Eli Hughes / CMSIS_DSP_401

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_interpolate_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?

UserRevisionLine numberNew 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_interpolate_f32.c
emh203 0:3d9c67d97d6f 9 *
emh203 0:3d9c67d97d6f 10 * Description: FIR interpolation 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 * @defgroup FIR_Interpolate Finite Impulse Response (FIR) Interpolator
emh203 0:3d9c67d97d6f 45 *
emh203 0:3d9c67d97d6f 46 * These functions combine an upsampler (zero stuffer) and an FIR filter.
emh203 0:3d9c67d97d6f 47 * They are used in multirate systems for increasing the sample rate of a signal without introducing high frequency images.
emh203 0:3d9c67d97d6f 48 * Conceptually, the functions are equivalent to the block diagram below:
emh203 0:3d9c67d97d6f 49 * \image html FIRInterpolator.gif "Components included in the FIR Interpolator functions"
emh203 0:3d9c67d97d6f 50 * After upsampling by a factor of <code>L</code>, the signal should be filtered by a lowpass filter with a normalized
emh203 0:3d9c67d97d6f 51 * cutoff frequency of <code>1/L</code> in order to eliminate high frequency copies of the spectrum.
emh203 0:3d9c67d97d6f 52 * The user of the function is responsible for providing the filter coefficients.
emh203 0:3d9c67d97d6f 53 *
emh203 0:3d9c67d97d6f 54 * The FIR interpolator functions provided in the CMSIS DSP Library combine the upsampler and FIR filter in an efficient manner.
emh203 0:3d9c67d97d6f 55 * The upsampler inserts <code>L-1</code> zeros between each sample.
emh203 0:3d9c67d97d6f 56 * Instead of multiplying by these zero values, the FIR filter is designed to skip them.
emh203 0:3d9c67d97d6f 57 * This leads to an efficient implementation without any wasted effort.
emh203 0:3d9c67d97d6f 58 * The functions operate on blocks of input and output data.
emh203 0:3d9c67d97d6f 59 * <code>pSrc</code> points to an array of <code>blockSize</code> input values and
emh203 0:3d9c67d97d6f 60 * <code>pDst</code> points to an array of <code>blockSize*L</code> output values.
emh203 0:3d9c67d97d6f 61 *
emh203 0:3d9c67d97d6f 62 * The library provides separate functions for Q15, Q31, and floating-point data types.
emh203 0:3d9c67d97d6f 63 *
emh203 0:3d9c67d97d6f 64 * \par Algorithm:
emh203 0:3d9c67d97d6f 65 * The functions use a polyphase filter structure:
emh203 0:3d9c67d97d6f 66 * <pre>
emh203 0:3d9c67d97d6f 67 * y[n] = b[0] * x[n] + b[L] * x[n-1] + ... + b[L*(phaseLength-1)] * x[n-phaseLength+1]
emh203 0:3d9c67d97d6f 68 * y[n+1] = b[1] * x[n] + b[L+1] * x[n-1] + ... + b[L*(phaseLength-1)+1] * x[n-phaseLength+1]
emh203 0:3d9c67d97d6f 69 * ...
emh203 0:3d9c67d97d6f 70 * y[n+(L-1)] = b[L-1] * x[n] + b[2*L-1] * x[n-1] + ....+ b[L*(phaseLength-1)+(L-1)] * x[n-phaseLength+1]
emh203 0:3d9c67d97d6f 71 * </pre>
emh203 0:3d9c67d97d6f 72 * This approach is more efficient than straightforward upsample-then-filter algorithms.
emh203 0:3d9c67d97d6f 73 * With this method the computation is reduced by a factor of <code>1/L</code> when compared to using a standard FIR filter.
emh203 0:3d9c67d97d6f 74 * \par
emh203 0:3d9c67d97d6f 75 * <code>pCoeffs</code> points to a coefficient array of size <code>numTaps</code>.
emh203 0:3d9c67d97d6f 76 * <code>numTaps</code> must be a multiple of the interpolation factor <code>L</code> and this is checked by the
emh203 0:3d9c67d97d6f 77 * initialization functions.
emh203 0:3d9c67d97d6f 78 * Internally, the function divides the FIR filter's impulse response into shorter filters of length
emh203 0:3d9c67d97d6f 79 * <code>phaseLength=numTaps/L</code>.
emh203 0:3d9c67d97d6f 80 * Coefficients are stored in time reversed order.
emh203 0:3d9c67d97d6f 81 * \par
emh203 0:3d9c67d97d6f 82 * <pre>
emh203 0:3d9c67d97d6f 83 * {b[numTaps-1], b[numTaps-2], b[N-2], ..., b[1], b[0]}
emh203 0:3d9c67d97d6f 84 * </pre>
emh203 0:3d9c67d97d6f 85 * \par
emh203 0:3d9c67d97d6f 86 * <code>pState</code> points to a state array of size <code>blockSize + phaseLength - 1</code>.
