pIn = pSrc; /* input pointer initialization */ |
emilmont | 1:fdd22bb7aa52 | 82 | q31_t *pOut = pDst; /* output pointer initialization */ |
emilmont | 1:fdd22bb7aa52 | 83 | q31_t *pState = S->pState; /* pState pointer initialization */ |
emilmont | 1:fdd22bb7aa52 | 84 | q31_t *pCoeffs = S->pCoeffs; /* coeff pointer initialization */ |
emilmont | 1:fdd22bb7aa52 | 85 | q31_t Xn1, Xn2, Yn1, Yn2; /* Filter state variables */ |
emilmont | 1:fdd22bb7aa52 | 86 | q31_t b0, b1, b2, a1, a2; /* Filter coefficients */ |
emilmont | 1:fdd22bb7aa52 | 87 | q31_t Xn; /* temporary input */ |
emilmont | 1:fdd22bb7aa52 | 88 | uint32_t sample, stage = S->numStages; /* loop counters */ |
emilmont | 1:fdd22bb7aa52 | 89 | |
emilmont | 1:fdd22bb7aa52 | 90 | |
emilmont | 1:fdd22bb7aa52 | 91 | #ifndef ARM_MATH_CM0 |
emilmont | 1:fdd22bb7aa52 | 92 | |
emilmont | 1:fdd22bb7aa52 | 93 | q31_t acc_l, acc_h; /* temporary output variables */ |
emilmont | 1:fdd22bb7aa52 | 94 | |
emilmont | 1:fdd22bb7aa52 | 95 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
emilmont | 1:fdd22bb7aa52 | 96 | |
emilmont | 1:fdd22bb7aa52 | 97 | do |
emilmont | 1:fdd22bb7aa52 | 98 | { |
emilmont | 1:fdd22bb7aa52 | 99 | /* Reading the coefficients */ |
emilmont | 1:fdd22bb7aa52 | 100 | b0 = *pCoeffs++; |
emilmont | 1:fdd22bb7aa52 | 101 | b1 = *pCoeffs++; |
emilmont | 1:fdd22bb7aa52 | 102 | b2 = *pCoeffs++; |
emilmont | 1:fdd22bb7aa52 | 103 | a1 = *pCoeffs++; |
emilmont | 1:fdd22bb7aa52 | 104 | a2 = *pCoeffs++; |
emilmont | 1:fdd22bb7aa52 | 105 | |
emilmont | 1:fdd22bb7aa52 | 106 | /* Reading the state values */ |
emilmont | 1:fdd22bb7aa52 | 107 | Xn1 = pState[0]; |
emilmont | 1:fdd22bb7aa52 | 108 | Xn2 = pState[1]; |
emilmont | 1:fdd22bb7aa52 | 109 | Yn1 = pState[2]; |
emilmont | 1:fdd22bb7aa52 | 110 | Yn2 = pState[3]; |
emilmont | 1:fdd22bb7aa52 | 111 | |
emilmont | 1:fdd22bb7aa52 | 112 | /* Apply loop unrolling and compute 4 output values simultaneously. */ |
emilmont | 1:fdd22bb7aa52 | 113 | /* The variable acc hold output values that are being computed: |
emilmont | 1:fdd22bb7aa52 | 114 | * |
emilmont | 1:fdd22bb7aa52 | 115 | * acc = b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] |
emilmont | 1:fdd22bb7aa52 | 116 | */ |
emilmont | 1:fdd22bb7aa52 | 117 | |
emilmont | 1:fdd22bb7aa52 | 118 | sample = blockSize >> 2u; |
emilmont | 1:fdd22bb7aa52 | 119 | |
emilmont | 1:fdd22bb7aa52 | 120 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
emilmont | 1:fdd22bb7aa52 | 121 | ** a second loop below computes the remaining 1 to 3 samples. */ |
emilmont | 1:fdd22bb7aa52 | 122 | while(sample > 0u) |
emilmont | 1:fdd22bb7aa52 | 123 | { |
