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_lattice_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_lattice_f32.c |
emh203 | 0:3d9c67d97d6f | 9 | * |
emh203 | 0:3d9c67d97d6f | 10 | * Description: Processing function for the floating-point FIR Lattice filter. |
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_Lattice Finite Impulse Response (FIR) Lattice Filters |
emh203 | 0:3d9c67d97d6f | 49 | * |
emh203 | 0:3d9c67d97d6f | 50 | * This set of functions implements Finite Impulse Response (FIR) lattice filters |
emh203 | 0:3d9c67d97d6f | 51 | * for Q15, Q31 and floating-point data types. Lattice filters are used in a |
emh203 | 0:3d9c67d97d6f | 52 | * variety of adaptive filter applications. The filter structure is feedforward and |
emh203 | 0:3d9c67d97d6f | 53 | * the net impulse response is finite length. |
emh203 | 0:3d9c67d97d6f | 54 | * The functions operate on blocks |
emh203 | 0:3d9c67d97d6f | 55 | * of input and output data and each call to the function processes |
emh203 | 0:3d9c67d97d6f | 56 | * <code>blockSize</code> samples through the filter. <code>pSrc</code> and |
emh203 | 0:3d9c67d97d6f | 57 | * <code>pDst</code> point to input and output arrays containing <code>blockSize</code> values. |
emh203 | 0:3d9c67d97d6f | 58 | * |
emh203 | 0:3d9c67d97d6f | 59 | * \par Algorithm: |
emh203 | 0:3d9c67d97d6f | 60 | * \image html FIRLattice.gif "Finite Impulse Response Lattice filter" |
emh203 | 0:3d9c67d97d6f | 61 | * The following difference equation is implemented: |
emh203 | 0:3d9c67d97d6f | 62 | * <pre> |
emh203 | 0:3d9c67d97d6f | 63 | * f0[n] = g0[n] = x[n] |
emh203 | 0:3d9c67d97d6f | 64 | * fm[n] = fm-1[n] + km * gm-1[n-1] for m = 1, 2, ...M |
emh203 | 0:3d9c67d97d6f | 65 | * gm[n] = km * fm-1[n] + gm-1[n-1] for m = 1, 2, ...M |
emh203 | 0:3d9c67d97d6f | 66 | * y[n] = fM[n] |
emh203 | 0:3d9c67d97d6f | 67 | * </pre> |
emh203 | 0:3d9c67d97d6f | 68 | * \par |
emh203 | 0:3d9c67d97d6f | 69 | * <code>pCoeffs</code> points to tha array of reflection coefficients of size <code>numStages</code>. |
emh203 | 0:3d9c67d97d6f | 70 | * Reflection Coefficients are stored in the following order. |
emh203 | 0:3d9c67d97d6f | 71 | * \par |
emh203 | 0:3d9c67d97d6f | 72 | * <pre> |
emh203 | 0:3d9c67d97d6f | 73 | * {k1, k2, ..., kM} |
emh203 | 0:3d9c67d97d6f | 74 | * </pre> |
emh203 | 0:3d9c67d97d6f | 75 | * where M is number of stages |
emh203 | 0:3d9c67d97d6f | 76 | * \par |
emh203 | 0:3d9c67d97d6f | 77 | * <code>pState</code> points to a state array of size <code>numStages</code>. |
emh203 | 0:3d9c67d97d6f | 78 | * The state variables (g values) hold previous inputs and are stored in the following order. |
emh203 | 0:3d9c67d97d6f | 79 | * <pre> |
emh203 | 0:3d9c67d97d6f | 80 | * {g0[n], g1[n], g2[n] ...gM-1[n]} |
emh203 | 0:3d9c67d97d6f | 81 | * </pre> |
emh203 | 0:3d9c67d97d6f | 82 | * The state variables are updated after each block of data is processed; the coefficients are untouched. |
emh203 | 0:3d9c67d97d6f | 83 | * \par Instance Structure |
emh203 | 0:3d9c67d97d6f | 84 | * The coefficients and state variables for a filter are stored together in an instance data structure. |
emh203 | 0:3d9c67d97d6f | 85 | * A separate instance structure must be defined for each filter. |
emh203 | 0:3d9c67d97d6f | 86 | * Coefficient arrays may be shared among several instances while state variable arrays cannot be shared. |
emh203 | 0:3d9c67d97d6f | 87 | * There are separate instance structure declarations for each of the 3 supported data types. |
emh203 | 0:3d9c67d97d6f | 88 | * |
emh203 | 0:3d9c67d97d6f | 89 | * \par Initialization Functions |