emh203 0:3d9c67d97d6f 87 * Samples in the state buffer are stored in the order:
emh203 0:3d9c67d97d6f 88 * \par
emh203 0:3d9c67d97d6f 89 * <pre>
emh203 0:3d9c67d97d6f 90 * {x[n-phaseLength+1], x[n-phaseLength], x[n-phaseLength-1], x[n-phaseLength-2]....x[0], x[1], ..., x[blockSize-1]}
emh203 0:3d9c67d97d6f 91 * </pre>
emh203 0:3d9c67d97d6f 92 * The state variables are updated after each block of data is processed, the coefficients are untouched.
emh203 0:3d9c67d97d6f 93 *
emh203 0:3d9c67d97d6f 94 * \par Instance Structure
emh203 0:3d9c67d97d6f 95 * The coefficients and state variables for a filter are stored together in an instance data structure.
emh203 0:3d9c67d97d6f 96 * A separate instance structure must be defined for each filter.
emh203 0:3d9c67d97d6f 97 * Coefficient arrays may be shared among several instances while state variable array should be allocated separately.
emh203 0:3d9c67d97d6f 98 * There are separate instance structure declarations for each of the 3 supported data types.
emh203 0:3d9c67d97d6f 99 *
emh203 0:3d9c67d97d6f 100 * \par Initialization Functions
emh203 0:3d9c67d97d6f 101 * There is also an associated initialization function for each data type.
emh203 0:3d9c67d97d6f 102 * The initialization function performs the following operations:
emh203 0:3d9c67d97d6f 103 * - Sets the values of the internal structure fields.
emh203 0:3d9c67d97d6f 104 * - Zeros out the values in the state buffer.
emh203 0:3d9c67d97d6f 105 * - Checks to make sure that the length of the filter is a multiple of the interpolation factor.
emh203 0:3d9c67d97d6f 106 * To do this manually without calling the init function, assign the follow subfields of the instance structure:
emh203 0:3d9c67d97d6f 107 * L (interpolation factor), pCoeffs, phaseLength (numTaps / L), pState. Also set all of the values in pState to zero.
emh203 0:3d9c67d97d6f 108 *
emh203 0:3d9c67d97d6f 109 * \par
emh203 0:3d9c67d97d6f 110 * Use of the initialization function is optional.
emh203 0:3d9c67d97d6f 111 * However, if the initialization function is used, then the instance structure cannot be placed into a const data section.
emh203 0:3d9c67d97d6f 112 * To place an instance structure into a const data section, the instance structure must be manually initialized.
emh203 0:3d9c67d97d6f 113 * The code below statically initializes each of the 3 different data type filter instance structures
emh203 0:3d9c67d97d6f 114 * <pre>
emh203 0:3d9c67d97d6f 115 * arm_fir_interpolate_instance_f32 S = {L, phaseLength, pCoeffs, pState};
emh203 0:3d9c67d97d6f 116 * arm_fir_interpolate_instance_q31 S = {L, phaseLength, pCoeffs, pState};
emh203 0:3d9c67d97d6f 117 * arm_fir_interpolate_instance_q15 S = {L, phaseLength, pCoeffs, pState};
emh203 0:3d9c67d97d6f 118 * </pre>
emh203 0:3d9c67d97d6f 119 * where <code>L</code> is the interpolation factor; <code>phaseLength=numTaps/L</code> is the
emh203 0:3d9c67d97d6f 120 * length of each of the shorter FIR filters used internally,
emh203 0:3d9c67d97d6f 121 * <code>pCoeffs</code> is the address of the coefficient buffer;
emh203 0:3d9c67d97d6f 122 * <code>pState</code> is the address of the state buffer.
emh203 0:3d9c67d97d6f 123 * Be sure to set the values in the state buffer to zeros when doing static initialization.
emh203 0:3d9c67d97d6f 124 *
emh203 0:3d9c67d97d6f 125 * \par Fixed-Point Behavior
emh203 0:3d9c67d97d6f 126 * Care must be taken when using the fixed-point versions of the FIR interpolate filter functions.
emh203 0:3d9c67d97d6f 127 * In particular, the overflow and saturation behavior of the accumulator used in each function must be considered.
emh203 0:3d9c67d97d6f 128 * Refer to the function specific documentation below for usage guidelines.
emh203 0:3d9c67d97d6f 129 */
emh203 0:3d9c67d97d6f 130
emh203 0:3d9c67d97d6f 131 /**
emh203 0:3d9c67d97d6f 132 * @addtogroup FIR_Interpolate
emh203 0:3d9c67d97d6f 133 * @{
emh203 0:3d9c67d97d6f 134 */
emh203 0:3d9c67d97d6f 135
emh203 0:3d9c67d97d6f 136 /**
emh203 0:3d9c67d97d6f 137 * @brief Processing function for the floating-point FIR interpolator.
emh203 0:3d9c67d97d6f 138 * @param[in] *S points to an instance of the floating-point FIR interpolator structure.
emh203 0:3d9c67d97d6f 139 * @param[in] *pSrc points to the block of input data.
emh203 0:3d9c67d97d6f 140 * @param[out] *pDst points to the block of output data.