emilmont | 1:fdd22bb7aa52 | 124 | /* Read the input */ |
emilmont | 1:fdd22bb7aa52 | 125 | Xn = *pIn++; |
emilmont | 1:fdd22bb7aa52 | 126 | |
emilmont | 1:fdd22bb7aa52 | 127 | /* acc = b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 128 | |
emilmont | 1:fdd22bb7aa52 | 129 | /* acc = b0 * x[n] */ |
emilmont | 1:fdd22bb7aa52 | 130 | acc = (q63_t) b0 *Xn; |
emilmont | 1:fdd22bb7aa52 | 131 | /* acc += b1 * x[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 132 | acc += (q63_t) b1 *Xn1; |
emilmont | 1:fdd22bb7aa52 | 133 | /* acc += b[2] * x[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 134 | acc += (q63_t) b2 *Xn2; |
emilmont | 1:fdd22bb7aa52 | 135 | /* acc += a1 * y[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 136 | acc += (q63_t) a1 *Yn1; |
emilmont | 1:fdd22bb7aa52 | 137 | /* acc += a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 138 | acc += (q63_t) a2 *Yn2; |
emilmont | 1:fdd22bb7aa52 | 139 | |
emilmont | 1:fdd22bb7aa52 | 140 | /* The result is converted to 1.31 , Yn2 variable is reused */ |
emilmont | 1:fdd22bb7aa52 | 141 | |
emilmont | 1:fdd22bb7aa52 | 142 | /* Calc lower part of acc */ |
emilmont | 1:fdd22bb7aa52 | 143 | acc_l = acc & 0xffffffff; |
emilmont | 1:fdd22bb7aa52 | 144 | |
emilmont | 1:fdd22bb7aa52 | 145 | /* Calc upper part of acc */ |
emilmont | 1:fdd22bb7aa52 | 146 | acc_h = (acc >> 32) & 0xffffffff; |
emilmont | 1:fdd22bb7aa52 | 147 | |
emilmont | 1:fdd22bb7aa52 | 148 | /* Apply shift for lower part of acc and upper part of acc */ |
emilmont | 1:fdd22bb7aa52 | 149 | Yn2 = (uint32_t) acc_l >> lShift | acc_h << uShift; |
emilmont | 1:fdd22bb7aa52 | 150 | |
emilmont | 1:fdd22bb7aa52 | 151 | /* Store the output in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 152 | *pOut++ = Yn2; |
emilmont | 1:fdd22bb7aa52 | 153 | |
emilmont | 1:fdd22bb7aa52 | 154 | /* Read the second input */ |
emilmont | 1:fdd22bb7aa52 | 155 | Xn2 = *pIn++; |
emilmont | 1:fdd22bb7aa52 | 156 | |
emilmont | 1:fdd22bb7aa52 | 157 | /* acc = b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 158 | |
emilmont | 1:fdd22bb7aa52 | 159 | /* acc = b0 * x[n] */ |
emilmont | 1:fdd22bb7aa52 | 160 | acc = (q63_t) b0 *Xn2; |
emilmont | 1:fdd22bb7aa52 | 161 | /* acc += b1 * x[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 162 | acc += (q63_t) b1 *Xn; |
emilmont | 1:fdd22bb7aa52 | 163 | /* acc += b[2] * x[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 164 | acc += (q63_t) b2 *Xn1; |
emilmont | 1:fdd22bb7aa52 | 165 | /* acc += a1 * y[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 166 | acc += (q63_t) a1 *Yn2; |
emilmont | 1:fdd22bb7aa52 | 167 | /* acc += a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 168 | acc += (q63_t) a2 *Yn1; |
