V4.0.1 of the ARM CMSIS DSP libraries. Note that arm_bitreversal2.s, arm_cfft_f32.c and arm_rfft_fast_f32.c had to be removed. arm_bitreversal2.s will not assemble with the online tools. So, the fast f32 FFT functions are not yet available. All the other FFT functions are available.
Dependents: MPU9150_Example fir_f32 fir_f32 MPU9150_nucleo_noni2cdev ... more
FilteringFunctions/arm_fir_decimate_f32.c@0:3d9c67d97d6f, 2014-07-28 (annotated)
- Committer:
- emh203
- Date:
- Mon Jul 28 15:03:15 2014 +0000
- Revision:
- 0:3d9c67d97d6f
1st working commit. Had to remove arm_bitreversal2.s arm_cfft_f32.c and arm_rfft_fast_f32.c. The .s will not assemble. For now I removed these functions so we could at least have a library for the other functions.
Who changed what in which revision?
User | Revision | Line number | New contents of line |
---|---|---|---|
emh203 | 0:3d9c67d97d6f | 1 | /* ---------------------------------------------------------------------- |
emh203 | 0:3d9c67d97d6f | 2 | * Copyright (C) 2010-2014 ARM Limited. All rights reserved. |
emh203 | 0:3d9c67d97d6f | 3 | * |
emh203 | 0:3d9c67d97d6f | 4 | * $Date: 12. March 2014 |
emh203 | 0:3d9c67d97d6f | 5 | * $Revision: V1.4.3 |
emh203 | 0:3d9c67d97d6f | 6 | * |
emh203 | 0:3d9c67d97d6f | 7 | * Project: CMSIS DSP Library |
emh203 | 0:3d9c67d97d6f | 8 | * Title: arm_fir_decimate_f32.c |
emh203 | 0:3d9c67d97d6f | 9 | * |
emh203 | 0:3d9c67d97d6f | 10 | * Description: FIR decimation for floating-point sequences. |
emh203 | 0:3d9c67d97d6f | 11 | * |
emh203 | 0:3d9c67d97d6f | 12 | * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0 |
emh203 | 0:3d9c67d97d6f | 13 | * |
emh203 | 0:3d9c67d97d6f | 14 | * Redistribution and use in source and binary forms, with or without |
emh203 | 0:3d9c67d97d6f | 15 | * modification, are permitted provided that the following conditions |
emh203 | 0:3d9c67d97d6f | 16 | * are met: |
emh203 | 0:3d9c67d97d6f | 17 | * - Redistributions of source code must retain the above copyright |
emh203 | 0:3d9c67d97d6f | 18 | * notice, this list of conditions and the following disclaimer. |
emh203 | 0:3d9c67d97d6f | 19 | * - Redistributions in binary form must reproduce the above copyright |
emh203 | 0:3d9c67d97d6f | 20 | * notice, this list of conditions and the following disclaimer in |
emh203 | 0:3d9c67d97d6f | 21 | * the documentation and/or other materials provided with the |
emh203 | 0:3d9c67d97d6f | 22 | * distribution. |
emh203 | 0:3d9c67d97d6f | 23 | * - Neither the name of ARM LIMITED nor the names of its contributors |
emh203 | 0:3d9c67d97d6f | 24 | * may be used to endorse or promote products derived from this |
emh203 | 0:3d9c67d97d6f | 25 | * software without specific prior written permission. |
emh203 | 0:3d9c67d97d6f | 26 | * |
emh203 | 0:3d9c67d97d6f | 27 | * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
emh203 | 0:3d9c67d97d6f | 28 | * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
emh203 | 0:3d9c67d97d6f | 29 | * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS |
emh203 | 0:3d9c67d97d6f | 30 | * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE |
emh203 | 0:3d9c67d97d6f | 31 | * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, |
emh203 | 0:3d9c67d97d6f | 32 | * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, |
emh203 | 0:3d9c67d97d6f | 33 | * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; |
emh203 | 0:3d9c67d97d6f | 34 | * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER |
emh203 | 0:3d9c67d97d6f | 35 | * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
emh203 | 0:3d9c67d97d6f | 36 | * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN |
emh203 | 0:3d9c67d97d6f | 37 | * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE |
emh203 | 0:3d9c67d97d6f | 38 | * POSSIBILITY OF SUCH DAMAGE. |
emh203 | 0:3d9c67d97d6f | 39 | * -------------------------------------------------------------------- */ |
emh203 | 0:3d9c67d97d6f | 40 | |
emh203 | 0:3d9c67d97d6f | 41 | #include "arm_math.h" |
emh203 | 0:3d9c67d97d6f | 42 | |
emh203 | 0:3d9c67d97d6f | 43 | /** |
emh203 | 0:3d9c67d97d6f | 44 | * @ingroup groupFilters |
emh203 | 0:3d9c67d97d6f | 45 | */ |