emh203 0:3d9c67d97d6f 141 * @param[in] blockSize number of input samples to process per call.
emh203 0:3d9c67d97d6f 142 * @return none.
emh203 0:3d9c67d97d6f 143 */
emh203 0:3d9c67d97d6f 144 #ifndef ARM_MATH_CM0_FAMILY
emh203 0:3d9c67d97d6f 145
emh203 0:3d9c67d97d6f 146 /* Run the below code for Cortex-M4 and Cortex-M3 */
emh203 0:3d9c67d97d6f 147
emh203 0:3d9c67d97d6f 148 void arm_fir_interpolate_f32(
emh203 0:3d9c67d97d6f 149 const arm_fir_interpolate_instance_f32 * S,
emh203 0:3d9c67d97d6f 150 float32_t * pSrc,
emh203 0:3d9c67d97d6f 151 float32_t * pDst,
emh203 0:3d9c67d97d6f 152 uint32_t blockSize)
emh203 0:3d9c67d97d6f 153 {
emh203 0:3d9c67d97d6f 154 float32_t *pState = S->pState; /* State pointer */
emh203 0:3d9c67d97d6f 155 float32_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emh203 0:3d9c67d97d6f 156 float32_t *pStateCurnt; /* Points to the current sample of the state */
emh203 0:3d9c67d97d6f 157 float32_t *ptr1, *ptr2; /* Temporary pointers for state and coefficient buffers */
emh203 0:3d9c67d97d6f 158 float32_t sum0; /* Accumulators */
emh203 0:3d9c67d97d6f 159 float32_t x0, c0; /* Temporary variables to hold state and coefficient values */
emh203 0:3d9c67d97d6f 160 uint32_t i, blkCnt, j; /* Loop counters */
emh203 0:3d9c67d97d6f 161 uint16_t phaseLen = S->phaseLength, tapCnt; /* Length of each polyphase filter component */
emh203 0:3d9c67d97d6f 162 float32_t acc0, acc1, acc2, acc3;
emh203 0:3d9c67d97d6f 163 float32_t x1, x2, x3;
emh203 0:3d9c67d97d6f 164 uint32_t blkCntN4;
emh203 0:3d9c67d97d6f 165 float32_t c1, c2, c3;
emh203 0:3d9c67d97d6f 166
emh203 0:3d9c67d97d6f 167 /* S->pState buffer contains previous frame (phaseLen - 1) samples */
emh203 0:3d9c67d97d6f 168 /* pStateCurnt points to the location where the new input data should be written */
emh203 0:3d9c67d97d6f 169 pStateCurnt = S->pState + (phaseLen - 1u);
emh203 0:3d9c67d97d6f 170
emh203 0:3d9c67d97d6f 171 /* Initialise blkCnt */
emh203 0:3d9c67d97d6f 172 blkCnt = blockSize / 4;
emh203 0:3d9c67d97d6f 173 blkCntN4 = blockSize - (4 * blkCnt);
emh203 0:3d9c67d97d6f 174
emh203 0:3d9c67d97d6f 175 /* Samples loop unrolled by 4 */
emh203 0:3d9c67d97d6f 176 while(blkCnt > 0u)
emh203 0:3d9c67d97d6f 177 {
emh203 0:3d9c67d97d6f 178 /* Copy new input sample into the state buffer */
emh203 0:3d9c67d97d6f 179 *pStateCurnt++ = *pSrc++;
emh203 0:3d9c67d97d6f 180 *pStateCurnt++ = *pSrc++;
emh203 0:3d9c67d97d6f 181 *pStateCurnt++ = *pSrc++;
emh203 0:3d9c67d97d6f 182 *pStateCurnt++ = *pSrc++;
emh203 0:3d9c67d97d6f 183
emh203 0:3d9c67d97d6f 184 /* Address modifier index of coefficient buffer */
emh203 0:3d9c67d97d6f 185 j = 1u;
emh203 0:3d9c67d97d6f 186
emh203 0:3d9c67d97d6f 187 /* Loop over the Interpolation factor. */
emh203 0:3d9c67d97d6f 188 i = (S->L);
emh203 0:3d9c67d97d6f 189
emh203 0:3d9c67d97d6f 190 while(i > 0u)
emh203 0:3d9c67d97d6f 191 {
emh203 0:3d9c67d97d6f 192 /* Set accumulator to zero */
emh203 0:3d9c67d97d6f 193 acc0 = 0.0f;
emh203 0:3d9c67d97d6f 194 acc1 = 0.0f;
emh203 0:3d9c67d97d6f 195 acc2 = 0.0f;
emh203 0:3d9c67d97d6f 196 acc3 = 0.0f;
emh203 0:3d9c67d97d6f 197
emh203 0:3d9c67d97d6f 198 /* Initialize state pointer */
emh203 0:3d9c67d97d6f 199 ptr1 = pState;
emh203 0:3d9c67d97d6f 200
emh203 0:3d9c67d97d6f 201 /* Initialize coefficient pointer */
emh203 0:3d9c67d97d6f 202 ptr2 = pCoeffs + (S->L - j);
emh203 0:3d9c67d97d6f 203
emh203 0:3d9c67d97d6f 204 /* Loop over the polyPhase length. Unroll by a factor of 4.