emh203 | 0:3d9c67d97d6f | 90 | * There is also an associated initialization function for each data type. |
emh203 | 0:3d9c67d97d6f | 91 | * The initialization function performs the following operations: |
emh203 | 0:3d9c67d97d6f | 92 | * - Sets the values of the internal structure fields. |
emh203 | 0:3d9c67d97d6f | 93 | * - Zeros out the values in the state buffer. |
emh203 | 0:3d9c67d97d6f | 94 | * To do this manually without calling the init function, assign the follow subfields of the instance structure: |
emh203 | 0:3d9c67d97d6f | 95 | * numStages, pCoeffs, pState. Also set all of the values in pState to zero. |
emh203 | 0:3d9c67d97d6f | 96 | * |
emh203 | 0:3d9c67d97d6f | 97 | * \par |
emh203 | 0:3d9c67d97d6f | 98 | * Use of the initialization function is optional. |
emh203 | 0:3d9c67d97d6f | 99 | * However, if the initialization function is used, then the instance structure cannot be placed into a const data section. |
emh203 | 0:3d9c67d97d6f | 100 | * To place an instance structure into a const data section, the instance structure must be manually initialized. |
emh203 | 0:3d9c67d97d6f | 101 | * Set the values in the state buffer to zeros and then manually initialize the instance structure as follows: |
emh203 | 0:3d9c67d97d6f | 102 | * <pre> |
emh203 | 0:3d9c67d97d6f | 103 | *arm_fir_lattice_instance_f32 S = {numStages, pState, pCoeffs}; |
emh203 | 0:3d9c67d97d6f | 104 | *arm_fir_lattice_instance_q31 S = {numStages, pState, pCoeffs}; |
emh203 | 0:3d9c67d97d6f | 105 | *arm_fir_lattice_instance_q15 S = {numStages, pState, pCoeffs}; |
emh203 | 0:3d9c67d97d6f | 106 | * </pre> |
emh203 | 0:3d9c67d97d6f | 107 | * \par |
emh203 | 0:3d9c67d97d6f | 108 | * where <code>numStages</code> is the number of stages in the filter; <code>pState</code> is the address of the state buffer; |
emh203 | 0:3d9c67d97d6f | 109 | * <code>pCoeffs</code> is the address of the coefficient buffer. |
emh203 | 0:3d9c67d97d6f | 110 | * \par Fixed-Point Behavior |
emh203 | 0:3d9c67d97d6f | 111 | * Care must be taken when using the fixed-point versions of the FIR Lattice filter functions. |
emh203 | 0:3d9c67d97d6f | 112 | * In particular, the overflow and saturation behavior of the accumulator used in each function must be considered. |
emh203 | 0:3d9c67d97d6f | 113 | * Refer to the function specific documentation below for usage guidelines. |
emh203 | 0:3d9c67d97d6f | 114 | */ |
emh203 | 0:3d9c67d97d6f | 115 | |
emh203 | 0:3d9c67d97d6f | 116 | /** |
emh203 | 0:3d9c67d97d6f | 117 | * @addtogroup FIR_Lattice |
emh203 | 0:3d9c67d97d6f | 118 | * @{ |
emh203 | 0:3d9c67d97d6f | 119 | */ |
emh203 | 0:3d9c67d97d6f | 120 | |
emh203 | 0:3d9c67d97d6f | 121 | |
emh203 | 0:3d9c67d97d6f | 122 | /** |
emh203 | 0:3d9c67d97d6f | 123 | * @brief Processing function for the floating-point FIR lattice filter. |
emh203 | 0:3d9c67d97d6f | 124 | * @param[in] *S points to an instance of the floating-point FIR lattice structure. |
emh203 | 0:3d9c67d97d6f | 125 | * @param[in] *pSrc points to the block of input data. |
emh203 | 0:3d9c67d97d6f | 126 | * @param[out] *pDst points to the block of output data |
emh203 | 0:3d9c67d97d6f | 127 | * @param[in] blockSize number of samples to process. |
emh203 | 0:3d9c67d97d6f | 128 | * @return none. |
emh203 | 0:3d9c67d97d6f | 129 | */ |
emh203 | 0:3d9c67d97d6f | 130 | |
emh203 | 0:3d9c67d97d6f | 131 | void arm_fir_lattice_f32( |
emh203 | 0:3d9c67d97d6f | 132 | const arm_fir_lattice_instance_f32 * S, |
emh203 | 0:3d9c67d97d6f | 133 | float32_t * pSrc, |
emh203 | 0:3d9c67d97d6f | 134 | float32_t * pDst, |
emh203 | 0:3d9c67d97d6f | 135 | uint32_t blockSize) |
