mbed official / mbed-dsp

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_lattice_f32.c@5:3762170b6d4d, 2015-11-20 (annotated)

Committer:
mbed_official
Date:
Fri Nov 20 08:45:18 2015 +0000
Revision:
5:3762170b6d4d
Parent:
3:7a284390b0ce 
Synchronized with git revision 2eb940b9a73af188d3004a2575fdfbb05febe62b



Full URL: https://github.com/mbedmicro/mbed/commit/2eb940b9a73af188d3004a2575fdfbb05febe62b/



Added option to build rpc library. closes #1426

Who changed what in which revision?

UserRevisionLine numberNew contents of line
emilmont 1:fdd22bb7aa52 1 /* ----------------------------------------------------------------------
mbed_official 5:3762170b6d4d 2 * Copyright (C) 2010-2014 ARM Limited. All rights reserved.
emilmont 1:fdd22bb7aa52 3 *
mbed_official 5:3762170b6d4d 4 * $Date: 19. March 2015
mbed_official 5:3762170b6d4d 5 * $Revision: V.1.4.5
emilmont 1:fdd22bb7aa52 6 *
emilmont 2:da51fb522205 7 * Project: CMSIS DSP Library
emilmont 2:da51fb522205 8 * Title: arm_fir_lattice_f32.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 2:da51fb522205 10 * Description: Processing function for the floating-point FIR Lattice filter.
emilmont 1:fdd22bb7aa52 11 *
emilmont 1:fdd22bb7aa52 12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
emilmont 1:fdd22bb7aa52 13 *
mbed_official 3:7a284390b0ce 14 * Redistribution and use in source and binary forms, with or without
mbed_official 3:7a284390b0ce 15 * modification, are permitted provided that the following conditions
mbed_official 3:7a284390b0ce 16 * are met:
mbed_official 3:7a284390b0ce 17 * - Redistributions of source code must retain the above copyright
mbed_official 3:7a284390b0ce 18 * notice, this list of conditions and the following disclaimer.
mbed_official 3:7a284390b0ce 19 * - Redistributions in binary form must reproduce the above copyright
mbed_official 3:7a284390b0ce 20 * notice, this list of conditions and the following disclaimer in
mbed_official 3:7a284390b0ce 21 * the documentation and/or other materials provided with the
mbed_official 3:7a284390b0ce 22 * distribution.
mbed_official 3:7a284390b0ce 23 * - Neither the name of ARM LIMITED nor the names of its contributors
mbed_official 3:7a284390b0ce 24 * may be used to endorse or promote products derived from this
mbed_official 3:7a284390b0ce 25 * software without specific prior written permission.
mbed_official 3:7a284390b0ce 26 *
mbed_official 3:7a284390b0ce 27 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
mbed_official 3:7a284390b0ce 28 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
mbed_official 3:7a284390b0ce 29 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
mbed_official 3:7a284390b0ce 30 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
mbed_official 3:7a284390b0ce 31 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
mbed_official 3:7a284390b0ce 32 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
mbed_official 3:7a284390b0ce 33 * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
mbed_official 3:7a284390b0ce 34 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
mbed_official 3:7a284390b0ce 35 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
mbed_official 3:7a284390b0ce 36 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
mbed_official 3:7a284390b0ce 37 * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
mbed_official 3:7a284390b0ce 38 * POSSIBILITY OF SUCH DAMAGE.
emilmont 1:fdd22bb7aa52 39 * -------------------------------------------------------------------- */
emilmont 1:fdd22bb7aa52 40
emilmont 1:fdd22bb7aa52 41 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 42
emilmont 1:fdd22bb7aa52 43 /**
emilmont 1:fdd22bb7aa52 44 * @ingroup groupFilters
emilmont 1:fdd22bb7aa52 45 */
emilmont 1:fdd22bb7aa52 46
emilmont 1:fdd22bb7aa52 47 /**
emilmont 1:fdd22bb7aa52 48 * @defgroup FIR_Lattice Finite Impulse Response (FIR) Lattice Filters
emilmont 1:fdd22bb7aa52 49 *
emilmont 1:fdd22bb7aa52 50 * This set of functions implements Finite Impulse Response (FIR) lattice filters
emilmont 1:fdd22bb7aa52 51 * for Q15, Q31 and floating-point data types. Lattice filters are used in a
emilmont 1:fdd22bb7aa52 52 * variety of adaptive filter applications. The filter structure is feedforward and
emilmont 1:fdd22bb7aa52 53 * the net impulse response is finite length.
emilmont 1:fdd22bb7aa52 54 * The functions operate on blocks
emilmont 1:fdd22bb7aa52 55 * of input and output data and each call to the function processes
emilmont 1:fdd22bb7aa52 56 * <code>blockSize</code> samples through the filter. <code>pSrc</code> and
emilmont 1:fdd22bb7aa52 57 * <code>pDst</code> point to input and output arrays containing <code>blockSize</code> values.