emh203 0:3d9c67d97d6f 205 ** Repeat until we've computed numTaps-(4*S->L) coefficients. */
emh203 0:3d9c67d97d6f 206 tapCnt = phaseLen >> 2u;
emh203 0:3d9c67d97d6f 207
emh203 0:3d9c67d97d6f 208 x0 = *(ptr1++);
emh203 0:3d9c67d97d6f 209 x1 = *(ptr1++);
emh203 0:3d9c67d97d6f 210 x2 = *(ptr1++);
emh203 0:3d9c67d97d6f 211
emh203 0:3d9c67d97d6f 212 while(tapCnt > 0u)
emh203 0:3d9c67d97d6f 213 {
emh203 0:3d9c67d97d6f 214
emh203 0:3d9c67d97d6f 215 /* Read the input sample */
emh203 0:3d9c67d97d6f 216 x3 = *(ptr1++);
emh203 0:3d9c67d97d6f 217
emh203 0:3d9c67d97d6f 218 /* Read the coefficient */
emh203 0:3d9c67d97d6f 219 c0 = *(ptr2);
emh203 0:3d9c67d97d6f 220
emh203 0:3d9c67d97d6f 221 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 222 acc0 += x0 * c0;
emh203 0:3d9c67d97d6f 223 acc1 += x1 * c0;
emh203 0:3d9c67d97d6f 224 acc2 += x2 * c0;
emh203 0:3d9c67d97d6f 225 acc3 += x3 * c0;
emh203 0:3d9c67d97d6f 226
emh203 0:3d9c67d97d6f 227 /* Read the coefficient */
emh203 0:3d9c67d97d6f 228 c1 = *(ptr2 + S->L);
emh203 0:3d9c67d97d6f 229
emh203 0:3d9c67d97d6f 230 /* Read the input sample */
emh203 0:3d9c67d97d6f 231 x0 = *(ptr1++);
emh203 0:3d9c67d97d6f 232
emh203 0:3d9c67d97d6f 233 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 234 acc0 += x1 * c1;
emh203 0:3d9c67d97d6f 235 acc1 += x2 * c1;
emh203 0:3d9c67d97d6f 236 acc2 += x3 * c1;
emh203 0:3d9c67d97d6f 237 acc3 += x0 * c1;
emh203 0:3d9c67d97d6f 238
emh203 0:3d9c67d97d6f 239 /* Read the coefficient */
emh203 0:3d9c67d97d6f 240 c2 = *(ptr2 + S->L * 2);
emh203 0:3d9c67d97d6f 241
emh203 0:3d9c67d97d6f 242 /* Read the input sample */
emh203 0:3d9c67d97d6f 243 x1 = *(ptr1++);
emh203 0:3d9c67d97d6f 244
emh203 0:3d9c67d97d6f 245 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 246 acc0 += x2 * c2;
emh203 0:3d9c67d97d6f 247 acc1 += x3 * c2;
emh203 0:3d9c67d97d6f 248 acc2 += x0 * c2;
emh203 0:3d9c67d97d6f 249 acc3 += x1 * c2;
emh203 0:3d9c67d97d6f 250
emh203 0:3d9c67d97d6f 251 /* Read the coefficient */
emh203 0:3d9c67d97d6f 252 c3 = *(ptr2 + S->L * 3);
emh203 0:3d9c67d97d6f 253
emh203 0:3d9c67d97d6f 254 /* Read the input sample */
emh203 0:3d9c67d97d6f 255 x2 = *(ptr1++);
emh203 0:3d9c67d97d6f 256
emh203 0:3d9c67d97d6f 257 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 258 acc0 += x3 * c3;
emh203 0:3d9c67d97d6f 259 acc1 += x0 * c3;
emh203 0:3d9c67d97d6f 260 acc2 += x1 * c3;
emh203 0:3d9c67d97d6f 261 acc3 += x2 * c3;
emh203 0:3d9c67d97d6f 262
emh203 0:3d9c67d97d6f 263
emh203 0:3d9c67d97d6f 264 /* Upsampling is done by stuffing L-1 zeros between each sample.