emh203 | 0:3d9c67d97d6f | 136 | { |
emh203 | 0:3d9c67d97d6f | 137 | float32_t *pState; /* State pointer */ |
emh203 | 0:3d9c67d97d6f | 138 | float32_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */ |
emh203 | 0:3d9c67d97d6f | 139 | float32_t *px; /* temporary state pointer */ |
emh203 | 0:3d9c67d97d6f | 140 | float32_t *pk; /* temporary coefficient pointer */ |
emh203 | 0:3d9c67d97d6f | 141 | |
emh203 | 0:3d9c67d97d6f | 142 | |
emh203 | 0:3d9c67d97d6f | 143 | #ifndef ARM_MATH_CM0_FAMILY |
emh203 | 0:3d9c67d97d6f | 144 | |
emh203 | 0:3d9c67d97d6f | 145 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
emh203 | 0:3d9c67d97d6f | 146 | |
emh203 | 0:3d9c67d97d6f | 147 | float32_t fcurr1, fnext1, gcurr1, gnext1; /* temporary variables for first sample in loop unrolling */ |
emh203 | 0:3d9c67d97d6f | 148 | float32_t fcurr2, fnext2, gnext2; /* temporary variables for second sample in loop unrolling */ |
emh203 | 0:3d9c67d97d6f | 149 | float32_t fcurr3, fnext3, gnext3; /* temporary variables for third sample in loop unrolling */ |
emh203 | 0:3d9c67d97d6f | 150 | float32_t fcurr4, fnext4, gnext4; /* temporary variables for fourth sample in loop unrolling */ |
emh203 | 0:3d9c67d97d6f | 151 | uint32_t numStages = S->numStages; /* Number of stages in the filter */ |
emh203 | 0:3d9c67d97d6f | 152 | uint32_t blkCnt, stageCnt; /* temporary variables for counts */ |
emh203 | 0:3d9c67d97d6f | 153 | |
emh203 | 0:3d9c67d97d6f | 154 | gcurr1 = 0.0f; |
emh203 | 0:3d9c67d97d6f | 155 | pState = &S->pState[0]; |
emh203 | 0:3d9c67d97d6f | 156 | |
emh203 | 0:3d9c67d97d6f | 157 | blkCnt = blockSize >> 2; |
emh203 | 0:3d9c67d97d6f | 158 | |
emh203 | 0:3d9c67d97d6f | 159 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
emh203 | 0:3d9c67d97d6f | 160 | a second loop below computes the remaining 1 to 3 samples. */ |
emh203 | 0:3d9c67d97d6f | 161 | while(blkCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 162 | { |
emh203 | 0:3d9c67d97d6f | 163 | |
emh203 | 0:3d9c67d97d6f | 164 | /* Read two samples from input buffer */ |
emh203 | 0:3d9c67d97d6f | 165 | /* f0(n) = x(n) */ |
emh203 | 0:3d9c67d97d6f | 166 | fcurr1 = *pSrc++; |
emh203 | 0:3d9c67d97d6f | 167 | fcurr2 = *pSrc++; |
emh203 | 0:3d9c67d97d6f | 168 | |
emh203 | 0:3d9c67d97d6f | 169 | /* Initialize coeff pointer */ |
emh203 | 0:3d9c67d97d6f | 170 | pk = (pCoeffs); |
emh203 | 0:3d9c67d97d6f | 171 | |
emh203 | 0:3d9c67d97d6f | 172 | /* Initialize state pointer */ |
emh203 | 0:3d9c67d97d6f | 173 | px = pState; |
emh203 | 0:3d9c67d97d6f | 174 | |
emh203 | 0:3d9c67d97d6f | 175 | /* Read g0(n-1) from state */ |
emh203 | 0:3d9c67d97d6f | 176 | gcurr1 = *px; |
emh203 | 0:3d9c67d97d6f | 177 | |
emh203 | 0:3d9c67d97d6f | 178 | /* Process first sample for first tap */ |
emh203 | 0:3d9c67d97d6f | 179 | /* f1(n) = f0(n) + K1 * g0(n-1) */ |
emh203 | 0:3d9c67d97d6f | 180 | fnext1 = fcurr1 + ((*pk) * gcurr1); |
emh203 | 0:3d9c67d97d6f | 181 | /* g1(n) = f0(n) * K1 + g0(n-1) */ |
emh203 | 0:3d9c67d97d6f | 182 | gnext1 = (fcurr1 * (*pk)) + gcurr1; |
emh203 | 0:3d9c67d97d6f | 183 | |
emh203 | 0:3d9c67d97d6f | 184 | /* Process second sample for first tap */ |
emh203 | 0:3d9c67d97d6f | 185 | /* for sample 2 processing */ |
emh203 | 0:3d9c67d97d6f | 186 | fnext2 = fcurr2 + ((*pk) * fcurr1); |
emh203 | 0:3d9c67d97d6f | 187 | gnext2 = (fcurr2 * (*pk)) + fcurr1; |
emh203 | 0:3d9c67d97d6f | 188 | |
emh203 | 0:3d9c67d97d6f | 189 | /* Read next two samples from input buffer */ |
emh203 | 0:3d9c67d97d6f | 190 | /* f0(n+2) = x(n+2) */ |
emh203 | 0:3d9c67d97d6f | 191 | fcurr3 = *pSrc++; |