emilmont 1:fdd22bb7aa52 58 *
emilmont 1:fdd22bb7aa52 59 * \par Algorithm:
emilmont 1:fdd22bb7aa52 60 * \image html FIRLattice.gif "Finite Impulse Response Lattice filter"
emilmont 1:fdd22bb7aa52 61 * The following difference equation is implemented:
emilmont 1:fdd22bb7aa52 62 * <pre>
emilmont 1:fdd22bb7aa52 63 * f0[n] = g0[n] = x[n]
emilmont 1:fdd22bb7aa52 64 * fm[n] = fm-1[n] + km * gm-1[n-1] for m = 1, 2, ...M
emilmont 1:fdd22bb7aa52 65 * gm[n] = km * fm-1[n] + gm-1[n-1] for m = 1, 2, ...M
emilmont 1:fdd22bb7aa52 66 * y[n] = fM[n]
emilmont 1:fdd22bb7aa52 67 * </pre>
emilmont 1:fdd22bb7aa52 68 * \par
emilmont 1:fdd22bb7aa52 69 * <code>pCoeffs</code> points to tha array of reflection coefficients of size <code>numStages</code>.
emilmont 1:fdd22bb7aa52 70 * Reflection Coefficients are stored in the following order.
emilmont 1:fdd22bb7aa52 71 * \par
emilmont 1:fdd22bb7aa52 72 * <pre>
emilmont 1:fdd22bb7aa52 73 * {k1, k2, ..., kM}
emilmont 1:fdd22bb7aa52 74 * </pre>
emilmont 1:fdd22bb7aa52 75 * where M is number of stages
emilmont 1:fdd22bb7aa52 76 * \par
emilmont 1:fdd22bb7aa52 77 * <code>pState</code> points to a state array of size <code>numStages</code>.
emilmont 1:fdd22bb7aa52 78 * The state variables (g values) hold previous inputs and are stored in the following order.
emilmont 1:fdd22bb7aa52 79 * <pre>
emilmont 1:fdd22bb7aa52 80 * {g0[n], g1[n], g2[n] ...gM-1[n]}
emilmont 1:fdd22bb7aa52 81 * </pre>
emilmont 1:fdd22bb7aa52 82 * The state variables are updated after each block of data is processed; the coefficients are untouched.
emilmont 1:fdd22bb7aa52 83 * \par Instance Structure
emilmont 1:fdd22bb7aa52 84 * The coefficients and state variables for a filter are stored together in an instance data structure.
emilmont 1:fdd22bb7aa52 85 * A separate instance structure must be defined for each filter.
emilmont 1:fdd22bb7aa52 86 * Coefficient arrays may be shared among several instances while state variable arrays cannot be shared.
emilmont 1:fdd22bb7aa52 87 * There are separate instance structure declarations for each of the 3 supported data types.
emilmont 1:fdd22bb7aa52 88 *
emilmont 1:fdd22bb7aa52 89 * \par Initialization Functions
emilmont 1:fdd22bb7aa52 90 * There is also an associated initialization function for each data type.
emilmont 1:fdd22bb7aa52 91 * The initialization function performs the following operations:
emilmont 1:fdd22bb7aa52 92 * - Sets the values of the internal structure fields.
emilmont 1:fdd22bb7aa52 93 * - Zeros out the values in the state buffer.
mbed_official 3:7a284390b0ce 94 * To do this manually without calling the init function, assign the follow subfields of the instance structure:
mbed_official 3:7a284390b0ce 95 * numStages, pCoeffs, pState. Also set all of the values in pState to zero.
emilmont 1:fdd22bb7aa52 96 *
emilmont 1:fdd22bb7aa52 97 * \par
emilmont 1:fdd22bb7aa52 98 * Use of the initialization function is optional.
emilmont 1:fdd22bb7aa52 99 * However, if the initialization function is used, then the instance structure cannot be placed into a const data section.
emilmont 1:fdd22bb7aa52 100 * To place an instance structure into a const data section, the instance structure must be manually initialized.
emilmont 1:fdd22bb7aa52 101 * Set the values in the state buffer to zeros and then manually initialize the instance structure as follows:
emilmont 1:fdd22bb7aa52 102 * <pre>
emilmont 1:fdd22bb7aa52 103 *arm_fir_lattice_instance_f32 S = {numStages, pState, pCoeffs};
emilmont 1:fdd22bb7aa52 104 *arm_fir_lattice_instance_q31 S = {numStages, pState, pCoeffs};
emilmont 1:fdd22bb7aa52 105 *arm_fir_lattice_instance_q15 S = {numStages, pState, pCoeffs};
emilmont 1:fdd22bb7aa52 106 * </pre>
emilmont 1:fdd22bb7aa52 107 * \par
emilmont 1:fdd22bb7aa52 108 * where <code>numStages</code> is the number of stages in the filter; <code>pState</code> is the address of the state buffer;
emilmont 1:fdd22bb7aa52 109 * <code>pCoeffs</code> is the address of the coefficient buffer.
emilmont 1:fdd22bb7aa52 110 * \par Fixed-Point Behavior
emilmont 1:fdd22bb7aa52 111 * Care must be taken when using the fixed-point versions of the FIR Lattice filter functions.
emilmont 1:fdd22bb7aa52 112 * In particular, the overflow and saturation behavior of the accumulator used in each function must be considered.
emilmont 1:fdd22bb7aa52 113 * Refer to the function specific documentation below for usage guidelines.