emh203 0:3d9c67d97d6f 265 * So instead of multiplying zeros with coefficients,
emh203 0:3d9c67d97d6f 266 * Increment the coefficient pointer by interpolation factor times. */
emh203 0:3d9c67d97d6f 267 ptr2 += 4 * S->L;
emh203 0:3d9c67d97d6f 268
emh203 0:3d9c67d97d6f 269 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 270 tapCnt--;
emh203 0:3d9c67d97d6f 271 }
emh203 0:3d9c67d97d6f 272
emh203 0:3d9c67d97d6f 273 /* If the polyPhase length is not a multiple of 4, compute the remaining filter taps */
emh203 0:3d9c67d97d6f 274 tapCnt = phaseLen % 0x4u;
emh203 0:3d9c67d97d6f 275
emh203 0:3d9c67d97d6f 276 while(tapCnt > 0u)
emh203 0:3d9c67d97d6f 277 {
emh203 0:3d9c67d97d6f 278
emh203 0:3d9c67d97d6f 279 /* Read the input sample */
emh203 0:3d9c67d97d6f 280 x3 = *(ptr1++);
emh203 0:3d9c67d97d6f 281
emh203 0:3d9c67d97d6f 282 /* Read the coefficient */
emh203 0:3d9c67d97d6f 283 c0 = *(ptr2);
emh203 0:3d9c67d97d6f 284
emh203 0:3d9c67d97d6f 285 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 286 acc0 += x0 * c0;
emh203 0:3d9c67d97d6f 287 acc1 += x1 * c0;
emh203 0:3d9c67d97d6f 288 acc2 += x2 * c0;
emh203 0:3d9c67d97d6f 289 acc3 += x3 * c0;
emh203 0:3d9c67d97d6f 290
emh203 0:3d9c67d97d6f 291 /* Increment the coefficient pointer by interpolation factor times. */
emh203 0:3d9c67d97d6f 292 ptr2 += S->L;
emh203 0:3d9c67d97d6f 293
emh203 0:3d9c67d97d6f 294 /* update states for next sample processing */
emh203 0:3d9c67d97d6f 295 x0 = x1;
emh203 0:3d9c67d97d6f 296 x1 = x2;
emh203 0:3d9c67d97d6f 297 x2 = x3;
emh203 0:3d9c67d97d6f 298
emh203 0:3d9c67d97d6f 299 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 300 tapCnt--;
emh203 0:3d9c67d97d6f 301 }
emh203 0:3d9c67d97d6f 302
emh203 0:3d9c67d97d6f 303 /* The result is in the accumulator, store in the destination buffer. */
emh203 0:3d9c67d97d6f 304 *pDst = acc0;
emh203 0:3d9c67d97d6f 305 *(pDst + S->L) = acc1;
emh203 0:3d9c67d97d6f 306 *(pDst + 2 * S->L) = acc2;
emh203 0:3d9c67d97d6f 307 *(pDst + 3 * S->L) = acc3;
emh203 0:3d9c67d97d6f 308
emh203 0:3d9c67d97d6f 309 pDst++;
emh203 0:3d9c67d97d6f 310
emh203 0:3d9c67d97d6f 311 /* Increment the address modifier index of coefficient buffer */
emh203 0:3d9c67d97d6f 312 j++;
emh203 0:3d9c67d97d6f 313
emh203 0:3d9c67d97d6f 314 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 315 i--;
emh203 0:3d9c67d97d6f 316 }
emh203 0:3d9c67d97d6f 317
emh203 0:3d9c67d97d6f 318 /* Advance the state pointer by 1
emh203 0:3d9c67d97d6f 319 * to process the next group of interpolation factor number samples */
emh203 0:3d9c67d97d6f 320 pState = pState + 4;
emh203 0:3d9c67d97d6f 321
emh203 0:3d9c67d97d6f 322 pDst += S->L * 3;
emh203 0:3d9c67d97d6f 323
emh203 0:3d9c67d97d6f 324 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 325 blkCnt--;
emh203 0:3d9c67d97d6f 326 }
emh203 0:3d9c67d97d6f 327
emh203 0:3d9c67d97d6f 328 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
emh203 0:3d9c67d97d6f 329 ** No loop unrolling is used. */
emh203 0:3d9c67d97d6f 330
emh203 0:3d9c67d97d6f 331 while(blkCntN4 > 0u)
emh203 0:3d9c67d97d6f 332 {
emh203 0:3d9c67d97d6f 333 /* Copy new input sample into the state buffer */
emh203 0:3d9c67d97d6f 334 *pStateCurnt++ = *pSrc++;
emh203 0:3d9c67d97d6f 335
emh203 0:3d9c67d97d6f 336 /* Address modifier index of coefficient buffer */
emh203 0:3d9c67d97d6f 337 j = 1u;
emh203 0:3d9c67d97d6f 338
emh203 0:3d9c67d97d6f 339 /* Loop over the Interpolation factor. */
emh203 0:3d9c67d97d6f 340 i = S->L;
emh203 0:3d9c67d97d6f 341 while(i > 0u)
emh203 0:3d9c67d97d6f 342 {
emh203 0:3d9c67d97d6f 343 /* Set accumulator to zero */
emh203 0:3d9c67d97d6f 344 sum0 = 0.0f;
emh203 0:3d9c67d97d6f 345
emh203 0:3d9c67d97d6f 346 /* Initialize state pointer */
emh203 0:3d9c67d97d6f 347 ptr1 = pState;
emh203 0:3d9c67d97d6f 348
emh203 0:3d9c67d97d6f 349 /* Initialize coefficient pointer */
emh203 0:3d9c67d97d6f 350 ptr2 = pCoeffs + (S->L - j);
emh203 0:3d9c67d97d6f 351
emh203 0:3d9c67d97d6f 352 /* Loop over the polyPhase length. Unroll by a factor of 4.