emilmont | 1:fdd22bb7aa52 | 169 | |
emilmont | 1:fdd22bb7aa52 | 170 | |
emilmont | 1:fdd22bb7aa52 | 171 | /* The result is converted to 1.31, Yn1 variable is reused */ |
emilmont | 1:fdd22bb7aa52 | 172 | |
emilmont | 1:fdd22bb7aa52 | 173 | /* Calc lower part of acc */ |
emilmont | 1:fdd22bb7aa52 | 174 | acc_l = acc & 0xffffffff; |
emilmont | 1:fdd22bb7aa52 | 175 | |
emilmont | 1:fdd22bb7aa52 | 176 | /* Calc upper part of acc */ |
emilmont | 1:fdd22bb7aa52 | 177 | acc_h = (acc >> 32) & 0xffffffff; |
emilmont | 1:fdd22bb7aa52 | 178 | |
emilmont | 1:fdd22bb7aa52 | 179 | |
emilmont | 1:fdd22bb7aa52 | 180 | /* Apply shift for lower part of acc and upper part of acc */ |
emilmont | 1:fdd22bb7aa52 | 181 | Yn1 = (uint32_t) acc_l >> lShift | acc_h << uShift; |
emilmont | 1:fdd22bb7aa52 | 182 | |
emilmont | 1:fdd22bb7aa52 | 183 | /* Store the output in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 184 | *pOut++ = Yn1; |
emilmont | 1:fdd22bb7aa52 | 185 | |
emilmont | 1:fdd22bb7aa52 | 186 | /* Read the third input */ |
emilmont | 1:fdd22bb7aa52 | 187 | Xn1 = *pIn++; |
emilmont | 1:fdd22bb7aa52 | 188 | |
emilmont | 1:fdd22bb7aa52 | 189 | /* acc = b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 190 | |
emilmont | 1:fdd22bb7aa52 | 191 | /* acc = b0 * x[n] */ |
emilmont | 1:fdd22bb7aa52 | 192 | acc = (q63_t) b0 *Xn1; |
emilmont | 1:fdd22bb7aa52 | 193 | /* acc += b1 * x[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 194 | acc += (q63_t) b1 *Xn2; |
emilmont | 1:fdd22bb7aa52 | 195 | /* acc += b[2] * x[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 196 | acc += (q63_t) b2 *Xn; |
emilmont | 1:fdd22bb7aa52 | 197 | /* acc += a1 * y[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 198 | acc += (q63_t) a1 *Yn1; |
emilmont | 1:fdd22bb7aa52 | 199 | /* acc += a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 200 | acc += (q63_t) a2 *Yn2; |
emilmont | 1:fdd22bb7aa52 | 201 | |
emilmont | 1:fdd22bb7aa52 | 202 | /* The result is converted to 1.31, Yn2 variable is reused */ |
emilmont | 1:fdd22bb7aa52 | 203 | /* Calc lower part of acc */ |
emilmont | 1:fdd22bb7aa52 | 204 | acc_l = acc & 0xffffffff; |
emilmont | 1:fdd22bb7aa52 | 205 | |
emilmont | 1:fdd22bb7aa52 | 206 | /* Calc upper part of acc */ |
emilmont | 1:fdd22bb7aa52 | 207 | acc_h = (acc >> 32) & 0xffffffff; |
emilmont | 1:fdd22bb7aa52 | 208 | |
emilmont | 1:fdd22bb7aa52 | 209 | |
emilmont | 1:fdd22bb7aa52 | 210 | /* Apply shift for lower part of acc and upper part of acc */ |
emilmont | 1:fdd22bb7aa52 | 211 | Yn2 = (uint32_t) acc_l >> lShift | acc_h << uShift; |
emilmont | 1:fdd22bb7aa52 | 212 | |