emh203 | 0:3d9c67d97d6f | 192 | fcurr4 = *pSrc++; |
emh203 | 0:3d9c67d97d6f | 193 | |
emh203 | 0:3d9c67d97d6f | 194 | /* Copy only last input samples into the state buffer |
emh203 | 0:3d9c67d97d6f | 195 | which will be used for next four samples processing */ |
emh203 | 0:3d9c67d97d6f | 196 | *px++ = fcurr4; |
emh203 | 0:3d9c67d97d6f | 197 | |
emh203 | 0:3d9c67d97d6f | 198 | /* Process third sample for first tap */ |
emh203 | 0:3d9c67d97d6f | 199 | fnext3 = fcurr3 + ((*pk) * fcurr2); |
emh203 | 0:3d9c67d97d6f | 200 | gnext3 = (fcurr3 * (*pk)) + fcurr2; |
emh203 | 0:3d9c67d97d6f | 201 | |
emh203 | 0:3d9c67d97d6f | 202 | /* Process fourth sample for first tap */ |
emh203 | 0:3d9c67d97d6f | 203 | fnext4 = fcurr4 + ((*pk) * fcurr3); |
emh203 | 0:3d9c67d97d6f | 204 | gnext4 = (fcurr4 * (*pk++)) + fcurr3; |
emh203 | 0:3d9c67d97d6f | 205 | |
emh203 | 0:3d9c67d97d6f | 206 | /* Update of f values for next coefficient set processing */ |
emh203 | 0:3d9c67d97d6f | 207 | fcurr1 = fnext1; |
emh203 | 0:3d9c67d97d6f | 208 | fcurr2 = fnext2; |
emh203 | 0:3d9c67d97d6f | 209 | fcurr3 = fnext3; |
emh203 | 0:3d9c67d97d6f | 210 | fcurr4 = fnext4; |
emh203 | 0:3d9c67d97d6f | 211 | |
emh203 | 0:3d9c67d97d6f | 212 | /* Loop unrolling. Process 4 taps at a time . */ |
emh203 | 0:3d9c67d97d6f | 213 | stageCnt = (numStages - 1u) >> 2u; |
emh203 | 0:3d9c67d97d6f | 214 | |
emh203 | 0:3d9c67d97d6f | 215 | /* Loop over the number of taps. Unroll by a factor of 4. |
emh203 | 0:3d9c67d97d6f | 216 | ** Repeat until we've computed numStages-3 coefficients. */ |
emh203 | 0:3d9c67d97d6f | 217 | |
emh203 | 0:3d9c67d97d6f | 218 | /* Process 2nd, 3rd, 4th and 5th taps ... here */ |
emh203 | 0:3d9c67d97d6f | 219 | while(stageCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 220 | { |
emh203 | 0:3d9c67d97d6f | 221 | /* Read g1(n-1), g3(n-1) .... from state */ |
emh203 | 0:3d9c67d97d6f | 222 | gcurr1 = *px; |
emh203 | 0:3d9c67d97d6f | 223 | |
emh203 | 0:3d9c67d97d6f | 224 | /* save g1(n) in state buffer */ |
emh203 | 0:3d9c67d97d6f | 225 | *px++ = gnext4; |
emh203 | 0:3d9c67d97d6f | 226 | |
emh203 | 0:3d9c67d97d6f | 227 | /* Process first sample for 2nd, 6th .. tap */ |
emh203 | 0:3d9c67d97d6f | 228 | /* Sample processing for K2, K6.... */ |
emh203 | 0:3d9c67d97d6f | 229 | /* f2(n) = f1(n) + K2 * g1(n-1) */ |
emh203 | 0:3d9c67d97d6f | 230 | fnext1 = fcurr1 + ((*pk) * gcurr1); |
emh203 | 0:3d9c67d97d6f | 231 | /* Process second sample for 2nd, 6th .. tap */ |
emh203 | 0:3d9c67d97d6f | 232 | /* for sample 2 processing */ |
emh203 | 0:3d9c67d97d6f | 233 | fnext2 = fcurr2 + ((*pk) * gnext1); |
emh203 | 0:3d9c67d97d6f | 234 | /* Process third sample for 2nd, 6th .. tap */ |
emh203 | 0:3d9c67d97d6f | 235 | fnext3 = fcurr3 + ((*pk) * gnext2); |
emh203 | 0:3d9c67d97d6f | 236 | /* Process fourth sample for 2nd, 6th .. tap */ |
emh203 | 0:3d9c67d97d6f | 237 | fnext4 = fcurr4 + ((*pk) * gnext3); |
emh203 | 0:3d9c67d97d6f | 238 | |
emh203 | 0:3d9c67d97d6f | 239 | /* g2(n) = f1(n) * K2 + g1(n-1) */ |
emh203 | 0:3d9c67d97d6f | 240 | /* Calculation of state values for next stage */ |
emh203 | 0:3d9c67d97d6f | 241 | gnext4 = (fcurr4 * (*pk)) + gnext3; |
emh203 | 0:3d9c67d97d6f | 242 | gnext3 = (fcurr3 * (*pk)) + gnext2; |
emh203 | 0:3d9c67d97d6f | 243 | gnext2 = (fcurr2 * (*pk)) + gnext1; |
emh203 | 0:3d9c67d97d6f | 244 | gnext1 = (fcurr1 * (*pk++)) + gcurr1; |
emh203 | 0:3d9c67d97d6f | 245 | |
emh203 | 0:3d9c67d97d6f | 246 | |
emh203 | 0:3d9c67d97d6f | 247 | /* Read g2(n-1), g4(n-1) .... from state */ |
emh203 | 0:3d9c67d97d6f | 248 | gcurr1 = *px; |
emh203 | 0:3d9c67d97d6f | 249 | |