emilmont 1:fdd22bb7aa52 114 */
emilmont 1:fdd22bb7aa52 115
emilmont 1:fdd22bb7aa52 116 /**
emilmont 1:fdd22bb7aa52 117 * @addtogroup FIR_Lattice
emilmont 1:fdd22bb7aa52 118 * @{
emilmont 1:fdd22bb7aa52 119 */
emilmont 1:fdd22bb7aa52 120
emilmont 1:fdd22bb7aa52 121
emilmont 1:fdd22bb7aa52 122 /**
emilmont 1:fdd22bb7aa52 123 * @brief Processing function for the floating-point FIR lattice filter.
emilmont 1:fdd22bb7aa52 124 * @param[in] *S points to an instance of the floating-point FIR lattice structure.
emilmont 1:fdd22bb7aa52 125 * @param[in] *pSrc points to the block of input data.
emilmont 1:fdd22bb7aa52 126 * @param[out] *pDst points to the block of output data
emilmont 1:fdd22bb7aa52 127 * @param[in] blockSize number of samples to process.
emilmont 1:fdd22bb7aa52 128 * @return none.
emilmont 1:fdd22bb7aa52 129 */
emilmont 1:fdd22bb7aa52 130
emilmont 1:fdd22bb7aa52 131 void arm_fir_lattice_f32(
emilmont 1:fdd22bb7aa52 132 const arm_fir_lattice_instance_f32 * S,
emilmont 1:fdd22bb7aa52 133 float32_t * pSrc,
emilmont 1:fdd22bb7aa52 134 float32_t * pDst,
emilmont 1:fdd22bb7aa52 135 uint32_t blockSize)
emilmont 1:fdd22bb7aa52 136 {
emilmont 1:fdd22bb7aa52 137 float32_t *pState; /* State pointer */
emilmont 1:fdd22bb7aa52 138 float32_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emilmont 1:fdd22bb7aa52 139 float32_t *px; /* temporary state pointer */
emilmont 1:fdd22bb7aa52 140 float32_t *pk; /* temporary coefficient pointer */
emilmont 1:fdd22bb7aa52 141
emilmont 1:fdd22bb7aa52 142
mbed_official 3:7a284390b0ce 143 #ifndef ARM_MATH_CM0_FAMILY
emilmont 1:fdd22bb7aa52 144
emilmont 1:fdd22bb7aa52 145 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 146
emilmont 1:fdd22bb7aa52 147 float32_t fcurr1, fnext1, gcurr1, gnext1; /* temporary variables for first sample in loop unrolling */
emilmont 1:fdd22bb7aa52 148 float32_t fcurr2, fnext2, gnext2; /* temporary variables for second sample in loop unrolling */
emilmont 1:fdd22bb7aa52 149 float32_t fcurr3, fnext3, gnext3; /* temporary variables for third sample in loop unrolling */
emilmont 1:fdd22bb7aa52 150 float32_t fcurr4, fnext4, gnext4; /* temporary variables for fourth sample in loop unrolling */
emilmont 1:fdd22bb7aa52 151 uint32_t numStages = S->numStages; /* Number of stages in the filter */
emilmont 1:fdd22bb7aa52 152 uint32_t blkCnt, stageCnt; /* temporary variables for counts */
emilmont 1:fdd22bb7aa52 153
emilmont 1:fdd22bb7aa52 154 gcurr1 = 0.0f;
emilmont 1:fdd22bb7aa52 155 pState = &S->pState[0];
emilmont 1:fdd22bb7aa52 156
emilmont 1:fdd22bb7aa52 157 blkCnt = blockSize >> 2;
emilmont 1:fdd22bb7aa52 158
emilmont 1:fdd22bb7aa52 159 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
emilmont 1:fdd22bb7aa52 160 a second loop below computes the remaining 1 to 3 samples. */
emilmont 1:fdd22bb7aa52 161 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 162 {
emilmont 1:fdd22bb7aa52 163
emilmont 1:fdd22bb7aa52 164 /* Read two samples from input buffer */
emilmont 1:fdd22bb7aa52 165 /* f0(n) = x(n) */
emilmont 1:fdd22bb7aa52 166 fcurr1 = *pSrc++;
emilmont 1:fdd22bb7aa52 167 fcurr2 = *pSrc++;
emilmont 1:fdd22bb7aa52 168
emilmont 1:fdd22bb7aa52 169 /* Initialize coeff pointer */
emilmont 1:fdd22bb7aa52 170 pk = (pCoeffs);
emilmont 1:fdd22bb7aa52 171
emilmont 1:fdd22bb7aa52 172 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 173 px = pState;
emilmont 1:fdd22bb7aa52 174