emh203 | 0:3d9c67d97d6f | 46 | |
emh203 | 0:3d9c67d97d6f | 47 | /** |
emh203 | 0:3d9c67d97d6f | 48 | * @defgroup FIR_decimate Finite Impulse Response (FIR) Decimator |
emh203 | 0:3d9c67d97d6f | 49 | * |
emh203 | 0:3d9c67d97d6f | 50 | * These functions combine an FIR filter together with a decimator. |
emh203 | 0:3d9c67d97d6f | 51 | * They are used in multirate systems for reducing the sample rate of a signal without introducing aliasing distortion. |
emh203 | 0:3d9c67d97d6f | 52 | * Conceptually, the functions are equivalent to the block diagram below: |
emh203 | 0:3d9c67d97d6f | 53 | * \image html FIRDecimator.gif "Components included in the FIR Decimator functions" |
emh203 | 0:3d9c67d97d6f | 54 | * When decimating by a factor of <code>M</code>, the signal should be prefiltered by a lowpass filter with a normalized |
emh203 | 0:3d9c67d97d6f | 55 | * cutoff frequency of <code>1/M</code> in order to prevent aliasing distortion. |
emh203 | 0:3d9c67d97d6f | 56 | * The user of the function is responsible for providing the filter coefficients. |
emh203 | 0:3d9c67d97d6f | 57 | * |
emh203 | 0:3d9c67d97d6f | 58 | * The FIR decimator functions provided in the CMSIS DSP Library combine the FIR filter and the decimator in an efficient manner. |
emh203 | 0:3d9c67d97d6f | 59 | * Instead of calculating all of the FIR filter outputs and discarding <code>M-1</code> out of every <code>M</code>, only the |
emh203 | 0:3d9c67d97d6f | 60 | * samples output by the decimator are computed. |
emh203 | 0:3d9c67d97d6f | 61 | * The functions operate on blocks of input and output data. |
emh203 | 0:3d9c67d97d6f | 62 | * <code>pSrc</code> points to an array of <code>blockSize</code> input values and |
emh203 | 0:3d9c67d97d6f | 63 | * <code>pDst</code> points to an array of <code>blockSize/M</code> output values. |
emh203 | 0:3d9c67d97d6f | 64 | * In order to have an integer number of output samples <code>blockSize</code> |
emh203 | 0:3d9c67d97d6f | 65 | * must always be a multiple of the decimation factor <code>M</code>. |
emh203 | 0:3d9c67d97d6f | 66 | * |
emh203 | 0:3d9c67d97d6f | 67 | * The library provides separate functions for Q15, Q31 and floating-point data types. |
emh203 | 0:3d9c67d97d6f | 68 | * |
emh203 | 0:3d9c67d97d6f | 69 | * \par Algorithm: |
emh203 | 0:3d9c67d97d6f | 70 | * The FIR portion of the algorithm uses the standard form filter: |
emh203 | 0:3d9c67d97d6f | 71 | * <pre> |
emh203 | 0:3d9c67d97d6f | 72 | * y[n] = b[0] * x[n] + b[1] * x[n-1] + b[2] * x[n-2] + ...+ b[numTaps-1] * x[n-numTaps+1] |
emh203 | 0:3d9c67d97d6f | 73 | * </pre> |
emh203 | 0:3d9c67d97d6f | 74 | * where, <code>b[n]</code> are the filter coefficients. |
emh203 | 0:3d9c67d97d6f | 75 | * \par |
emh203 | 0:3d9c67d97d6f | 76 | * The <code>pCoeffs</code> points to a coefficient array of size <code>numTaps</code>. |
emh203 | 0:3d9c67d97d6f | 77 | * Coefficients are stored in time reversed order. |
emh203 | 0:3d9c67d97d6f | 78 | * \par |
emh203 | 0:3d9c67d97d6f | 79 | * <pre> |
emh203 | 0:3d9c67d97d6f | 80 | * {b[numTaps-1], b[numTaps-2], b[N-2], ..., b[1], b[0]} |
emh203 | 0:3d9c67d97d6f | 81 | * </pre> |
emh203 | 0:3d9c67d97d6f | 82 | * \par |
emh203 | 0:3d9c67d97d6f | 83 | * <code>pState</code> points to a state array of size <code>numTaps + blockSize - 1</code>. |
emh203 | 0:3d9c67d97d6f | 84 | * Samples in the state buffer are stored in the order: |
emh203 | 0:3d9c67d97d6f | 85 | * \par |
emh203 | 0:3d9c67d97d6f | 86 | * <pre> |
emh203 | 0:3d9c67d97d6f | 87 | * {x[n-numTaps+1], x[n-numTaps], x[n-numTaps-1], x[n-numTaps-2]....x[0], x[1], ..., x[blockSize-1]} |
emh203 | 0:3d9c67d97d6f | 88 | * </pre> |
emh203 | 0:3d9c67d97d6f | 89 | * The state variables are updated after each block of data is processed, the coefficients are untouched. |
emh203 | 0:3d9c67d97d6f | 90 | * |
emh203 | 0:3d9c67d97d6f | 91 | * \par Instance Structure |
emh203 | 0:3d9c67d97d6f | 92 | * The coefficients and state variables for a filter are stored together in an instance data structure. |
emh203 | 0:3d9c67d97d6f | 93 | * A separate instance structure must be defined for each filter. |