emh203 | 0:3d9c67d97d6f | 250 | /* save g2(n) in state buffer */ |
emh203 | 0:3d9c67d97d6f | 251 | *px++ = gnext4; |
emh203 | 0:3d9c67d97d6f | 252 | |
emh203 | 0:3d9c67d97d6f | 253 | /* Sample processing for K3, K7.... */ |
emh203 | 0:3d9c67d97d6f | 254 | /* Process first sample for 3rd, 7th .. tap */ |
emh203 | 0:3d9c67d97d6f | 255 | /* f3(n) = f2(n) + K3 * g2(n-1) */ |
emh203 | 0:3d9c67d97d6f | 256 | fcurr1 = fnext1 + ((*pk) * gcurr1); |
emh203 | 0:3d9c67d97d6f | 257 | /* Process second sample for 3rd, 7th .. tap */ |
emh203 | 0:3d9c67d97d6f | 258 | fcurr2 = fnext2 + ((*pk) * gnext1); |
emh203 | 0:3d9c67d97d6f | 259 | /* Process third sample for 3rd, 7th .. tap */ |
emh203 | 0:3d9c67d97d6f | 260 | fcurr3 = fnext3 + ((*pk) * gnext2); |
emh203 | 0:3d9c67d97d6f | 261 | /* Process fourth sample for 3rd, 7th .. tap */ |
emh203 | 0:3d9c67d97d6f | 262 | fcurr4 = fnext4 + ((*pk) * gnext3); |
emh203 | 0:3d9c67d97d6f | 263 | |
emh203 | 0:3d9c67d97d6f | 264 | /* Calculation of state values for next stage */ |
emh203 | 0:3d9c67d97d6f | 265 | /* g3(n) = f2(n) * K3 + g2(n-1) */ |
emh203 | 0:3d9c67d97d6f | 266 | gnext4 = (fnext4 * (*pk)) + gnext3; |
emh203 | 0:3d9c67d97d6f | 267 | gnext3 = (fnext3 * (*pk)) + gnext2; |
emh203 | 0:3d9c67d97d6f | 268 | gnext2 = (fnext2 * (*pk)) + gnext1; |
emh203 | 0:3d9c67d97d6f | 269 | gnext1 = (fnext1 * (*pk++)) + gcurr1; |
emh203 | 0:3d9c67d97d6f | 270 | |
emh203 | 0:3d9c67d97d6f | 271 | |
emh203 | 0:3d9c67d97d6f | 272 | /* Read g1(n-1), g3(n-1) .... from state */ |
emh203 | 0:3d9c67d97d6f | 273 | gcurr1 = *px; |
emh203 | 0:3d9c67d97d6f | 274 | |
emh203 | 0:3d9c67d97d6f | 275 | /* save g3(n) in state buffer */ |
emh203 | 0:3d9c67d97d6f | 276 | *px++ = gnext4; |
emh203 | 0:3d9c67d97d6f | 277 | |
emh203 | 0:3d9c67d97d6f | 278 | /* Sample processing for K4, K8.... */ |
emh203 | 0:3d9c67d97d6f | 279 | /* Process first sample for 4th, 8th .. tap */ |
emh203 | 0:3d9c67d97d6f | 280 | /* f4(n) = f3(n) + K4 * g3(n-1) */ |
emh203 | 0:3d9c67d97d6f | 281 | fnext1 = fcurr1 + ((*pk) * gcurr1); |
emh203 | 0:3d9c67d97d6f | 282 | /* Process second sample for 4th, 8th .. tap */ |
emh203 | 0:3d9c67d97d6f | 283 | /* for sample 2 processing */ |
emh203 | 0:3d9c67d97d6f | 284 | fnext2 = fcurr2 + ((*pk) * gnext1); |
emh203 | 0:3d9c67d97d6f | 285 | /* Process third sample for 4th, 8th .. tap */ |
emh203 | 0:3d9c67d97d6f | 286 | fnext3 = fcurr3 + ((*pk) * gnext2); |
emh203 | 0:3d9c67d97d6f | 287 | /* Process fourth sample for 4th, 8th .. tap */ |
emh203 | 0:3d9c67d97d6f | 288 | fnext4 = fcurr4 + ((*pk) * gnext3); |
emh203 | 0:3d9c67d97d6f | 289 | |
emh203 | 0:3d9c67d97d6f | 290 | /* g4(n) = f3(n) * K4 + g3(n-1) */ |
emh203 | 0:3d9c67d97d6f | 291 | /* Calculation of state values for next stage */ |
emh203 | 0:3d9c67d97d6f | 292 | gnext4 = (fcurr4 * (*pk)) + gnext3; |
emh203 | 0:3d9c67d97d6f | 293 | gnext3 = (fcurr3 * (*pk)) + gnext2; |
emh203 | 0:3d9c67d97d6f | 294 | gnext2 = (fcurr2 * (*pk)) + gnext1; |
emh203 | 0:3d9c67d97d6f | 295 | gnext1 = (fcurr1 * (*pk++)) + gcurr1; |
emh203 | 0:3d9c67d97d6f | 296 | |
emh203 | 0:3d9c67d97d6f | 297 | /* Read g2(n-1), g4(n-1) .... from state */ |
emh203 | 0:3d9c67d97d6f | 298 | gcurr1 = *px; |
emh203 | 0:3d9c67d97d6f | 299 | |
emh203 | 0:3d9c67d97d6f | 300 | /* save g4(n) in state buffer */ |
emh203 | 0:3d9c67d97d6f | 301 | *px++ = gnext4; |
emh203 | 0:3d9c67d97d6f | 302 | |
emh203 | 0:3d9c67d97d6f | 303 | /* Sample processing for K5, K9.... */ |
emh203 | 0:3d9c67d97d6f | 304 | /* Process first sample for 5th, 9th .. tap */ |
emh203 | 0:3d9c67d97d6f | 305 | /* f5(n) = f4(n) + K5 * g4(n-1) */ |