emilmont | 1:fdd22bb7aa52 | 213 | /* Store the output in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 214 | *pOut++ = Yn2; |
emilmont | 1:fdd22bb7aa52 | 215 | |
emilmont | 1:fdd22bb7aa52 | 216 | /* Read the forth input */ |
emilmont | 1:fdd22bb7aa52 | 217 | Xn = *pIn++; |
emilmont | 1:fdd22bb7aa52 | 218 | |
emilmont | 1:fdd22bb7aa52 | 219 | /* acc = b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 220 | |
emilmont | 1:fdd22bb7aa52 | 221 | /* acc = b0 * x[n] */ |
emilmont | 1:fdd22bb7aa52 | 222 | acc = (q63_t) b0 *Xn; |
emilmont | 1:fdd22bb7aa52 | 223 | /* acc += b1 * x[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 224 | acc += (q63_t) b1 *Xn1; |
emilmont | 1:fdd22bb7aa52 | 225 | /* acc += b[2] * x[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 226 | acc += (q63_t) b2 *Xn2; |
emilmont | 1:fdd22bb7aa52 | 227 | /* acc += a1 * y[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 228 | acc += (q63_t) a1 *Yn2; |
emilmont | 1:fdd22bb7aa52 | 229 | /* acc += a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 230 | acc += (q63_t) a2 *Yn1; |
emilmont | 1:fdd22bb7aa52 | 231 | |
emilmont | 1:fdd22bb7aa52 | 232 | /* The result is converted to 1.31, Yn1 variable is reused */ |
emilmont | 1:fdd22bb7aa52 | 233 | /* Calc lower part of acc */ |
emilmont | 1:fdd22bb7aa52 | 234 | acc_l = acc & 0xffffffff; |
emilmont | 1:fdd22bb7aa52 | 235 | |
emilmont | 1:fdd22bb7aa52 | 236 | /* Calc upper part of acc */ |
emilmont | 1:fdd22bb7aa52 | 237 | acc_h = (acc >> 32) & 0xffffffff; |
emilmont | 1:fdd22bb7aa52 | 238 | |
emilmont | 1:fdd22bb7aa52 | 239 | /* Apply shift for lower part of acc and upper part of acc */ |
emilmont | 1:fdd22bb7aa52 | 240 | Yn1 = (uint32_t) acc_l >> lShift | acc_h << uShift; |
emilmont | 1:fdd22bb7aa52 | 241 | |
emilmont | 1:fdd22bb7aa52 | 242 | /* Every time after the output is computed state should be updated. */ |
emilmont | 1:fdd22bb7aa52 | 243 | /* The states should be updated as: */ |
emilmont | 1:fdd22bb7aa52 | 244 | /* Xn2 = Xn1 */ |
emilmont | 1:fdd22bb7aa52 | 245 | /* Xn1 = Xn */ |
emilmont | 1:fdd22bb7aa52 | 246 | /* Yn2 = Yn1 */ |
emilmont | 1:fdd22bb7aa52 | 247 | /* Yn1 = acc */ |
emilmont | 1:fdd22bb7aa52 | 248 | Xn2 = Xn1; |
emilmont | 1:fdd22bb7aa52 | 249 | Xn1 = Xn; |
emilmont | 1:fdd22bb7aa52 | 250 | |
emilmont | 1:fdd22bb7aa52 | 251 | /* Store the output in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 252 | *pOut++ = Yn1; |
emilmont | 1:fdd22bb7aa52 | 253 | |
emilmont | 1:fdd22bb7aa52 | 254 | /* decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 255 | sample--; |
emilmont | 1:fdd22bb7aa52 | 256 | } |
emilmont | 1:fdd22bb7aa52 | 257 | |