emh203 | 0:3d9c67d97d6f | 94 | * Coefficient arrays may be shared among several instances while state variable array should be allocated separately. |
emh203 | 0:3d9c67d97d6f | 95 | * There are separate instance structure declarations for each of the 3 supported data types. |
emh203 | 0:3d9c67d97d6f | 96 | * |
emh203 | 0:3d9c67d97d6f | 97 | * \par Initialization Functions |
emh203 | 0:3d9c67d97d6f | 98 | * There is also an associated initialization function for each data type. |
emh203 | 0:3d9c67d97d6f | 99 | * The initialization function performs the following operations: |
emh203 | 0:3d9c67d97d6f | 100 | * - Sets the values of the internal structure fields. |
emh203 | 0:3d9c67d97d6f | 101 | * - Zeros out the values in the state buffer. |
emh203 | 0:3d9c67d97d6f | 102 | * - Checks to make sure that the size of the input is a multiple of the decimation factor. |
emh203 | 0:3d9c67d97d6f | 103 | * To do this manually without calling the init function, assign the follow subfields of the instance structure: |
emh203 | 0:3d9c67d97d6f | 104 | * numTaps, pCoeffs, M (decimation factor), pState. Also set all of the values in pState to zero. |
emh203 | 0:3d9c67d97d6f | 105 | * |
emh203 | 0:3d9c67d97d6f | 106 | * \par |
emh203 | 0:3d9c67d97d6f | 107 | * Use of the initialization function is optional. |
emh203 | 0:3d9c67d97d6f | 108 | * However, if the initialization function is used, then the instance structure cannot be placed into a const data section. |
emh203 | 0:3d9c67d97d6f | 109 | * To place an instance structure into a const data section, the instance structure must be manually initialized. |
emh203 | 0:3d9c67d97d6f | 110 | * The code below statically initializes each of the 3 different data type filter instance structures |
emh203 | 0:3d9c67d97d6f | 111 | * <pre> |
emh203 | 0:3d9c67d97d6f | 112 | *arm_fir_decimate_instance_f32 S = {M, numTaps, pCoeffs, pState}; |
emh203 | 0:3d9c67d97d6f | 113 | *arm_fir_decimate_instance_q31 S = {M, numTaps, pCoeffs, pState}; |
emh203 | 0:3d9c67d97d6f | 114 | *arm_fir_decimate_instance_q15 S = {M, numTaps, pCoeffs, pState}; |
emh203 | 0:3d9c67d97d6f | 115 | * </pre> |
emh203 | 0:3d9c67d97d6f | 116 | * where <code>M</code> is the decimation factor; <code>numTaps</code> is the number of filter coefficients in the filter; |
emh203 | 0:3d9c67d97d6f | 117 | * <code>pCoeffs</code> is the address of the coefficient buffer; |
emh203 | 0:3d9c67d97d6f | 118 | * <code>pState</code> is the address of the state buffer. |
emh203 | 0:3d9c67d97d6f | 119 | * Be sure to set the values in the state buffer to zeros when doing static initialization. |
emh203 | 0:3d9c67d97d6f | 120 | * |
emh203 | 0:3d9c67d97d6f | 121 | * \par Fixed-Point Behavior |
emh203 | 0:3d9c67d97d6f | 122 | * Care must be taken when using the fixed-point versions of the FIR decimate filter functions. |
emh203 | 0:3d9c67d97d6f | 123 | * In particular, the overflow and saturation behavior of the accumulator used in each function must be considered. |
emh203 | 0:3d9c67d97d6f | 124 | * Refer to the function specific documentation below for usage guidelines. |
emh203 | 0:3d9c67d97d6f | 125 | */ |
emh203 | 0:3d9c67d97d6f | 126 | |
emh203 | 0:3d9c67d97d6f | 127 | /** |
emh203 | 0:3d9c67d97d6f | 128 | * @addtogroup FIR_decimate |
emh203 | 0:3d9c67d97d6f | 129 | * @{ |
emh203 | 0:3d9c67d97d6f | 130 | */ |
emh203 | 0:3d9c67d97d6f | 131 | |
emh203 | 0:3d9c67d97d6f | 132 | /** |
emh203 | 0:3d9c67d97d6f | 133 | * @brief Processing function for the floating-point FIR decimator. |
emh203 | 0:3d9c67d97d6f | 134 | * @param[in] *S points to an instance of the floating-point FIR decimator structure. |
emh203 | 0:3d9c67d97d6f | 135 | * @param[in] *pSrc points to the block of input data. |
emh203 | 0:3d9c67d97d6f | 136 | * @param[out] *pDst points to the block of output data. |
emh203 | 0:3d9c67d97d6f | 137 | * @param[in] blockSize number of input samples to process per call. |
emh203 | 0:3d9c67d97d6f | 138 | * @return none. |
emh203 | 0:3d9c67d97d6f | 139 | */ |
emh203 | 0:3d9c67d97d6f | 140 | |
emh203 | 0:3d9c67d97d6f | 141 | void arm_fir_decimate_f32( |
emh203 | 0:3d9c67d97d6f | 142 | const arm_fir_decimate_instance_f32 * S, |