emh203 | 0:3d9c67d97d6f | 306 | fcurr1 = fnext1 + ((*pk) * gcurr1); |
emh203 | 0:3d9c67d97d6f | 307 | /* Process second sample for 5th, 9th .. tap */ |
emh203 | 0:3d9c67d97d6f | 308 | fcurr2 = fnext2 + ((*pk) * gnext1); |
emh203 | 0:3d9c67d97d6f | 309 | /* Process third sample for 5th, 9th .. tap */ |
emh203 | 0:3d9c67d97d6f | 310 | fcurr3 = fnext3 + ((*pk) * gnext2); |
emh203 | 0:3d9c67d97d6f | 311 | /* Process fourth sample for 5th, 9th .. tap */ |
emh203 | 0:3d9c67d97d6f | 312 | fcurr4 = fnext4 + ((*pk) * gnext3); |
emh203 | 0:3d9c67d97d6f | 313 | |
emh203 | 0:3d9c67d97d6f | 314 | /* Calculation of state values for next stage */ |
emh203 | 0:3d9c67d97d6f | 315 | /* g5(n) = f4(n) * K5 + g4(n-1) */ |
emh203 | 0:3d9c67d97d6f | 316 | gnext4 = (fnext4 * (*pk)) + gnext3; |
emh203 | 0:3d9c67d97d6f | 317 | gnext3 = (fnext3 * (*pk)) + gnext2; |
emh203 | 0:3d9c67d97d6f | 318 | gnext2 = (fnext2 * (*pk)) + gnext1; |
emh203 | 0:3d9c67d97d6f | 319 | gnext1 = (fnext1 * (*pk++)) + gcurr1; |
emh203 | 0:3d9c67d97d6f | 320 | |
emh203 | 0:3d9c67d97d6f | 321 | stageCnt--; |
emh203 | 0:3d9c67d97d6f | 322 | } |
emh203 | 0:3d9c67d97d6f | 323 | |
emh203 | 0:3d9c67d97d6f | 324 | /* If the (filter length -1) is not a multiple of 4, compute the remaining filter taps */ |
emh203 | 0:3d9c67d97d6f | 325 | stageCnt = (numStages - 1u) % 0x4u; |
emh203 | 0:3d9c67d97d6f | 326 | |
emh203 | 0:3d9c67d97d6f | 327 | while(stageCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 328 | { |
emh203 | 0:3d9c67d97d6f | 329 | gcurr1 = *px; |
emh203 | 0:3d9c67d97d6f | 330 | |
emh203 | 0:3d9c67d97d6f | 331 | /* save g value in state buffer */ |
emh203 | 0:3d9c67d97d6f | 332 | *px++ = gnext4; |
emh203 | 0:3d9c67d97d6f | 333 | |
emh203 | 0:3d9c67d97d6f | 334 | /* Process four samples for last three taps here */ |
emh203 | 0:3d9c67d97d6f | 335 | fnext1 = fcurr1 + ((*pk) * gcurr1); |
emh203 | 0:3d9c67d97d6f | 336 | fnext2 = fcurr2 + ((*pk) * gnext1); |
emh203 | 0:3d9c67d97d6f | 337 | fnext3 = fcurr3 + ((*pk) * gnext2); |
emh203 | 0:3d9c67d97d6f | 338 | fnext4 = fcurr4 + ((*pk) * gnext3); |
emh203 | 0:3d9c67d97d6f | 339 | |
emh203 | 0:3d9c67d97d6f | 340 | /* g1(n) = f0(n) * K1 + g0(n-1) */ |
emh203 | 0:3d9c67d97d6f | 341 | gnext4 = (fcurr4 * (*pk)) + gnext3; |
emh203 | 0:3d9c67d97d6f | 342 | gnext3 = (fcurr3 * (*pk)) + gnext2; |
emh203 | 0:3d9c67d97d6f | 343 | gnext2 = (fcurr2 * (*pk)) + gnext1; |
emh203 | 0:3d9c67d97d6f | 344 | gnext1 = (fcurr1 * (*pk++)) + gcurr1; |
emh203 | 0:3d9c67d97d6f | 345 | |
emh203 | 0:3d9c67d97d6f | 346 | /* Update of f values for next coefficient set processing */ |
emh203 | 0:3d9c67d97d6f | 347 | fcurr1 = fnext1; |
emh203 | 0:3d9c67d97d6f | 348 | fcurr2 = fnext2; |
emh203 | 0:3d9c67d97d6f | 349 | fcurr3 = fnext3; |
emh203 | 0:3d9c67d97d6f | 350 | fcurr4 = fnext4; |
emh203 | 0:3d9c67d97d6f | 351 | |
emh203 | 0:3d9c67d97d6f | 352 | stageCnt--; |
emh203 | 0:3d9c67d97d6f | 353 | |
emh203 | 0:3d9c67d97d6f | 354 | } |
emh203 | 0:3d9c67d97d6f | 355 | |
emh203 | 0:3d9c67d97d6f | 356 | /* The results in the 4 accumulators, store in the destination buffer. */ |
emh203 | 0:3d9c67d97d6f | 357 | /* y(n) = fN(n) */ |
emh203 | 0:3d9c67d97d6f | 358 | *pDst++ = fcurr1; |
emh203 | 0:3d9c67d97d6f | 359 | *pDst++ = fcurr2; |
emh203 | 0:3d9c67d97d6f | 360 | *pDst++ = fcurr3; |
emh203 | 0:3d9c67d97d6f | 361 | *pDst++ = fcurr4; |
emh203 | 0:3d9c67d97d6f | 362 | |
emh203 | 0:3d9c67d97d6f | 363 | blkCnt--; |
emh203 | 0:3d9c67d97d6f | 364 | } |
emh203 | 0:3d9c67d97d6f | 365 | |
emh203 | 0:3d9c67d97d6f | 366 | /* If the blockSize is not a multiple of 4, compute any remaining output samples here. |