emilmont 1:fdd22bb7aa52 175 /* Read g0(n-1) from state */
emilmont 1:fdd22bb7aa52 176 gcurr1 = *px;
emilmont 1:fdd22bb7aa52 177
emilmont 1:fdd22bb7aa52 178 /* Process first sample for first tap */
emilmont 1:fdd22bb7aa52 179 /* f1(n) = f0(n) + K1 * g0(n-1) */
emilmont 1:fdd22bb7aa52 180 fnext1 = fcurr1 + ((*pk) * gcurr1);
emilmont 1:fdd22bb7aa52 181 /* g1(n) = f0(n) * K1 + g0(n-1) */
emilmont 1:fdd22bb7aa52 182 gnext1 = (fcurr1 * (*pk)) + gcurr1;
emilmont 1:fdd22bb7aa52 183
emilmont 1:fdd22bb7aa52 184 /* Process second sample for first tap */
emilmont 1:fdd22bb7aa52 185 /* for sample 2 processing */
emilmont 1:fdd22bb7aa52 186 fnext2 = fcurr2 + ((*pk) * fcurr1);
emilmont 1:fdd22bb7aa52 187 gnext2 = (fcurr2 * (*pk)) + fcurr1;
emilmont 1:fdd22bb7aa52 188
emilmont 1:fdd22bb7aa52 189 /* Read next two samples from input buffer */
emilmont 1:fdd22bb7aa52 190 /* f0(n+2) = x(n+2) */
emilmont 1:fdd22bb7aa52 191 fcurr3 = *pSrc++;
emilmont 1:fdd22bb7aa52 192 fcurr4 = *pSrc++;
emilmont 1:fdd22bb7aa52 193
emilmont 1:fdd22bb7aa52 194 /* Copy only last input samples into the state buffer
emilmont 1:fdd22bb7aa52 195 which will be used for next four samples processing */
emilmont 1:fdd22bb7aa52 196 *px++ = fcurr4;
emilmont 1:fdd22bb7aa52 197
emilmont 1:fdd22bb7aa52 198 /* Process third sample for first tap */
emilmont 1:fdd22bb7aa52 199 fnext3 = fcurr3 + ((*pk) * fcurr2);
emilmont 1:fdd22bb7aa52 200 gnext3 = (fcurr3 * (*pk)) + fcurr2;
emilmont 1:fdd22bb7aa52 201
emilmont 1:fdd22bb7aa52 202 /* Process fourth sample for first tap */
emilmont 1:fdd22bb7aa52 203 fnext4 = fcurr4 + ((*pk) * fcurr3);
emilmont 1:fdd22bb7aa52 204 gnext4 = (fcurr4 * (*pk++)) + fcurr3;
emilmont 1:fdd22bb7aa52 205
emilmont 1:fdd22bb7aa52 206 /* Update of f values for next coefficient set processing */
emilmont 1:fdd22bb7aa52 207 fcurr1 = fnext1;
emilmont 1:fdd22bb7aa52 208 fcurr2 = fnext2;
emilmont 1:fdd22bb7aa52 209 fcurr3 = fnext3;
emilmont 1:fdd22bb7aa52 210 fcurr4 = fnext4;
emilmont 1:fdd22bb7aa52 211
emilmont 1:fdd22bb7aa52 212 /* Loop unrolling. Process 4 taps at a time . */
emilmont 1:fdd22bb7aa52 213 stageCnt = (numStages - 1u) >> 2u;
emilmont 1:fdd22bb7aa52 214
emilmont 1:fdd22bb7aa52 215 /* Loop over the number of taps. Unroll by a factor of 4.
emilmont 1:fdd22bb7aa52 216 ** Repeat until we've computed numStages-3 coefficients. */
emilmont 1:fdd22bb7aa52 217
emilmont 1:fdd22bb7aa52 218 /* Process 2nd, 3rd, 4th and 5th taps ... here */
emilmont 1:fdd22bb7aa52 219 while(stageCnt > 0u)
emilmont 1:fdd22bb7aa52 220 {
emilmont 1:fdd22bb7aa52 221 /* Read g1(n-1), g3(n-1) .... from state */
emilmont 1:fdd22bb7aa52 222 gcurr1 = *px;
emilmont 1:fdd22bb7aa52 223
emilmont 1:fdd22bb7aa52 224 /* save g1(n) in state buffer */
emilmont 1:fdd22bb7aa52 225 *px++ = gnext4;
emilmont 1:fdd22bb7aa52 226
emilmont 1:fdd22bb7aa52 227 /* Process first sample for 2nd, 6th .. tap */
emilmont 1:fdd22bb7aa52 228 /* Sample processing for K2, K6.... */
emilmont 1:fdd22bb7aa52 229 /* f2(n) = f1(n) + K2 * g1(n-1) */
emilmont 1:fdd22bb7aa52 230 fnext1 = fcurr1 + ((*pk) * gcurr1);
emilmont 1:fdd22bb7aa52 231 /* Process second sample for 2nd, 6th .. tap */
emilmont 1:fdd22bb7aa52 232 /* for sample 2 processing */
emilmont 1:fdd22bb7aa52 233 fnext2 = fcurr2 + ((*pk) * gnext1);
emilmont 1:fdd22bb7aa52 234 /* Process third sample for 2nd, 6th .. tap */
emilmont 1:fdd22bb7aa52 235 fnext3 = fcurr3 + ((*pk) * gnext2);