emh203 | 0:3d9c67d97d6f | 143 | float32_t * pSrc, |
emh203 | 0:3d9c67d97d6f | 144 | float32_t * pDst, |
emh203 | 0:3d9c67d97d6f | 145 | uint32_t blockSize) |
emh203 | 0:3d9c67d97d6f | 146 | { |
emh203 | 0:3d9c67d97d6f | 147 | float32_t *pState = S->pState; /* State pointer */ |
emh203 | 0:3d9c67d97d6f | 148 | float32_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */ |
emh203 | 0:3d9c67d97d6f | 149 | float32_t *pStateCurnt; /* Points to the current sample of the state */ |
emh203 | 0:3d9c67d97d6f | 150 | float32_t *px, *pb; /* Temporary pointers for state and coefficient buffers */ |
emh203 | 0:3d9c67d97d6f | 151 | float32_t sum0; /* Accumulator */ |
emh203 | 0:3d9c67d97d6f | 152 | float32_t x0, c0; /* Temporary variables to hold state and coefficient values */ |
emh203 | 0:3d9c67d97d6f | 153 | uint32_t numTaps = S->numTaps; /* Number of filter coefficients in the filter */ |
emh203 | 0:3d9c67d97d6f | 154 | uint32_t i, tapCnt, blkCnt, outBlockSize = blockSize / S->M; /* Loop counters */ |
emh203 | 0:3d9c67d97d6f | 155 | |
emh203 | 0:3d9c67d97d6f | 156 | #ifndef ARM_MATH_CM0_FAMILY |
emh203 | 0:3d9c67d97d6f | 157 | |
emh203 | 0:3d9c67d97d6f | 158 | uint32_t blkCntN4; |
emh203 | 0:3d9c67d97d6f | 159 | float32_t *px0, *px1, *px2, *px3; |
emh203 | 0:3d9c67d97d6f | 160 | float32_t acc0, acc1, acc2, acc3; |
emh203 | 0:3d9c67d97d6f | 161 | float32_t x1, x2, x3; |
emh203 | 0:3d9c67d97d6f | 162 | |
emh203 | 0:3d9c67d97d6f | 163 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
emh203 | 0:3d9c67d97d6f | 164 | |
emh203 | 0:3d9c67d97d6f | 165 | /* S->pState buffer contains previous frame (numTaps - 1) samples */ |
emh203 | 0:3d9c67d97d6f | 166 | /* pStateCurnt points to the location where the new input data should be written */ |
emh203 | 0:3d9c67d97d6f | 167 | pStateCurnt = S->pState + (numTaps - 1u); |
emh203 | 0:3d9c67d97d6f | 168 | |
emh203 | 0:3d9c67d97d6f | 169 | /* Total number of output samples to be computed */ |
emh203 | 0:3d9c67d97d6f | 170 | blkCnt = outBlockSize / 4; |
emh203 | 0:3d9c67d97d6f | 171 | blkCntN4 = outBlockSize - (4 * blkCnt); |
emh203 | 0:3d9c67d97d6f | 172 | |
emh203 | 0:3d9c67d97d6f | 173 | while(blkCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 174 | { |
emh203 | 0:3d9c67d97d6f | 175 | /* Copy 4 * decimation factor number of new input samples into the state buffer */ |
emh203 | 0:3d9c67d97d6f | 176 | i = 4 * S->M; |
emh203 | 0:3d9c67d97d6f | 177 | |
emh203 | 0:3d9c67d97d6f | 178 | do |
emh203 | 0:3d9c67d97d6f | 179 | { |
emh203 | 0:3d9c67d97d6f | 180 | *pStateCurnt++ = *pSrc++; |
emh203 | 0:3d9c67d97d6f | 181 | |
emh203 | 0:3d9c67d97d6f | 182 | } while(--i); |
emh203 | 0:3d9c67d97d6f | 183 | |
emh203 | 0:3d9c67d97d6f | 184 | /* Set accumulators to zero */ |
emh203 | 0:3d9c67d97d6f | 185 | acc0 = 0.0f; |
emh203 | 0:3d9c67d97d6f | 186 | acc1 = 0.0f; |
emh203 | 0:3d9c67d97d6f | 187 | acc2 = 0.0f; |
emh203 | 0:3d9c67d97d6f | 188 | acc3 = 0.0f; |
emh203 | 0:3d9c67d97d6f | 189 | |
emh203 | 0:3d9c67d97d6f | 190 | /* Initialize state pointer for all the samples */ |
emh203 | 0:3d9c67d97d6f | 191 | px0 = pState; |
emh203 | 0:3d9c67d97d6f | 192 | px1 = pState + S->M; |
emh203 | 0:3d9c67d97d6f | 193 | px2 = pState + 2 * S->M; |
emh203 | 0:3d9c67d97d6f | 194 | px3 = pState + 3 * S->M; |
emh203 | 0:3d9c67d97d6f | 195 | |
emh203 | 0:3d9c67d97d6f | 196 | /* Initialize coeff pointer */ |
emh203 | 0:3d9c67d97d6f | 197 | pb = pCoeffs; |
emh203 | 0:3d9c67d97d6f | 198 | |
emh203 | 0:3d9c67d97d6f | 199 | /* Loop unrolling. Process 4 taps at a time. */ |
emh203 | 0:3d9c67d97d6f | 200 | tapCnt = numTaps >> 2; |
emh203 | 0:3d9c67d97d6f | 201 | |
emh203 | 0:3d9c67d97d6f | 202 | /* Loop over the number of taps. Unroll by a factor of 4. |
emh203 | 0:3d9c67d97d6f | 203 | ** Repeat until we've computed numTaps-4 coefficients. */ |
emh203 | 0:3d9c67d97d6f | 204 | |
emh203 | 0:3d9c67d97d6f | 205 | while(tapCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 206 | { |
emh203 | 0:3d9c67d97d6f | 207 | /* Read the b[numTaps-1] coefficient */ |
emh203 | 0:3d9c67d97d6f | 208 | c0 = *(pb++); |
emh203 | 0:3d9c67d97d6f | 209 | |