emilmont | 1:fdd22bb7aa52 | 258 | /* If the blockSize is not a multiple of 4, compute any remaining output samples here. |
emilmont | 1:fdd22bb7aa52 | 259 | ** No loop unrolling is used. */ |
emilmont | 1:fdd22bb7aa52 | 260 | sample = (blockSize & 0x3u); |
emilmont | 1:fdd22bb7aa52 | 261 | |
emilmont | 1:fdd22bb7aa52 | 262 | while(sample > 0u) |
emilmont | 1:fdd22bb7aa52 | 263 | { |
emilmont | 1:fdd22bb7aa52 | 264 | /* Read the input */ |
emilmont | 1:fdd22bb7aa52 | 265 | Xn = *pIn++; |
emilmont | 1:fdd22bb7aa52 | 266 | |
emilmont | 1:fdd22bb7aa52 | 267 | /* acc = b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 268 | |
emilmont | 1:fdd22bb7aa52 | 269 | /* acc = b0 * x[n] */ |
emilmont | 1:fdd22bb7aa52 | 270 | acc = (q63_t) b0 *Xn; |
emilmont | 1:fdd22bb7aa52 | 271 | /* acc += b1 * x[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 272 | acc += (q63_t) b1 *Xn1; |
emilmont | 1:fdd22bb7aa52 | 273 | /* acc += b[2] * x[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 274 | acc += (q63_t) b2 *Xn2; |
emilmont | 1:fdd22bb7aa52 | 275 | /* acc += a1 * y[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 276 | acc += (q63_t) a1 *Yn1; |
emilmont | 1:fdd22bb7aa52 | 277 | /* acc += a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 278 | acc += (q63_t) a2 *Yn2; |
emilmont | 1:fdd22bb7aa52 | 279 | |
emilmont | 1:fdd22bb7aa52 | 280 | /* The result is converted to 1.31 */ |
emilmont | 1:fdd22bb7aa52 | 281 | acc = acc >> lShift; |
emilmont | 1:fdd22bb7aa52 | 282 | |
emilmont | 1:fdd22bb7aa52 | 283 | /* Every time after the output is computed state should be updated. */ |
emilmont | 1:fdd22bb7aa52 | 284 | /* The states should be updated as: */ |
emilmont | 1:fdd22bb7aa52 | 285 | /* Xn2 = Xn1 */ |
emilmont | 1:fdd22bb7aa52 | 286 | /* Xn1 = Xn */ |
emilmont | 1:fdd22bb7aa52 | 287 | /* Yn2 = Yn1 */ |
emilmont | 1:fdd22bb7aa52 | 288 | /* Yn1 = acc */ |
emilmont | 1:fdd22bb7aa52 | 289 | Xn2 = Xn1; |
emilmont | 1:fdd22bb7aa52 | 290 | Xn1 = Xn; |
emilmont | 1:fdd22bb7aa52 | 291 | Yn2 = Yn1; |
emilmont | 1:fdd22bb7aa52 | 292 | Yn1 = (q31_t) acc; |
emilmont | 1:fdd22bb7aa52 | 293 | |
emilmont | 1:fdd22bb7aa52 | 294 | /* Store the output in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 295 | *pOut++ = (q31_t) acc; |
emilmont | 1:fdd22bb7aa52 | 296 | |
emilmont | 1:fdd22bb7aa52 | 297 | /* decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 298 | sample--; |
emilmont | 1:fdd22bb7aa52 | 299 | } |
emilmont | 1:fdd22bb7aa52 | 300 | |
emilmont | 1:fdd22bb7aa52 | 301 | /* The first stage goes from the input buffer to the output buffer. */ |
emilmont | 1:fdd22bb7aa52 | 302 | /* Subsequent stages occur in-place in the output buffer */ |