emh203 0:3d9c67d97d6f 353 ** Repeat until we've computed numTaps-(4*S->L) coefficients. */
emh203 0:3d9c67d97d6f 354 tapCnt = phaseLen >> 2u;
emh203 0:3d9c67d97d6f 355 while(tapCnt > 0u)
emh203 0:3d9c67d97d6f 356 {
emh203 0:3d9c67d97d6f 357
emh203 0:3d9c67d97d6f 358 /* Read the coefficient */
emh203 0:3d9c67d97d6f 359 c0 = *(ptr2);
emh203 0:3d9c67d97d6f 360
emh203 0:3d9c67d97d6f 361 /* Upsampling is done by stuffing L-1 zeros between each sample.
emh203 0:3d9c67d97d6f 362 * So instead of multiplying zeros with coefficients,
emh203 0:3d9c67d97d6f 363 * Increment the coefficient pointer by interpolation factor times. */
emh203 0:3d9c67d97d6f 364 ptr2 += S->L;
emh203 0:3d9c67d97d6f 365
emh203 0:3d9c67d97d6f 366 /* Read the input sample */
emh203 0:3d9c67d97d6f 367 x0 = *(ptr1++);
emh203 0:3d9c67d97d6f 368
emh203 0:3d9c67d97d6f 369 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 370 sum0 += x0 * c0;
emh203 0:3d9c67d97d6f 371
emh203 0:3d9c67d97d6f 372 /* Read the coefficient */
emh203 0:3d9c67d97d6f 373 c0 = *(ptr2);
emh203 0:3d9c67d97d6f 374
emh203 0:3d9c67d97d6f 375 /* Increment the coefficient pointer by interpolation factor times. */
emh203 0:3d9c67d97d6f 376 ptr2 += S->L;
emh203 0:3d9c67d97d6f 377
emh203 0:3d9c67d97d6f 378 /* Read the input sample */
emh203 0:3d9c67d97d6f 379 x0 = *(ptr1++);
emh203 0:3d9c67d97d6f 380
emh203 0:3d9c67d97d6f 381 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 382 sum0 += x0 * c0;
emh203 0:3d9c67d97d6f 383
emh203 0:3d9c67d97d6f 384 /* Read the coefficient */
emh203 0:3d9c67d97d6f 385 c0 = *(ptr2);
emh203 0:3d9c67d97d6f 386
emh203 0:3d9c67d97d6f 387 /* Increment the coefficient pointer by interpolation factor times. */
emh203 0:3d9c67d97d6f 388 ptr2 += S->L;
emh203 0:3d9c67d97d6f 389
emh203 0:3d9c67d97d6f 390 /* Read the input sample */
emh203 0:3d9c67d97d6f 391 x0 = *(ptr1++);
emh203 0:3d9c67d97d6f 392
emh203 0:3d9c67d97d6f 393 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 394 sum0 += x0 * c0;
emh203 0:3d9c67d97d6f 395
emh203 0:3d9c67d97d6f 396 /* Read the coefficient */
emh203 0:3d9c67d97d6f 397 c0 = *(ptr2);
emh203 0:3d9c67d97d6f 398
emh203 0:3d9c67d97d6f 399 /* Increment the coefficient pointer by interpolation factor times. */
emh203 0:3d9c67d97d6f 400 ptr2 += S->L;
emh203 0:3d9c67d97d6f 401
emh203 0:3d9c67d97d6f 402 /* Read the input sample */
emh203 0:3d9c67d97d6f 403 x0 = *(ptr1++);
emh203 0:3d9c67d97d6f 404
emh203 0:3d9c67d97d6f 405 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 406 sum0 += x0 * c0;
emh203 0:3d9c67d97d6f 407
emh203 0:3d9c67d97d6f 408 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 409 tapCnt--;
emh203 0:3d9c67d97d6f 410 }
emh203 0:3d9c67d97d6f 411
emh203 0:3d9c67d97d6f 412 /* If the polyPhase length is not a multiple of 4, compute the remaining filter taps */
emh203 0:3d9c67d97d6f 413 tapCnt = phaseLen % 0x4u;
emh203 0:3d9c67d97d6f 414
emh203 0:3d9c67d97d6f 415 while(tapCnt > 0u)
emh203 0:3d9c67d97d6f 416 {
emh203 0:3d9c67d97d6f 417 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 418 sum0 += *(ptr1++) * (*ptr2);
emh203 0:3d9c67d97d6f 419
emh203 0:3d9c67d97d6f 420 /* Increment the coefficient pointer by interpolation factor times. */
emh203 0:3d9c67d97d6f 421 ptr2 += S->L;
emh203 0:3d9c67d97d6f 422
emh203 0:3d9c67d97d6f 423 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 424 tapCnt--;
emh203 0:3d9c67d97d6f 425 }
emh203 0:3d9c67d97d6f 426
emh203 0:3d9c67d97d6f 427 /* The result is in the accumulator, store in the destination buffer. */
emh203 0:3d9c67d97d6f 428 *pDst++ = sum0;
emh203 0:3d9c67d97d6f 429
emh203 0:3d9c67d97d6f 430 /* Increment the address modifier index of coefficient buffer */
emh203 0:3d9c67d97d6f 431 j++;
emh203 0:3d9c67d97d6f 432
emh203 0:3d9c67d97d6f 433 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 434 i--;
emh203 0:3d9c67d97d6f 435 }
emh203 0:3d9c67d97d6f 436
emh203 0:3d9c67d97d6f 437 /* Advance the state pointer by 1
emh203 0:3d9c67d97d6f 438 * to process the next group of interpolation factor number samples */
emh203 0:3d9c67d97d6f 439 pState = pState + 1;
emh203 0:3d9c67d97d6f 440
emh203 0:3d9c67d97d6f 441 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 442 blkCntN4--;
emh203 0:3d9c67d97d6f 443 }
emh203 0:3d9c67d97d6f 444
emh203 0:3d9c67d97d6f 445 /* Processing is complete.