emh203 | 0:3d9c67d97d6f | 210 | /* Read x[n-numTaps-1] sample for acc0 */ |
emh203 | 0:3d9c67d97d6f | 211 | x0 = *(px0++); |
emh203 | 0:3d9c67d97d6f | 212 | /* Read x[n-numTaps-1] sample for acc1 */ |
emh203 | 0:3d9c67d97d6f | 213 | x1 = *(px1++); |
emh203 | 0:3d9c67d97d6f | 214 | /* Read x[n-numTaps-1] sample for acc2 */ |
emh203 | 0:3d9c67d97d6f | 215 | x2 = *(px2++); |
emh203 | 0:3d9c67d97d6f | 216 | /* Read x[n-numTaps-1] sample for acc3 */ |
emh203 | 0:3d9c67d97d6f | 217 | x3 = *(px3++); |
emh203 | 0:3d9c67d97d6f | 218 | |
emh203 | 0:3d9c67d97d6f | 219 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 220 | acc0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 221 | acc1 += x1 * c0; |
emh203 | 0:3d9c67d97d6f | 222 | acc2 += x2 * c0; |
emh203 | 0:3d9c67d97d6f | 223 | acc3 += x3 * c0; |
emh203 | 0:3d9c67d97d6f | 224 | |
emh203 | 0:3d9c67d97d6f | 225 | /* Read the b[numTaps-2] coefficient */ |
emh203 | 0:3d9c67d97d6f | 226 | c0 = *(pb++); |
emh203 | 0:3d9c67d97d6f | 227 | |
emh203 | 0:3d9c67d97d6f | 228 | /* Read x[n-numTaps-2] sample for acc0, acc1, acc2, acc3 */ |
emh203 | 0:3d9c67d97d6f | 229 | x0 = *(px0++); |
emh203 | 0:3d9c67d97d6f | 230 | x1 = *(px1++); |
emh203 | 0:3d9c67d97d6f | 231 | x2 = *(px2++); |
emh203 | 0:3d9c67d97d6f | 232 | x3 = *(px3++); |
emh203 | 0:3d9c67d97d6f | 233 | |
emh203 | 0:3d9c67d97d6f | 234 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 235 | acc0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 236 | acc1 += x1 * c0; |
emh203 | 0:3d9c67d97d6f | 237 | acc2 += x2 * c0; |
emh203 | 0:3d9c67d97d6f | 238 | acc3 += x3 * c0; |
emh203 | 0:3d9c67d97d6f | 239 | |
emh203 | 0:3d9c67d97d6f | 240 | /* Read the b[numTaps-3] coefficient */ |
emh203 | 0:3d9c67d97d6f | 241 | c0 = *(pb++); |
emh203 | 0:3d9c67d97d6f | 242 | |
emh203 | 0:3d9c67d97d6f | 243 | /* Read x[n-numTaps-3] sample acc0, acc1, acc2, acc3 */ |
emh203 | 0:3d9c67d97d6f | 244 | x0 = *(px0++); |
emh203 | 0:3d9c67d97d6f | 245 | x1 = *(px1++); |
emh203 | 0:3d9c67d97d6f | 246 | x2 = *(px2++); |
emh203 | 0:3d9c67d97d6f | 247 | x3 = *(px3++); |
emh203 | 0:3d9c67d97d6f | 248 | |
emh203 | 0:3d9c67d97d6f | 249 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 250 | acc0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 251 | acc1 += x1 * c0; |
emh203 | 0:3d9c67d97d6f | 252 | acc2 += x2 * c0; |
emh203 | 0:3d9c67d97d6f | 253 | acc3 += x3 * c0; |
emh203 | 0:3d9c67d97d6f | 254 | |
emh203 | 0:3d9c67d97d6f | 255 | /* Read the b[numTaps-4] coefficient */ |
emh203 | 0:3d9c67d97d6f | 256 | c0 = *(pb++); |
emh203 | 0:3d9c67d97d6f | 257 | |
emh203 | 0:3d9c67d97d6f | 258 | /* Read x[n-numTaps-4] sample acc0, acc1, acc2, acc3 */ |
emh203 | 0:3d9c67d97d6f | 259 | x0 = *(px0++); |
emh203 | 0:3d9c67d97d6f | 260 | x1 = *(px1++); |
emh203 | 0:3d9c67d97d6f | 261 | x2 = *(px2++); |
emh203 | 0:3d9c67d97d6f | 262 | x3 = *(px3++); |
emh203 | 0:3d9c67d97d6f | 263 | |
emh203 | 0:3d9c67d97d6f | 264 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 265 | acc0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 266 | acc1 += x1 * c0; |
emh203 | 0:3d9c67d97d6f | 267 | acc2 += x2 * c0; |
emh203 | 0:3d9c67d97d6f | 268 | acc3 += x3 * c0; |
emh203 | 0:3d9c67d97d6f | 269 | |
emh203 | 0:3d9c67d97d6f | 270 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 271 | tapCnt--; |
emh203 | 0:3d9c67d97d6f | 272 | } |
emh203 | 0:3d9c67d97d6f | 273 | |
emh203 | 0:3d9c67d97d6f | 274 | /* If the filter length is not a multiple of 4, compute the remaining filter taps */ |
emh203 | 0:3d9c67d97d6f | 275 | tapCnt = numTaps % 0x4u; |
emh203 | 0:3d9c67d97d6f | 276 | |
emh203 | 0:3d9c67d97d6f | 277 | while(tapCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 278 | { |
emh203 | 0:3d9c67d97d6f | 279 | /* Read coefficients */ |
emh203 | 0:3d9c67d97d6f | 280 | c0 = *(pb++); |
emh203 | 0:3d9c67d97d6f | 281 | |
emh203 | 0:3d9c67d97d6f | 282 | /* Fetch state variables for acc0, acc1, acc2, acc3 */ |
emh203 | 0:3d9c67d97d6f | 283 | x0 = *(px0++); |
emh203 | 0:3d9c67d97d6f | 284 | x1 = *(px1++); |
emh203 | 0:3d9c67d97d6f | 285 | x2 = *(px2++); |
emh203 | 0:3d9c67d97d6f | 286 | x3 = *(px3++); |
emh203 | 0:3d9c67d97d6f | 287 | |