emilmont 1:fdd22bb7aa52 236 /* Process fourth sample for 2nd, 6th .. tap */
emilmont 1:fdd22bb7aa52 237 fnext4 = fcurr4 + ((*pk) * gnext3);
emilmont 1:fdd22bb7aa52 238
emilmont 1:fdd22bb7aa52 239 /* g2(n) = f1(n) * K2 + g1(n-1) */
emilmont 1:fdd22bb7aa52 240 /* Calculation of state values for next stage */
emilmont 1:fdd22bb7aa52 241 gnext4 = (fcurr4 * (*pk)) + gnext3;
emilmont 1:fdd22bb7aa52 242 gnext3 = (fcurr3 * (*pk)) + gnext2;
emilmont 1:fdd22bb7aa52 243 gnext2 = (fcurr2 * (*pk)) + gnext1;
emilmont 1:fdd22bb7aa52 244 gnext1 = (fcurr1 * (*pk++)) + gcurr1;
emilmont 1:fdd22bb7aa52 245
emilmont 1:fdd22bb7aa52 246
emilmont 1:fdd22bb7aa52 247 /* Read g2(n-1), g4(n-1) .... from state */
emilmont 1:fdd22bb7aa52 248 gcurr1 = *px;
emilmont 1:fdd22bb7aa52 249
emilmont 1:fdd22bb7aa52 250 /* save g2(n) in state buffer */
emilmont 1:fdd22bb7aa52 251 *px++ = gnext4;
emilmont 1:fdd22bb7aa52 252
emilmont 1:fdd22bb7aa52 253 /* Sample processing for K3, K7.... */
emilmont 1:fdd22bb7aa52 254 /* Process first sample for 3rd, 7th .. tap */
emilmont 1:fdd22bb7aa52 255 /* f3(n) = f2(n) + K3 * g2(n-1) */
emilmont 1:fdd22bb7aa52 256 fcurr1 = fnext1 + ((*pk) * gcurr1);
emilmont 1:fdd22bb7aa52 257 /* Process second sample for 3rd, 7th .. tap */
emilmont 1:fdd22bb7aa52 258 fcurr2 = fnext2 + ((*pk) * gnext1);
emilmont 1:fdd22bb7aa52 259 /* Process third sample for 3rd, 7th .. tap */
emilmont 1:fdd22bb7aa52 260 fcurr3 = fnext3 + ((*pk) * gnext2);
emilmont 1:fdd22bb7aa52 261 /* Process fourth sample for 3rd, 7th .. tap */
emilmont 1:fdd22bb7aa52 262 fcurr4 = fnext4 + ((*pk) * gnext3);
emilmont 1:fdd22bb7aa52 263
emilmont 1:fdd22bb7aa52 264 /* Calculation of state values for next stage */
emilmont 1:fdd22bb7aa52 265 /* g3(n) = f2(n) * K3 + g2(n-1) */
emilmont 1:fdd22bb7aa52 266 gnext4 = (fnext4 * (*pk)) + gnext3;
emilmont 1:fdd22bb7aa52 267 gnext3 = (fnext3 * (*pk)) + gnext2;
emilmont 1:fdd22bb7aa52 268 gnext2 = (fnext2 * (*pk)) + gnext1;
emilmont 1:fdd22bb7aa52 269 gnext1 = (fnext1 * (*pk++)) + gcurr1;
emilmont 1:fdd22bb7aa52 270
emilmont 1:fdd22bb7aa52 271
emilmont 1:fdd22bb7aa52 272 /* Read g1(n-1), g3(n-1) .... from state */
emilmont 1:fdd22bb7aa52 273 gcurr1 = *px;
emilmont 1:fdd22bb7aa52 274
emilmont 1:fdd22bb7aa52 275 /* save g3(n) in state buffer */
emilmont 1:fdd22bb7aa52 276 *px++ = gnext4;
emilmont 1:fdd22bb7aa52 277
emilmont 1:fdd22bb7aa52 278 /* Sample processing for K4, K8.... */
emilmont 1:fdd22bb7aa52 279 /* Process first sample for 4th, 8th .. tap */
emilmont 1:fdd22bb7aa52 280 /* f4(n) = f3(n) + K4 * g3(n-1) */
emilmont 1:fdd22bb7aa52 281 fnext1 = fcurr1 + ((*pk) * gcurr1);
emilmont 1:fdd22bb7aa52 282 /* Process second sample for 4th, 8th .. tap */
emilmont 1:fdd22bb7aa52 283 /* for sample 2 processing */
emilmont 1:fdd22bb7aa52 284 fnext2 = fcurr2 + ((*pk) * gnext1);
emilmont 1:fdd22bb7aa52 285 /* Process third sample for 4th, 8th .. tap */
emilmont 1:fdd22bb7aa52 286 fnext3 = fcurr3 + ((*pk) * gnext2);
emilmont 1:fdd22bb7aa52 287 /* Process fourth sample for 4th, 8th .. tap */
emilmont 1:fdd22bb7aa52 288 fnext4 = fcurr4 + ((*pk) * gnext3);
emilmont 1:fdd22bb7aa52 289
emilmont 1:fdd22bb7aa52 290 /* g4(n) = f3(n) * K4 + g3(n-1) */
emilmont 1:fdd22bb7aa52 291 /* Calculation of state values for next stage */
emilmont 1:fdd22bb7aa52 292 gnext4 = (fcurr4 * (*pk)) + gnext3;
emilmont 1:fdd22bb7aa52 293 gnext3 = (fcurr3 * (*pk)) + gnext2;