emh203 | 0:3d9c67d97d6f | 367 | ** No loop unrolling is used. */ |
emh203 | 0:3d9c67d97d6f | 368 | blkCnt = blockSize % 0x4u; |
emh203 | 0:3d9c67d97d6f | 369 | |
emh203 | 0:3d9c67d97d6f | 370 | while(blkCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 371 | { |
emh203 | 0:3d9c67d97d6f | 372 | /* f0(n) = x(n) */ |
emh203 | 0:3d9c67d97d6f | 373 | fcurr1 = *pSrc++; |
emh203 | 0:3d9c67d97d6f | 374 | |
emh203 | 0:3d9c67d97d6f | 375 | /* Initialize coeff pointer */ |
emh203 | 0:3d9c67d97d6f | 376 | pk = (pCoeffs); |
emh203 | 0:3d9c67d97d6f | 377 | |
emh203 | 0:3d9c67d97d6f | 378 | /* Initialize state pointer */ |
emh203 | 0:3d9c67d97d6f | 379 | px = pState; |
emh203 | 0:3d9c67d97d6f | 380 | |
emh203 | 0:3d9c67d97d6f | 381 | /* read g2(n) from state buffer */ |
emh203 | 0:3d9c67d97d6f | 382 | gcurr1 = *px; |
emh203 | 0:3d9c67d97d6f | 383 | |
emh203 | 0:3d9c67d97d6f | 384 | /* for sample 1 processing */ |
emh203 | 0:3d9c67d97d6f | 385 | /* f1(n) = f0(n) + K1 * g0(n-1) */ |
emh203 | 0:3d9c67d97d6f | 386 | fnext1 = fcurr1 + ((*pk) * gcurr1); |
emh203 | 0:3d9c67d97d6f | 387 | /* g1(n) = f0(n) * K1 + g0(n-1) */ |
emh203 | 0:3d9c67d97d6f | 388 | gnext1 = (fcurr1 * (*pk++)) + gcurr1; |
emh203 | 0:3d9c67d97d6f | 389 | |
emh203 | 0:3d9c67d97d6f | 390 | /* save g1(n) in state buffer */ |
emh203 | 0:3d9c67d97d6f | 391 | *px++ = fcurr1; |
emh203 | 0:3d9c67d97d6f | 392 | |
emh203 | 0:3d9c67d97d6f | 393 | /* f1(n) is saved in fcurr1 |
emh203 | 0:3d9c67d97d6f | 394 | for next stage processing */ |
emh203 | 0:3d9c67d97d6f | 395 | fcurr1 = fnext1; |
emh203 | 0:3d9c67d97d6f | 396 | |
emh203 | 0:3d9c67d97d6f | 397 | stageCnt = (numStages - 1u); |
emh203 | 0:3d9c67d97d6f | 398 | |
emh203 | 0:3d9c67d97d6f | 399 | /* stage loop */ |
emh203 | 0:3d9c67d97d6f | 400 | while(stageCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 401 | { |
emh203 | 0:3d9c67d97d6f | 402 | /* read g2(n) from state buffer */ |
emh203 | 0:3d9c67d97d6f | 403 | gcurr1 = *px; |
emh203 | 0:3d9c67d97d6f | 404 | |
emh203 | 0:3d9c67d97d6f | 405 | /* save g1(n) in state buffer */ |
emh203 | 0:3d9c67d97d6f | 406 | *px++ = gnext1; |
emh203 | 0:3d9c67d97d6f | 407 | |
emh203 | 0:3d9c67d97d6f | 408 | /* Sample processing for K2, K3.... */ |
emh203 | 0:3d9c67d97d6f | 409 | /* f2(n) = f1(n) + K2 * g1(n-1) */ |
emh203 | 0:3d9c67d97d6f | 410 | fnext1 = fcurr1 + ((*pk) * gcurr1); |
emh203 | 0:3d9c67d97d6f | 411 | /* g2(n) = f1(n) * K2 + g1(n-1) */ |
emh203 | 0:3d9c67d97d6f | 412 | gnext1 = (fcurr1 * (*pk++)) + gcurr1; |
emh203 | 0:3d9c67d97d6f | 413 | |
emh203 | 0:3d9c67d97d6f | 414 | /* f1(n) is saved in fcurr1 |
emh203 | 0:3d9c67d97d6f | 415 | for next stage processing */ |
emh203 | 0:3d9c67d97d6f | 416 | fcurr1 = fnext1; |
emh203 | 0:3d9c67d97d6f | 417 | |
emh203 | 0:3d9c67d97d6f | 418 | stageCnt--; |
emh203 | 0:3d9c67d97d6f | 419 | |
emh203 | 0:3d9c67d97d6f | 420 | } |
emh203 | 0:3d9c67d97d6f | 421 | |
emh203 | 0:3d9c67d97d6f | 422 | /* y(n) = fN(n) */ |
emh203 | 0:3d9c67d97d6f | 423 | *pDst++ = fcurr1; |
emh203 | 0:3d9c67d97d6f | 424 | |
emh203 | 0:3d9c67d97d6f | 425 | blkCnt--; |
emh203 | 0:3d9c67d97d6f | 426 | |
emh203 | 0:3d9c67d97d6f | 427 | } |
emh203 | 0:3d9c67d97d6f | 428 | |
emh203 | 0:3d9c67d97d6f | 429 | #else |
emh203 | 0:3d9c67d97d6f | 430 | |
emh203 | 0:3d9c67d97d6f | 431 | /* Run the below code for Cortex-M0 */ |
emh203 | 0:3d9c67d97d6f | 432 | |
emh203 | 0:3d9c67d97d6f | 433 | float32_t fcurr, fnext, gcurr, gnext; /* temporary variables */ |
emh203 | 0:3d9c67d97d6f | 434 | uint32_t numStages = S->numStages; /* Length of the filter */ |