emilmont | 1:fdd22bb7aa52 | 303 | pIn = pDst; |
emilmont | 1:fdd22bb7aa52 | 304 | |
emilmont | 1:fdd22bb7aa52 | 305 | /* Reset to destination pointer */ |
emilmont | 1:fdd22bb7aa52 | 306 | pOut = pDst; |
emilmont | 1:fdd22bb7aa52 | 307 | |
emilmont | 1:fdd22bb7aa52 | 308 | /* Store the updated state variables back into the pState array */ |
emilmont | 1:fdd22bb7aa52 | 309 | *pState++ = Xn1; |
emilmont | 1:fdd22bb7aa52 | 310 | *pState++ = Xn2; |
emilmont | 1:fdd22bb7aa52 | 311 | *pState++ = Yn1; |
emilmont | 1:fdd22bb7aa52 | 312 | *pState++ = Yn2; |
emilmont | 1:fdd22bb7aa52 | 313 | |
emilmont | 1:fdd22bb7aa52 | 314 | } while(--stage); |
emilmont | 1:fdd22bb7aa52 | 315 | |
emilmont | 1:fdd22bb7aa52 | 316 | #else |
emilmont | 1:fdd22bb7aa52 | 317 | |
emilmont | 1:fdd22bb7aa52 | 318 | /* Run the below code for Cortex-M0 */ |
emilmont | 1:fdd22bb7aa52 | 319 | |
emilmont | 1:fdd22bb7aa52 | 320 | do |
emilmont | 1:fdd22bb7aa52 | 321 | { |
emilmont | 1:fdd22bb7aa52 | 322 | /* Reading the coefficients */ |
emilmont | 1:fdd22bb7aa52 | 323 | b0 = *pCoeffs++; |
emilmont | 1:fdd22bb7aa52 | 324 | b1 = *pCoeffs++; |
emilmont | 1:fdd22bb7aa52 | 325 | b2 = *pCoeffs++; |
emilmont | 1:fdd22bb7aa52 | 326 | a1 = *pCoeffs++; |
emilmont | 1:fdd22bb7aa52 | 327 | a2 = *pCoeffs++; |
emilmont | 1:fdd22bb7aa52 | 328 | |
emilmont | 1:fdd22bb7aa52 | 329 | /* Reading the state values */ |
emilmont | 1:fdd22bb7aa52 | 330 | Xn1 = pState[0]; |
emilmont | 1:fdd22bb7aa52 | 331 | Xn2 = pState[1]; |
emilmont | 1:fdd22bb7aa52 | 332 | Yn1 = pState[2]; |
emilmont | 1:fdd22bb7aa52 | 333 | Yn2 = pState[3]; |
emilmont | 1:fdd22bb7aa52 | 334 | |
emilmont | 1:fdd22bb7aa52 | 335 | /* The variables acc holds the output value that is computed: |
emilmont | 1:fdd22bb7aa52 | 336 | * acc = b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] |
emilmont | 1:fdd22bb7aa52 | 337 | */ |
emilmont | 1:fdd22bb7aa52 | 338 | |
emilmont | 1:fdd22bb7aa52 | 339 | sample = blockSize; |
emilmont | 1:fdd22bb7aa52 | 340 | |
emilmont | 1:fdd22bb7aa52 | 341 | while(sample > 0u) |
emilmont | 1:fdd22bb7aa52 | 342 | { |
emilmont | 1:fdd22bb7aa52 | 343 | /* Read the input */ |
emilmont | 1:fdd22bb7aa52 | 344 | Xn = *pIn++; |
emilmont | 1:fdd22bb7aa52 | 345 | |
emilmont | 1:fdd22bb7aa52 | 346 | /* acc = b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 347 | /* acc = b0 * x[n] */ |
emilmont | 1:fdd22bb7aa52 | 348 | acc = (q63_t) b0 *Xn; |
emilmont | 1:fdd22bb7aa52 | 349 | |
emilmont | 1:fdd22bb7aa52 | 350 | /* acc += b1 * x[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 351 | acc += (q63_t) b1 *Xn1; |
emilmont | 1:fdd22bb7aa52 | 352 | /* acc += b[2] * x[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 353 | acc += (q63_t) b2 *Xn2; |