emh203 0:3d9c67d97d6f 446 ** Now copy the last phaseLen - 1 samples to the satrt of the state buffer.
emh203 0:3d9c67d97d6f 447 ** This prepares the state buffer for the next function call. */
emh203 0:3d9c67d97d6f 448
emh203 0:3d9c67d97d6f 449 /* Points to the start of the state buffer */
emh203 0:3d9c67d97d6f 450 pStateCurnt = S->pState;
emh203 0:3d9c67d97d6f 451
emh203 0:3d9c67d97d6f 452 tapCnt = (phaseLen - 1u) >> 2u;
emh203 0:3d9c67d97d6f 453
emh203 0:3d9c67d97d6f 454 /* copy data */
emh203 0:3d9c67d97d6f 455 while(tapCnt > 0u)
emh203 0:3d9c67d97d6f 456 {
emh203 0:3d9c67d97d6f 457 *pStateCurnt++ = *pState++;
emh203 0:3d9c67d97d6f 458 *pStateCurnt++ = *pState++;
emh203 0:3d9c67d97d6f 459 *pStateCurnt++ = *pState++;
emh203 0:3d9c67d97d6f 460 *pStateCurnt++ = *pState++;
emh203 0:3d9c67d97d6f 461
emh203 0:3d9c67d97d6f 462 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 463 tapCnt--;
emh203 0:3d9c67d97d6f 464 }
emh203 0:3d9c67d97d6f 465
emh203 0:3d9c67d97d6f 466 tapCnt = (phaseLen - 1u) % 0x04u;
emh203 0:3d9c67d97d6f 467
emh203 0:3d9c67d97d6f 468 /* copy data */
emh203 0:3d9c67d97d6f 469 while(tapCnt > 0u)
emh203 0:3d9c67d97d6f 470 {
emh203 0:3d9c67d97d6f 471 *pStateCurnt++ = *pState++;
emh203 0:3d9c67d97d6f 472
emh203 0:3d9c67d97d6f 473 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 474 tapCnt--;
emh203 0:3d9c67d97d6f 475 }
emh203 0:3d9c67d97d6f 476 }
emh203 0:3d9c67d97d6f 477
emh203 0:3d9c67d97d6f 478 #else
emh203 0:3d9c67d97d6f 479
emh203 0:3d9c67d97d6f 480 /* Run the below code for Cortex-M0 */
emh203 0:3d9c67d97d6f 481
emh203 0:3d9c67d97d6f 482 void arm_fir_interpolate_f32(
emh203 0:3d9c67d97d6f 483 const arm_fir_interpolate_instance_f32 * S,
emh203 0:3d9c67d97d6f 484 float32_t * pSrc,
emh203 0:3d9c67d97d6f 485 float32_t * pDst,
emh203 0:3d9c67d97d6f 486 uint32_t blockSize)
emh203 0:3d9c67d97d6f 487 {
emh203 0:3d9c67d97d6f 488 float32_t *pState = S->pState; /* State pointer */
emh203 0:3d9c67d97d6f 489 float32_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emh203 0:3d9c67d97d6f 490 float32_t *pStateCurnt; /* Points to the current sample of the state */
emh203 0:3d9c67d97d6f 491 float32_t *ptr1, *ptr2; /* Temporary pointers for state and coefficient buffers */
emh203 0:3d9c67d97d6f 492
emh203 0:3d9c67d97d6f 493
emh203 0:3d9c67d97d6f 494 float32_t sum; /* Accumulator */
emh203 0:3d9c67d97d6f 495 uint32_t i, blkCnt; /* Loop counters */
emh203 0:3d9c67d97d6f 496 uint16_t phaseLen = S->phaseLength, tapCnt; /* Length of each polyphase filter component */
emh203 0:3d9c67d97d6f 497
emh203 0:3d9c67d97d6f 498
emh203 0:3d9c67d97d6f 499 /* S->pState buffer contains previous frame (phaseLen - 1) samples */
emh203 0:3d9c67d97d6f 500 /* pStateCurnt points to the location where the new input data should be written */
emh203 0:3d9c67d97d6f 501 pStateCurnt = S->pState + (phaseLen - 1u);
emh203 0:3d9c67d97d6f 502
emh203 0:3d9c67d97d6f 503 /* Total number of intput samples */
emh203 0:3d9c67d97d6f 504 blkCnt = blockSize;
emh203 0:3d9c67d97d6f 505
emh203 0:3d9c67d97d6f 506 /* Loop over the blockSize. */
emh203 0:3d9c67d97d6f 507 while(blkCnt > 0u)
emh203 0:3d9c67d97d6f 508 {