emh203 | 0:3d9c67d97d6f | 288 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 289 | acc0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 290 | acc1 += x1 * c0; |
emh203 | 0:3d9c67d97d6f | 291 | acc2 += x2 * c0; |
emh203 | 0:3d9c67d97d6f | 292 | acc3 += x3 * c0; |
emh203 | 0:3d9c67d97d6f | 293 | |
emh203 | 0:3d9c67d97d6f | 294 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 295 | tapCnt--; |
emh203 | 0:3d9c67d97d6f | 296 | } |
emh203 | 0:3d9c67d97d6f | 297 | |
emh203 | 0:3d9c67d97d6f | 298 | /* Advance the state pointer by the decimation factor |
emh203 | 0:3d9c67d97d6f | 299 | * to process the next group of decimation factor number samples */ |
emh203 | 0:3d9c67d97d6f | 300 | pState = pState + 4 * S->M; |
emh203 | 0:3d9c67d97d6f | 301 | |
emh203 | 0:3d9c67d97d6f | 302 | /* The result is in the accumulator, store in the destination buffer. */ |
emh203 | 0:3d9c67d97d6f | 303 | *pDst++ = acc0; |
emh203 | 0:3d9c67d97d6f | 304 | *pDst++ = acc1; |
emh203 | 0:3d9c67d97d6f | 305 | *pDst++ = acc2; |
emh203 | 0:3d9c67d97d6f | 306 | *pDst++ = acc3; |
emh203 | 0:3d9c67d97d6f | 307 | |
emh203 | 0:3d9c67d97d6f | 308 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 309 | blkCnt--; |
emh203 | 0:3d9c67d97d6f | 310 | } |
emh203 | 0:3d9c67d97d6f | 311 | |
emh203 | 0:3d9c67d97d6f | 312 | while(blkCntN4 > 0u) |
emh203 | 0:3d9c67d97d6f | 313 | { |
emh203 | 0:3d9c67d97d6f | 314 | /* Copy decimation factor number of new input samples into the state buffer */ |
emh203 | 0:3d9c67d97d6f | 315 | i = S->M; |
emh203 | 0:3d9c67d97d6f | 316 | |
emh203 | 0:3d9c67d97d6f | 317 | do |
emh203 | 0:3d9c67d97d6f | 318 | { |
emh203 | 0:3d9c67d97d6f | 319 | *pStateCurnt++ = *pSrc++; |
emh203 | 0:3d9c67d97d6f | 320 | |
emh203 | 0:3d9c67d97d6f | 321 | } while(--i); |
emh203 | 0:3d9c67d97d6f | 322 | |
emh203 | 0:3d9c67d97d6f | 323 | /* Set accumulator to zero */ |
emh203 | 0:3d9c67d97d6f | 324 | sum0 = 0.0f; |
emh203 | 0:3d9c67d97d6f | 325 | |
emh203 | 0:3d9c67d97d6f | 326 | /* Initialize state pointer */ |
emh203 | 0:3d9c67d97d6f | 327 | px = pState; |
emh203 | 0:3d9c67d97d6f | 328 | |
emh203 | 0:3d9c67d97d6f | 329 | /* Initialize coeff pointer */ |
emh203 | 0:3d9c67d97d6f | 330 | pb = pCoeffs; |
emh203 | 0:3d9c67d97d6f | 331 | |
emh203 | 0:3d9c67d97d6f | 332 | /* Loop unrolling. Process 4 taps at a time. */ |
emh203 | 0:3d9c67d97d6f | 333 | tapCnt = numTaps >> 2; |
emh203 | 0:3d9c67d97d6f | 334 | |
emh203 | 0:3d9c67d97d6f | 335 | /* Loop over the number of taps. Unroll by a factor of 4. |
emh203 | 0:3d9c67d97d6f | 336 | ** Repeat until we've computed numTaps-4 coefficients. */ |
emh203 | 0:3d9c67d97d6f | 337 | while(tapCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 338 | { |
emh203 | 0:3d9c67d97d6f | 339 | /* Read the b[numTaps-1] coefficient */ |
emh203 | 0:3d9c67d97d6f | 340 | c0 = *(pb++); |
emh203 | 0:3d9c67d97d6f | 341 | |
emh203 | 0:3d9c67d97d6f | 342 | /* Read x[n-numTaps-1] sample */ |
emh203 | 0:3d9c67d97d6f | 343 | x0 = *(px++); |
emh203 | 0:3d9c67d97d6f | 344 | |
emh203 | 0:3d9c67d97d6f | 345 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 346 | sum0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 347 | |
emh203 | 0:3d9c67d97d6f | 348 | /* Read the b[numTaps-2] coefficient */ |
emh203 | 0:3d9c67d97d6f | 349 | c0 = *(pb++); |
emh203 | 0:3d9c67d97d6f | 350 | |
emh203 | 0:3d9c67d97d6f | 351 | /* Read x[n-numTaps-2] sample */ |
emh203 | 0:3d9c67d97d6f | 352 | x0 = *(px++); |
emh203 | 0:3d9c67d97d6f | 353 | |
emh203 | 0:3d9c67d97d6f | 354 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 355 | sum0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 356 | |
emh203 | 0:3d9c67d97d6f | 357 | /* Read the b[numTaps-3] coefficient */ |
emh203 | 0:3d9c67d97d6f | 358 | c0 = *(pb++); |
emh203 | 0:3d9c67d97d6f | 359 | |
emh203 | 0:3d9c67d97d6f | 360 | /* Read x[n-numTaps-3] sample */ |
emh203 | 0:3d9c67d97d6f | 361 | x0 = *(px++); |
emh203 | 0:3d9c67d97d6f | 362 | |
emh203 | 0:3d9c67d97d6f | 363 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 364 | sum0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 365 | |
emh203 | 0:3d9c67d97d6f | 366 | /* Read the b[numTaps-4] coefficient */ |