emilmont 1:fdd22bb7aa52 294 gnext2 = (fcurr2 * (*pk)) + gnext1;
emilmont 1:fdd22bb7aa52 295 gnext1 = (fcurr1 * (*pk++)) + gcurr1;
emilmont 1:fdd22bb7aa52 296
emilmont 1:fdd22bb7aa52 297 /* Read g2(n-1), g4(n-1) .... from state */
emilmont 1:fdd22bb7aa52 298 gcurr1 = *px;
emilmont 1:fdd22bb7aa52 299
emilmont 1:fdd22bb7aa52 300 /* save g4(n) in state buffer */
emilmont 1:fdd22bb7aa52 301 *px++ = gnext4;
emilmont 1:fdd22bb7aa52 302
emilmont 1:fdd22bb7aa52 303 /* Sample processing for K5, K9.... */
emilmont 1:fdd22bb7aa52 304 /* Process first sample for 5th, 9th .. tap */
emilmont 1:fdd22bb7aa52 305 /* f5(n) = f4(n) + K5 * g4(n-1) */
emilmont 1:fdd22bb7aa52 306 fcurr1 = fnext1 + ((*pk) * gcurr1);
emilmont 1:fdd22bb7aa52 307 /* Process second sample for 5th, 9th .. tap */
emilmont 1:fdd22bb7aa52 308 fcurr2 = fnext2 + ((*pk) * gnext1);
emilmont 1:fdd22bb7aa52 309 /* Process third sample for 5th, 9th .. tap */
emilmont 1:fdd22bb7aa52 310 fcurr3 = fnext3 + ((*pk) * gnext2);
emilmont 1:fdd22bb7aa52 311 /* Process fourth sample for 5th, 9th .. tap */
emilmont 1:fdd22bb7aa52 312 fcurr4 = fnext4 + ((*pk) * gnext3);
emilmont 1:fdd22bb7aa52 313
emilmont 1:fdd22bb7aa52 314 /* Calculation of state values for next stage */
emilmont 1:fdd22bb7aa52 315 /* g5(n) = f4(n) * K5 + g4(n-1) */
emilmont 1:fdd22bb7aa52 316 gnext4 = (fnext4 * (*pk)) + gnext3;
emilmont 1:fdd22bb7aa52 317 gnext3 = (fnext3 * (*pk)) + gnext2;
emilmont 1:fdd22bb7aa52 318 gnext2 = (fnext2 * (*pk)) + gnext1;
emilmont 1:fdd22bb7aa52 319 gnext1 = (fnext1 * (*pk++)) + gcurr1;
emilmont 1:fdd22bb7aa52 320
emilmont 1:fdd22bb7aa52 321 stageCnt--;
emilmont 1:fdd22bb7aa52 322 }
emilmont 1:fdd22bb7aa52 323
emilmont 1:fdd22bb7aa52 324 /* If the (filter length -1) is not a multiple of 4, compute the remaining filter taps */
emilmont 1:fdd22bb7aa52 325 stageCnt = (numStages - 1u) % 0x4u;
emilmont 1:fdd22bb7aa52 326
emilmont 1:fdd22bb7aa52 327 while(stageCnt > 0u)
emilmont 1:fdd22bb7aa52 328 {
emilmont 1:fdd22bb7aa52 329 gcurr1 = *px;
emilmont 1:fdd22bb7aa52 330
emilmont 1:fdd22bb7aa52 331 /* save g value in state buffer */
emilmont 1:fdd22bb7aa52 332 *px++ = gnext4;
emilmont 1:fdd22bb7aa52 333
emilmont 1:fdd22bb7aa52 334 /* Process four samples for last three taps here */
emilmont 1:fdd22bb7aa52 335 fnext1 = fcurr1 + ((*pk) * gcurr1);
emilmont 1:fdd22bb7aa52 336 fnext2 = fcurr2 + ((*pk) * gnext1);
emilmont 1:fdd22bb7aa52 337 fnext3 = fcurr3 + ((*pk) * gnext2);
emilmont 1:fdd22bb7aa52 338 fnext4 = fcurr4 + ((*pk) * gnext3);
emilmont 1:fdd22bb7aa52 339
emilmont 1:fdd22bb7aa52 340 /* g1(n) = f0(n) * K1 + g0(n-1) */
emilmont 1:fdd22bb7aa52 341 gnext4 = (fcurr4 * (*pk)) + gnext3;
emilmont 1:fdd22bb7aa52 342 gnext3 = (fcurr3 * (*pk)) + gnext2;
emilmont 1:fdd22bb7aa52 343 gnext2 = (fcurr2 * (*pk)) + gnext1;
emilmont 1:fdd22bb7aa52 344 gnext1 = (fcurr1 * (*pk++)) + gcurr1;
emilmont 1:fdd22bb7aa52 345
emilmont 1:fdd22bb7aa52 346 /* Update of f values for next coefficient set processing */
emilmont 1:fdd22bb7aa52 347 fcurr1 = fnext1;
emilmont 1:fdd22bb7aa52 348 fcurr2 = fnext2;
emilmont 1:fdd22bb7aa52 349 fcurr3 = fnext3;
emilmont 1:fdd22bb7aa52 350 fcurr4 = fnext4;
emilmont 1:fdd22bb7aa52 351
emilmont 1:fdd22bb7aa52 352 stageCnt--;
emilmont 1:fdd22bb7aa52 353
emilmont 1:fdd22bb7aa52 354 }
emilmont 1:fdd22bb7aa52 355
emilmont 1:fdd22bb7aa52 356 /* The results in the 4 accumulators, store in the destination buffer. */