emh203 | 0:3d9c67d97d6f | 435 | uint32_t blkCnt, stageCnt; /* temporary variables for counts */ |
emh203 | 0:3d9c67d97d6f | 436 | |
emh203 | 0:3d9c67d97d6f | 437 | pState = &S->pState[0]; |
emh203 | 0:3d9c67d97d6f | 438 | |
emh203 | 0:3d9c67d97d6f | 439 | blkCnt = blockSize; |
emh203 | 0:3d9c67d97d6f | 440 | |
emh203 | 0:3d9c67d97d6f | 441 | while(blkCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 442 | { |
emh203 | 0:3d9c67d97d6f | 443 | /* f0(n) = x(n) */ |
emh203 | 0:3d9c67d97d6f | 444 | fcurr = *pSrc++; |
emh203 | 0:3d9c67d97d6f | 445 | |
emh203 | 0:3d9c67d97d6f | 446 | /* Initialize coeff pointer */ |
emh203 | 0:3d9c67d97d6f | 447 | pk = pCoeffs; |
emh203 | 0:3d9c67d97d6f | 448 | |
emh203 | 0:3d9c67d97d6f | 449 | /* Initialize state pointer */ |
emh203 | 0:3d9c67d97d6f | 450 | px = pState; |
emh203 | 0:3d9c67d97d6f | 451 | |
emh203 | 0:3d9c67d97d6f | 452 | /* read g0(n-1) from state buffer */ |
emh203 | 0:3d9c67d97d6f | 453 | gcurr = *px; |
emh203 | 0:3d9c67d97d6f | 454 | |
emh203 | 0:3d9c67d97d6f | 455 | /* for sample 1 processing */ |
emh203 | 0:3d9c67d97d6f | 456 | /* f1(n) = f0(n) + K1 * g0(n-1) */ |
emh203 | 0:3d9c67d97d6f | 457 | fnext = fcurr + ((*pk) * gcurr); |
emh203 | 0:3d9c67d97d6f | 458 | /* g1(n) = f0(n) * K1 + g0(n-1) */ |
emh203 | 0:3d9c67d97d6f | 459 | gnext = (fcurr * (*pk++)) + gcurr; |
emh203 | 0:3d9c67d97d6f | 460 | |
emh203 | 0:3d9c67d97d6f | 461 | /* save f0(n) in state buffer */ |
emh203 | 0:3d9c67d97d6f | 462 | *px++ = fcurr; |
emh203 | 0:3d9c67d97d6f | 463 | |
emh203 | 0:3d9c67d97d6f | 464 | /* f1(n) is saved in fcurr |
emh203 | 0:3d9c67d97d6f | 465 | for next stage processing */ |
emh203 | 0:3d9c67d97d6f | 466 | fcurr = fnext; |
emh203 | 0:3d9c67d97d6f | 467 | |
emh203 | 0:3d9c67d97d6f | 468 | stageCnt = (numStages - 1u); |
emh203 | 0:3d9c67d97d6f | 469 | |
emh203 | 0:3d9c67d97d6f | 470 | /* stage loop */ |
emh203 | 0:3d9c67d97d6f | 471 | while(stageCnt > 0u) |
emh203 | 0:3d9c67d97d6f | 472 | { |
emh203 | 0:3d9c67d97d6f | 473 | /* read g2(n) from state buffer */ |
emh203 | 0:3d9c67d97d6f | 474 | gcurr = *px; |
emh203 | 0:3d9c67d97d6f | 475 | |
emh203 | 0:3d9c67d97d6f | 476 | /* save g1(n) in state buffer */ |
emh203 | 0:3d9c67d97d6f | 477 | *px++ = gnext; |
emh203 | 0:3d9c67d97d6f | 478 | |
emh203 | 0:3d9c67d97d6f | 479 | /* Sample processing for K2, K3.... */ |
emh203 | 0:3d9c67d97d6f | 480 | /* f2(n) = f1(n) + K2 * g1(n-1) */ |
emh203 | 0:3d9c67d97d6f | 481 | fnext = fcurr + ((*pk) * gcurr); |
emh203 | 0:3d9c67d97d6f | 482 | /* g2(n) = f1(n) * K2 + g1(n-1) */ |
emh203 | 0:3d9c67d97d6f | 483 | gnext = (fcurr * (*pk++)) + gcurr; |
emh203 | 0:3d9c67d97d6f | 484 | |
emh203 | 0:3d9c67d97d6f | 485 | /* f1(n) is saved in fcurr1 |
emh203 | 0:3d9c67d97d6f | 486 | for next stage processing */ |
emh203 | 0:3d9c67d97d6f | 487 | fcurr = fnext; |
emh203 | 0:3d9c67d97d6f | 488 | |
emh203 | 0:3d9c67d97d6f | 489 | stageCnt--; |
emh203 | 0:3d9c67d97d6f | 490 | |
emh203 | 0:3d9c67d97d6f | 491 | } |
emh203 | 0:3d9c67d97d6f | 492 | |
emh203 | 0:3d9c67d97d6f | 493 | /* y(n) = fN(n) */ |
emh203 | 0:3d9c67d97d6f | 494 | *pDst++ = fcurr; |
emh203 | 0:3d9c67d97d6f | 495 | |
emh203 | 0:3d9c67d97d6f | 496 | blkCnt--; |
emh203 | 0:3d9c67d97d6f | 497 | |
emh203 | 0:3d9c67d97d6f | 498 | } |
emh203 | 0:3d9c67d97d6f | 499 | |
emh203 | 0:3d9c67d97d6f | 500 | #endif /* #ifndef ARM_MATH_CM0_FAMILY */ |
emh203 | 0:3d9c67d97d6f | 501 | |
emh203 | 0:3d9c67d97d6f | 502 | } |
emh203 | 0:3d9c67d97d6f | 503 | |
emh203 | 0:3d9c67d97d6f | 504 | /** |
emh203 | 0:3d9c67d97d6f | 505 | * @} end of FIR_Lattice group |
emh203 | 0:3d9c67d97d6f | 506 | */ |