emilmont | 1:fdd22bb7aa52 | 354 | /* acc += a1 * y[n-1] */ |
emilmont | 1:fdd22bb7aa52 | 355 | acc += (q63_t) a1 *Yn1; |
emilmont | 1:fdd22bb7aa52 | 356 | /* acc += a2 * y[n-2] */ |
emilmont | 1:fdd22bb7aa52 | 357 | acc += (q63_t) a2 *Yn2; |
emilmont | 1:fdd22bb7aa52 | 358 | |
emilmont | 1:fdd22bb7aa52 | 359 | /* The result is converted to 1.31 */ |
emilmont | 1:fdd22bb7aa52 | 360 | acc = acc >> lShift; |
emilmont | 1:fdd22bb7aa52 | 361 | |
emilmont | 1:fdd22bb7aa52 | 362 | /* Every time after the output is computed state should be updated. */ |
emilmont | 1:fdd22bb7aa52 | 363 | /* The states should be updated as: */ |
emilmont | 1:fdd22bb7aa52 | 364 | /* Xn2 = Xn1 */ |
emilmont | 1:fdd22bb7aa52 | 365 | /* Xn1 = Xn */ |
emilmont | 1:fdd22bb7aa52 | 366 | /* Yn2 = Yn1 */ |
emilmont | 1:fdd22bb7aa52 | 367 | /* Yn1 = acc */ |
emilmont | 1:fdd22bb7aa52 | 368 | Xn2 = Xn1; |
emilmont | 1:fdd22bb7aa52 | 369 | Xn1 = Xn; |
emilmont | 1:fdd22bb7aa52 | 370 | Yn2 = Yn1; |
emilmont | 1:fdd22bb7aa52 | 371 | Yn1 = (q31_t) acc; |
emilmont | 1:fdd22bb7aa52 | 372 | |
emilmont | 1:fdd22bb7aa52 | 373 | /* Store the output in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 374 | *pOut++ = (q31_t) acc; |
emilmont | 1:fdd22bb7aa52 | 375 | |
emilmont | 1:fdd22bb7aa52 | 376 | /* decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 377 | sample--; |
emilmont | 1:fdd22bb7aa52 | 378 | } |
emilmont | 1:fdd22bb7aa52 | 379 | |
emilmont | 1:fdd22bb7aa52 | 380 | /* The first stage goes from the input buffer to the output buffer. */ |
emilmont | 1:fdd22bb7aa52 | 381 | /* Subsequent stages occur in-place in the output buffer */ |
emilmont | 1:fdd22bb7aa52 | 382 | pIn = pDst; |
emilmont | 1:fdd22bb7aa52 | 383 | |
emilmont | 1:fdd22bb7aa52 | 384 | /* Reset to destination pointer */ |
emilmont | 1:fdd22bb7aa52 | 385 | pOut = pDst; |
emilmont | 1:fdd22bb7aa52 | 386 | |
emilmont | 1:fdd22bb7aa52 | 387 | /* Store the updated state variables back into the pState array */ |
emilmont | 1:fdd22bb7aa52 | 388 | *pState++ = Xn1; |
emilmont | 1:fdd22bb7aa52 | 389 | *pState++ = Xn2; |
emilmont | 1:fdd22bb7aa52 | 390 | *pState++ = Yn1; |
emilmont | 1:fdd22bb7aa52 | 391 | *pState++ = Yn2; |
emilmont | 1:fdd22bb7aa52 | 392 | |
emilmont | 1:fdd22bb7aa52 | 393 | } while(--stage); |
emilmont | 1:fdd22bb7aa52 | 394 | |
emilmont | 1:fdd22bb7aa52 | 395 | #endif /* #ifndef ARM_MATH_CM0 */ |
emilmont | 1:fdd22bb7aa52 | 396 | } |
emilmont | 1:fdd22bb7aa52 | 397 | |
emilmont | 1:fdd22bb7aa52 | 398 | /** |
emilmont | 1:fdd22bb7aa52 | 399 | * @} end of BiquadCascadeDF1 group |
emilmont | 1:fdd22bb7aa52 | 400 | */ |