emilmont 1:fdd22bb7aa52 357 /* y(n) = fN(n) */
emilmont 1:fdd22bb7aa52 358 *pDst++ = fcurr1;
emilmont 1:fdd22bb7aa52 359 *pDst++ = fcurr2;
emilmont 1:fdd22bb7aa52 360 *pDst++ = fcurr3;
emilmont 1:fdd22bb7aa52 361 *pDst++ = fcurr4;
emilmont 1:fdd22bb7aa52 362
emilmont 1:fdd22bb7aa52 363 blkCnt--;
emilmont 1:fdd22bb7aa52 364 }
emilmont 1:fdd22bb7aa52 365
emilmont 1:fdd22bb7aa52 366 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
emilmont 1:fdd22bb7aa52 367 ** No loop unrolling is used. */
emilmont 1:fdd22bb7aa52 368 blkCnt = blockSize % 0x4u;
emilmont 1:fdd22bb7aa52 369
emilmont 1:fdd22bb7aa52 370 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 371 {
emilmont 1:fdd22bb7aa52 372 /* f0(n) = x(n) */
emilmont 1:fdd22bb7aa52 373 fcurr1 = *pSrc++;
emilmont 1:fdd22bb7aa52 374
emilmont 1:fdd22bb7aa52 375 /* Initialize coeff pointer */
emilmont 1:fdd22bb7aa52 376 pk = (pCoeffs);
emilmont 1:fdd22bb7aa52 377
emilmont 1:fdd22bb7aa52 378 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 379 px = pState;
emilmont 1:fdd22bb7aa52 380
emilmont 1:fdd22bb7aa52 381 /* read g2(n) from state buffer */
emilmont 1:fdd22bb7aa52 382 gcurr1 = *px;
emilmont 1:fdd22bb7aa52 383
emilmont 1:fdd22bb7aa52 384 /* for sample 1 processing */
emilmont 1:fdd22bb7aa52 385 /* f1(n) = f0(n) + K1 * g0(n-1) */
emilmont 1:fdd22bb7aa52 386 fnext1 = fcurr1 + ((*pk) * gcurr1);
emilmont 1:fdd22bb7aa52 387 /* g1(n) = f0(n) * K1 + g0(n-1) */
emilmont 1:fdd22bb7aa52 388 gnext1 = (fcurr1 * (*pk++)) + gcurr1;
emilmont 1:fdd22bb7aa52 389
emilmont 1:fdd22bb7aa52 390 /* save g1(n) in state buffer */
emilmont 1:fdd22bb7aa52 391 *px++ = fcurr1;
emilmont 1:fdd22bb7aa52 392
emilmont 1:fdd22bb7aa52 393 /* f1(n) is saved in fcurr1
emilmont 1:fdd22bb7aa52 394 for next stage processing */
emilmont 1:fdd22bb7aa52 395 fcurr1 = fnext1;
emilmont 1:fdd22bb7aa52 396
emilmont 1:fdd22bb7aa52 397 stageCnt = (numStages - 1u);
emilmont 1:fdd22bb7aa52 398
emilmont 1:fdd22bb7aa52 399 /* stage loop */
emilmont 1:fdd22bb7aa52 400 while(stageCnt > 0u)
emilmont 1:fdd22bb7aa52 401 {
emilmont 1:fdd22bb7aa52 402 /* read g2(n) from state buffer */
emilmont 1:fdd22bb7aa52 403 gcurr1 = *px;
emilmont 1:fdd22bb7aa52 404
emilmont 1:fdd22bb7aa52 405 /* save g1(n) in state buffer */
emilmont 1:fdd22bb7aa52 406 *px++ = gnext1;
emilmont 1:fdd22bb7aa52 407
emilmont 1:fdd22bb7aa52 408 /* Sample processing for K2, K3.... */
emilmont 1:fdd22bb7aa52 409 /* f2(n) = f1(n) + K2 * g1(n-1) */
emilmont 1:fdd22bb7aa52 410 fnext1 = fcurr1 + ((*pk) * gcurr1);
emilmont 1:fdd22bb7aa52 411 /* g2(n) = f1(n) * K2 + g1(n-1) */
emilmont 1:fdd22bb7aa52 412 gnext1 = (fcurr1 * (*pk++)) + gcurr1;
emilmont 1:fdd22bb7aa52 413
emilmont 1:fdd22bb7aa52 414 /* f1(n) is saved in fcurr1
emilmont 1:fdd22bb7aa52 415 for next stage processing */
emilmont 1:fdd22bb7aa52 416 fcurr1 = fnext1;
emilmont 1:fdd22bb7aa52 417
emilmont 1:fdd22bb7aa52 418 stageCnt--;
emilmont 1:fdd22bb7aa52 419
emilmont 1:fdd22bb7aa52 420 }
emilmont 1:fdd22bb7aa52 421
emilmont 1:fdd22bb7aa52 422 /* y(n) = fN(n) */
emilmont 1:fdd22bb7aa52 423 *pDst++ = fcurr1;
emilmont 1:fdd22bb7aa52 424
emilmont 1:fdd22bb7aa52 425 blkCnt--;
emilmont 1:fdd22bb7aa52 426
emilmont 1:fdd22bb7aa52 427 }
emilmont 1:fdd22bb7aa52 428
emilmont 1:fdd22bb7aa52 429 #else
emilmont 1:fdd22bb7aa52 430
emilmont 1:fdd22bb7aa52 431 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 432