emh203 0:3d9c67d97d6f 509 /* Copy new input sample into the state buffer */
emh203 0:3d9c67d97d6f 510 *pStateCurnt++ = *pSrc++;
emh203 0:3d9c67d97d6f 511
emh203 0:3d9c67d97d6f 512 /* Loop over the Interpolation factor. */
emh203 0:3d9c67d97d6f 513 i = S->L;
emh203 0:3d9c67d97d6f 514
emh203 0:3d9c67d97d6f 515 while(i > 0u)
emh203 0:3d9c67d97d6f 516 {
emh203 0:3d9c67d97d6f 517 /* Set accumulator to zero */
emh203 0:3d9c67d97d6f 518 sum = 0.0f;
emh203 0:3d9c67d97d6f 519
emh203 0:3d9c67d97d6f 520 /* Initialize state pointer */
emh203 0:3d9c67d97d6f 521 ptr1 = pState;
emh203 0:3d9c67d97d6f 522
emh203 0:3d9c67d97d6f 523 /* Initialize coefficient pointer */
emh203 0:3d9c67d97d6f 524 ptr2 = pCoeffs + (i - 1u);
emh203 0:3d9c67d97d6f 525
emh203 0:3d9c67d97d6f 526 /* Loop over the polyPhase length */
emh203 0:3d9c67d97d6f 527 tapCnt = phaseLen;
emh203 0:3d9c67d97d6f 528
emh203 0:3d9c67d97d6f 529 while(tapCnt > 0u)
emh203 0:3d9c67d97d6f 530 {
emh203 0:3d9c67d97d6f 531 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 532 sum += *ptr1++ * *ptr2;
emh203 0:3d9c67d97d6f 533
emh203 0:3d9c67d97d6f 534 /* Increment the coefficient pointer by interpolation factor times. */
emh203 0:3d9c67d97d6f 535 ptr2 += S->L;
emh203 0:3d9c67d97d6f 536
emh203 0:3d9c67d97d6f 537 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 538 tapCnt--;
emh203 0:3d9c67d97d6f 539 }
emh203 0:3d9c67d97d6f 540
emh203 0:3d9c67d97d6f 541 /* The result is in the accumulator, store in the destination buffer. */
emh203 0:3d9c67d97d6f 542 *pDst++ = sum;
emh203 0:3d9c67d97d6f 543
emh203 0:3d9c67d97d6f 544 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 545 i--;
emh203 0:3d9c67d97d6f 546 }
emh203 0:3d9c67d97d6f 547
emh203 0:3d9c67d97d6f 548 /* Advance the state pointer by 1
emh203 0:3d9c67d97d6f 549 * to process the next group of interpolation factor number samples */
emh203 0:3d9c67d97d6f 550 pState = pState + 1;
emh203 0:3d9c67d97d6f 551
emh203 0:3d9c67d97d6f 552 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 553 blkCnt--;
emh203 0:3d9c67d97d6f 554 }
emh203 0:3d9c67d97d6f 555
emh203 0:3d9c67d97d6f 556 /* Processing is complete.
emh203 0:3d9c67d97d6f 557 ** Now copy the last phaseLen - 1 samples to the start of the state buffer.
emh203 0:3d9c67d97d6f 558 ** This prepares the state buffer for the next function call. */
emh203 0:3d9c67d97d6f 559
emh203 0:3d9c67d97d6f 560 /* Points to the start of the state buffer */
emh203 0:3d9c67d97d6f 561 pStateCurnt = S->pState;
emh203 0:3d9c67d97d6f 562
emh203 0:3d9c67d97d6f 563 tapCnt = phaseLen - 1u;
emh203 0:3d9c67d97d6f 564
emh203 0:3d9c67d97d6f 565 while(tapCnt > 0u)
emh203 0:3d9c67d97d6f 566 {
emh203 0:3d9c67d97d6f 567 *pStateCurnt++ = *pState++;
emh203 0:3d9c67d97d6f 568
emh203 0:3d9c67d97d6f 569 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 570 tapCnt--;
emh203 0:3d9c67d97d6f 571 }
emh203 0:3d9c67d97d6f 572
emh203 0:3d9c67d97d6f 573 }
emh203 0:3d9c67d97d6f 574
emh203 0:3d9c67d97d6f 575 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
emh203 0:3d9c67d97d6f 576
emh203 0:3d9c67d97d6f 577
emh203 0:3d9c67d97d6f 578
emh203 0:3d9c67d97d6f 579 /**
emh203 0:3d9c67d97d6f 580 * @} end of FIR_Interpolate group
emh203 0:3d9c67d97d6f 581 */