emh203 | 0:3d9c67d97d6f | 367 | c0 = *(pb++); |
emh203 | 0:3d9c67d97d6f | 368 | |
emh203 | 0:3d9c67d97d6f | 369 | /* Read x[n-numTaps-4] sample */ |
emh203 | 0:3d9c67d97d6f | 370 | x0 = *(px++); |
emh203 | 0:3d9c67d97d6f | 371 | |
emh203 | 0:3d9c67d97d6f | 372 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 373 | sum0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 374 | |
emh203 | 0:3d9c67d97d6f | 375 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 376 | tapCnt--; |
emh203 | 0:3d9c67d97d6f | 377 | } |
emh203 | 0:3d9c67d97d6f | 378 | |
emh203 | 0:3d9c67d97d6f | 379 | /* If the filter length is not a multiple of 4, compute the remaining filter taps */ |
emh203 | 0:3d9c67d97d6f | 380 | tapCnt = numTaps % 0x4u; |
emh203 | 0:3d9c67d97d6f | 381 | |
emh203 | 0:3d9c67d97d6f | 382 | while(tapCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 383 | { |
emh203 | 0:3d9c67d97d6f | 384 | /* Read coefficients */ |
emh203 | 0:3d9c67d97d6f | 385 | c0 = *(pb++); |
emh203 | 0:3d9c67d97d6f | 386 | |
emh203 | 0:3d9c67d97d6f | 387 | /* Fetch 1 state variable */ |
emh203 | 0:3d9c67d97d6f | 388 | x0 = *(px++); |
emh203 | 0:3d9c67d97d6f | 389 | |
emh203 | 0:3d9c67d97d6f | 390 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 391 | sum0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 392 | |
emh203 | 0:3d9c67d97d6f | 393 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 394 | tapCnt--; |
emh203 | 0:3d9c67d97d6f | 395 | } |
emh203 | 0:3d9c67d97d6f | 396 | |
emh203 | 0:3d9c67d97d6f | 397 | /* Advance the state pointer by the decimation factor |
emh203 | 0:3d9c67d97d6f | 398 | * to process the next group of decimation factor number samples */ |
emh203 | 0:3d9c67d97d6f | 399 | pState = pState + S->M; |
emh203 | 0:3d9c67d97d6f | 400 | |
emh203 | 0:3d9c67d97d6f | 401 | /* The result is in the accumulator, store in the destination buffer. */ |
emh203 | 0:3d9c67d97d6f | 402 | *pDst++ = sum0; |
emh203 | 0:3d9c67d97d6f | 403 | |
emh203 | 0:3d9c67d97d6f | 404 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 405 | blkCntN4--; |
emh203 | 0:3d9c67d97d6f | 406 | } |
emh203 | 0:3d9c67d97d6f | 407 | |
emh203 | 0:3d9c67d97d6f | 408 | /* Processing is complete. |
emh203 | 0:3d9c67d97d6f | 409 | ** Now copy the last numTaps - 1 samples to the satrt of the state buffer. |
emh203 | 0:3d9c67d97d6f | 410 | ** This prepares the state buffer for the next function call. */ |
emh203 | 0:3d9c67d97d6f | 411 | |
emh203 | 0:3d9c67d97d6f | 412 | /* Points to the start of the state buffer */ |
emh203 | 0:3d9c67d97d6f | 413 | pStateCurnt = S->pState; |
emh203 | 0:3d9c67d97d6f | 414 | |
emh203 | 0:3d9c67d97d6f | 415 | i = (numTaps - 1u) >> 2; |
emh203 | 0:3d9c67d97d6f | 416 | |
emh203 | 0:3d9c67d97d6f | 417 | /* copy data */ |
emh203 | 0:3d9c67d97d6f | 418 | while(i > 0u) |
emh203 | 0:3d9c67d97d6f | 419 | { |
emh203 | 0:3d9c67d97d6f | 420 | *pStateCurnt++ = *pState++; |
emh203 | 0:3d9c67d97d6f | 421 | *pStateCurnt++ = *pState++; |
emh203 | 0:3d9c67d97d6f | 422 | *pStateCurnt++ = *pState++; |
emh203 | 0:3d9c67d97d6f | 423 | *pStateCurnt++ = *pState++; |
emh203 | 0:3d9c67d97d6f | 424 | |
emh203 | 0:3d9c67d97d6f | 425 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 426 | i--; |
emh203 | 0:3d9c67d97d6f | 427 | } |
emh203 | 0:3d9c67d97d6f | 428 | |
emh203 | 0:3d9c67d97d6f | 429 | i = (numTaps - 1u) % 0x04u; |
emh203 | 0:3d9c67d97d6f | 430 | |
emh203 | 0:3d9c67d97d6f | 431 | /* copy data */ |
emh203 | 0:3d9c67d97d6f | 432 | while(i > 0u) |
emh203 | 0:3d9c67d97d6f | 433 | { |
emh203 | 0:3d9c67d97d6f | 434 | *pStateCurnt++ = *pState++; |
emh203 | 0:3d9c67d97d6f | 435 | |
emh203 | 0:3d9c67d97d6f | 436 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 437 | i--; |
emh203 | 0:3d9c67d97d6f | 438 | } |
emh203 | 0:3d9c67d97d6f | 439 | |
emh203 | 0:3d9c67d97d6f | 440 | #else |
emh203 | 0:3d9c67d97d6f | 441 | |
emh203 | 0:3d9c67d97d6f | 442 | /* Run the below code for Cortex-M0 */ |
emh203 | 0:3d9c67d97d6f | 443 | |
emh203 | 0:3d9c67d97d6f | 444 | /* S->pState buffer contains previous frame (numTaps - 1) samples */ |