emilmont 1:fdd22bb7aa52 433 float32_t fcurr, fnext, gcurr, gnext; /* temporary variables */
emilmont 1:fdd22bb7aa52 434 uint32_t numStages = S->numStages; /* Length of the filter */
emilmont 1:fdd22bb7aa52 435 uint32_t blkCnt, stageCnt; /* temporary variables for counts */
emilmont 1:fdd22bb7aa52 436
emilmont 1:fdd22bb7aa52 437 pState = &S->pState[0];
emilmont 1:fdd22bb7aa52 438
emilmont 1:fdd22bb7aa52 439 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 440
emilmont 1:fdd22bb7aa52 441 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 442 {
emilmont 1:fdd22bb7aa52 443 /* f0(n) = x(n) */
emilmont 1:fdd22bb7aa52 444 fcurr = *pSrc++;
emilmont 1:fdd22bb7aa52 445
emilmont 1:fdd22bb7aa52 446 /* Initialize coeff pointer */
emilmont 1:fdd22bb7aa52 447 pk = pCoeffs;
emilmont 1:fdd22bb7aa52 448
emilmont 1:fdd22bb7aa52 449 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 450 px = pState;
emilmont 1:fdd22bb7aa52 451
emilmont 1:fdd22bb7aa52 452 /* read g0(n-1) from state buffer */
emilmont 1:fdd22bb7aa52 453 gcurr = *px;
emilmont 1:fdd22bb7aa52 454
emilmont 1:fdd22bb7aa52 455 /* for sample 1 processing */
emilmont 1:fdd22bb7aa52 456 /* f1(n) = f0(n) + K1 * g0(n-1) */
emilmont 1:fdd22bb7aa52 457 fnext = fcurr + ((*pk) * gcurr);
emilmont 1:fdd22bb7aa52 458 /* g1(n) = f0(n) * K1 + g0(n-1) */
emilmont 1:fdd22bb7aa52 459 gnext = (fcurr * (*pk++)) + gcurr;
emilmont 1:fdd22bb7aa52 460
emilmont 1:fdd22bb7aa52 461 /* save f0(n) in state buffer */
emilmont 1:fdd22bb7aa52 462 *px++ = fcurr;
emilmont 1:fdd22bb7aa52 463
emilmont 1:fdd22bb7aa52 464 /* f1(n) is saved in fcurr
emilmont 1:fdd22bb7aa52 465 for next stage processing */
emilmont 1:fdd22bb7aa52 466 fcurr = fnext;
emilmont 1:fdd22bb7aa52 467
emilmont 1:fdd22bb7aa52 468 stageCnt = (numStages - 1u);
emilmont 1:fdd22bb7aa52 469
emilmont 1:fdd22bb7aa52 470 /* stage loop */
emilmont 1:fdd22bb7aa52 471 while(stageCnt > 0u)
emilmont 1:fdd22bb7aa52 472 {
emilmont 1:fdd22bb7aa52 473 /* read g2(n) from state buffer */
emilmont 1:fdd22bb7aa52 474 gcurr = *px;
emilmont 1:fdd22bb7aa52 475
emilmont 1:fdd22bb7aa52 476 /* save g1(n) in state buffer */
emilmont 1:fdd22bb7aa52 477 *px++ = gnext;
emilmont 1:fdd22bb7aa52 478
emilmont 1:fdd22bb7aa52 479 /* Sample processing for K2, K3.... */
emilmont 1:fdd22bb7aa52 480 /* f2(n) = f1(n) + K2 * g1(n-1) */
emilmont 1:fdd22bb7aa52 481 fnext = fcurr + ((*pk) * gcurr);
emilmont 1:fdd22bb7aa52 482 /* g2(n) = f1(n) * K2 + g1(n-1) */
emilmont 1:fdd22bb7aa52 483 gnext = (fcurr * (*pk++)) + gcurr;
emilmont 1:fdd22bb7aa52 484
emilmont 1:fdd22bb7aa52 485 /* f1(n) is saved in fcurr1
emilmont 1:fdd22bb7aa52 486 for next stage processing */
emilmont 1:fdd22bb7aa52 487 fcurr = fnext;
emilmont 1:fdd22bb7aa52 488
emilmont 1:fdd22bb7aa52 489 stageCnt--;
emilmont 1:fdd22bb7aa52 490
emilmont 1:fdd22bb7aa52 491 }
emilmont 1:fdd22bb7aa52 492
emilmont 1:fdd22bb7aa52 493 /* y(n) = fN(n) */
emilmont 1:fdd22bb7aa52 494 *pDst++ = fcurr;
emilmont 1:fdd22bb7aa52 495
emilmont 1:fdd22bb7aa52 496 blkCnt--;
emilmont 1:fdd22bb7aa52 497
emilmont 1:fdd22bb7aa52 498 }
emilmont 1:fdd22bb7aa52 499
mbed_official 3:7a284390b0ce 500 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
emilmont 1:fdd22bb7aa52 501
emilmont 1:fdd22bb7aa52 502 }
emilmont 1:fdd22bb7aa52 503
emilmont 1:fdd22bb7aa52 504 /**
emilmont 1:fdd22bb7aa52 505 * @} end of FIR_Lattice group
emilmont 1:fdd22bb7aa52 506 */