emh203 | 0:3d9c67d97d6f | 445 | /* pStateCurnt points to the location where the new input data should be written */ |
emh203 | 0:3d9c67d97d6f | 446 | pStateCurnt = S->pState + (numTaps - 1u); |
emh203 | 0:3d9c67d97d6f | 447 | |
emh203 | 0:3d9c67d97d6f | 448 | /* Total number of output samples to be computed */ |
emh203 | 0:3d9c67d97d6f | 449 | blkCnt = outBlockSize; |
emh203 | 0:3d9c67d97d6f | 450 | |
emh203 | 0:3d9c67d97d6f | 451 | while(blkCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 452 | { |
emh203 | 0:3d9c67d97d6f | 453 | /* Copy decimation factor number of new input samples into the state buffer */ |
emh203 | 0:3d9c67d97d6f | 454 | i = S->M; |
emh203 | 0:3d9c67d97d6f | 455 | |
emh203 | 0:3d9c67d97d6f | 456 | do |
emh203 | 0:3d9c67d97d6f | 457 | { |
emh203 | 0:3d9c67d97d6f | 458 | *pStateCurnt++ = *pSrc++; |
emh203 | 0:3d9c67d97d6f | 459 | |
emh203 | 0:3d9c67d97d6f | 460 | } while(--i); |
emh203 | 0:3d9c67d97d6f | 461 | |
emh203 | 0:3d9c67d97d6f | 462 | /* Set accumulator to zero */ |
emh203 | 0:3d9c67d97d6f | 463 | sum0 = 0.0f; |
emh203 | 0:3d9c67d97d6f | 464 | |
emh203 | 0:3d9c67d97d6f | 465 | /* Initialize state pointer */ |
emh203 | 0:3d9c67d97d6f | 466 | px = pState; |
emh203 | 0:3d9c67d97d6f | 467 | |
emh203 | 0:3d9c67d97d6f | 468 | /* Initialize coeff pointer */ |
emh203 | 0:3d9c67d97d6f | 469 | pb = pCoeffs; |
emh203 | 0:3d9c67d97d6f | 470 | |
emh203 | 0:3d9c67d97d6f | 471 | tapCnt = numTaps; |
emh203 | 0:3d9c67d97d6f | 472 | |
emh203 | 0:3d9c67d97d6f | 473 | while(tapCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 474 | { |
emh203 | 0:3d9c67d97d6f | 475 | /* Read coefficients */ |
emh203 | 0:3d9c67d97d6f | 476 | c0 = *pb++; |
emh203 | 0:3d9c67d97d6f | 477 | |
emh203 | 0:3d9c67d97d6f | 478 | /* Fetch 1 state variable */ |
emh203 | 0:3d9c67d97d6f | 479 | x0 = *px++; |
emh203 | 0:3d9c67d97d6f | 480 | |
emh203 | 0:3d9c67d97d6f | 481 | /* Perform the multiply-accumulate */ |
emh203 | 0:3d9c67d97d6f | 482 | sum0 += x0 * c0; |
emh203 | 0:3d9c67d97d6f | 483 | |
emh203 | 0:3d9c67d97d6f | 484 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 485 | tapCnt--; |
emh203 | 0:3d9c67d97d6f | 486 | } |
emh203 | 0:3d9c67d97d6f | 487 | |
emh203 | 0:3d9c67d97d6f | 488 | /* Advance the state pointer by the decimation factor |
emh203 | 0:3d9c67d97d6f | 489 | * to process the next group of decimation factor number samples */ |
emh203 | 0:3d9c67d97d6f | 490 | pState = pState + S->M; |
emh203 | 0:3d9c67d97d6f | 491 | |
emh203 | 0:3d9c67d97d6f | 492 | /* The result is in the accumulator, store in the destination buffer. */ |
emh203 | 0:3d9c67d97d6f | 493 | *pDst++ = sum0; |
emh203 | 0:3d9c67d97d6f | 494 | |
emh203 | 0:3d9c67d97d6f | 495 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 496 | blkCnt--; |
emh203 | 0:3d9c67d97d6f | 497 | } |
emh203 | 0:3d9c67d97d6f | 498 | |
emh203 | 0:3d9c67d97d6f | 499 | /* Processing is complete. |
emh203 | 0:3d9c67d97d6f | 500 | ** Now copy the last numTaps - 1 samples to the start of the state buffer. |
emh203 | 0:3d9c67d97d6f | 501 | ** This prepares the state buffer for the next function call. */ |
emh203 | 0:3d9c67d97d6f | 502 | |
emh203 | 0:3d9c67d97d6f | 503 | /* Points to the start of the state buffer */ |
emh203 | 0:3d9c67d97d6f | 504 | pStateCurnt = S->pState; |
emh203 | 0:3d9c67d97d6f | 505 | |
emh203 | 0:3d9c67d97d6f | 506 | /* Copy numTaps number of values */ |
emh203 | 0:3d9c67d97d6f | 507 | i = (numTaps - 1u); |
emh203 | 0:3d9c67d97d6f | 508 | |
emh203 | 0:3d9c67d97d6f | 509 | /* copy data */ |
emh203 | 0:3d9c67d97d6f | 510 | while(i > 0u) |
emh203 | 0:3d9c67d97d6f | 511 | { |
emh203 | 0:3d9c67d97d6f | 512 | *pStateCurnt++ = *pState++; |
emh203 | 0:3d9c67d97d6f | 513 | |
emh203 | 0:3d9c67d97d6f | 514 | /* Decrement the loop counter */ |
emh203 | 0:3d9c67d97d6f | 515 | i--; |
emh203 | 0:3d9c67d97d6f | 516 | } |
emh203 | 0:3d9c67d97d6f | 517 | |
emh203 | 0:3d9c67d97d6f | 518 | #endif /* #ifndef ARM_MATH_CM0_FAMILY */ |
emh203 | 0:3d9c67d97d6f | 519 | |
emh203 | 0:3d9c67d97d6f | 520 | } |
emh203 | 0:3d9c67d97d6f | 521 | |
emh203 | 0:3d9c67d97d6f | 522 | /** |
emh203 | 0:3d9c67d97d6f | 523 | * @} end of FIR_decimate group |
emh203 | 0:3d9c67d97d6f | 524 | */ |