arm_biquad_cascade_df1_32x64_q31.c

Go to the documentation of this file.

/* ----------------------------------------------------------------------    
* Copyright (C) 2010-2014 ARM Limited. All rights reserved.    
*    
* $Date:        19. October 2015
* $Revision:    V.1.4.5 a
*    
* Project:      CMSIS DSP Library    
* Title:        arm_biquad_cascade_df1_32x64_q31.c    
*    
* Description:  High precision Q31 Biquad cascade filter processing function    
*    
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
*  
* Redistribution and use in source and binary forms, with or without 
* modification, are permitted provided that the following conditions
* are met:
*   - Redistributions of source code must retain the above copyright
*     notice, this list of conditions and the following disclaimer.
*   - Redistributions in binary form must reproduce the above copyright
*     notice, this list of conditions and the following disclaimer in
*     the documentation and/or other materials provided with the 
*     distribution.
*   - Neither the name of ARM LIMITED nor the names of its contributors
*     may be used to endorse or promote products derived from this
*     software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE 
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
* BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
* ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.   
* -------------------------------------------------------------------- */

#include "arm_math.h"

void arm_biquad_cas_df1_32x64_q31(
  const arm_biquad_cas_df1_32x64_ins_q31 * S,
  q31_t * pSrc,
  q31_t * pDst,
  uint32_t blockSize)
{
  q31_t *pIn = pSrc;                             /*  input pointer initialization  */
  q31_t *pOut = pDst;                            /*  output pointer initialization */
  q63_t *pState = S->pState;                     /*  state pointer initialization  */
  q31_t *pCoeffs = S->pCoeffs;                   /*  coeff pointer initialization  */
  q63_t acc;                                     /*  accumulator                   */
  q31_t Xn1, Xn2;                                /*  Input Filter state variables        */
  q63_t Yn1, Yn2;                                /*  Output Filter state variables        */
  q31_t b0, b1, b2, a1, a2;                      /*  Filter coefficients           */
  q31_t Xn;                                      /*  temporary input               */
  int32_t shift = (int32_t) S->postShift + 1;    /*  Shift to be applied to the output */
  uint32_t sample, stage = S->numStages;         /*  loop counters                     */
  q31_t acc_l, acc_h;                            /*  temporary output               */
  uint32_t uShift = ((uint32_t) S->postShift + 1u);
  uint32_t lShift = 32u - uShift;                /*  Shift to be applied to the output */


#ifndef ARM_MATH_CM0_FAMILY

  /* Run the below code for Cortex-M4 and Cortex-M3 */

  do
  {
    /* Reading the coefficients */
    b0 = *pCoeffs++;
    b1 = *pCoeffs++;
    b2 = *pCoeffs++;
    a1 = *pCoeffs++;
    a2 = *pCoeffs++;

    /* Reading the state values */
    Xn1 = (q31_t) (pState[0]);
    Xn2 = (q31_t) (pState[1]);
    Yn1 = pState[2];
    Yn2 = pState[3];

    /* Apply loop unrolling and compute 4 output values simultaneously. */
    /* The variable acc hold output value that is being computed and    
     * stored in the destination buffer    
     * acc =  b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2]    
     */

    sample = blockSize >> 2u;

    /* First part of the processing with loop unrolling. Compute 4 outputs at a time.    
     ** a second loop below computes the remaining 1 to 3 samples. */
    while(sample > 0u)
    {
      /* Read the input */
      Xn = *pIn++;

      /* acc =  b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */

      /* acc =  b0 * x[n] */
      acc = (q63_t) Xn *b0;

      /* acc +=  b1 * x[n-1] */
      acc += (q63_t) Xn1 *b1;

      /* acc +=  b[2] * x[n-2] */
      acc += (q63_t) Xn2 *b2;

      /* acc +=  a1 * y[n-1] */
      acc += mult32x64(Yn1, a1);

      /* acc +=  a2 * y[n-2] */
      acc += mult32x64(Yn2, a2);

      /* The result is converted to 1.63 , Yn2 variable is reused */
      Yn2 = acc << shift;

      /* Calc lower part of acc */
      acc_l = acc & 0xffffffff;

      /* Calc upper part of acc */
      acc_h = (acc >> 32) & 0xffffffff;

      /* Apply shift for lower part of acc and upper part of acc */
      acc_h = (uint32_t) acc_l >> lShift | acc_h << uShift;

      /* Store the output in the destination buffer in 1.31 format. */
      *pOut = acc_h;

      /* Read the second input into Xn2, to reuse the value */
      Xn2 = *pIn++;

      /* acc =  b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */

      /* acc +=  b1 * x[n-1] */
      acc = (q63_t) Xn *b1;

      /* acc =  b0 * x[n] */
      acc += (q63_t) Xn2 *b0;

      /* acc +=  b[2] * x[n-2] */
      acc += (q63_t) Xn1 *b2;

      /* acc +=  a1 * y[n-1] */
      acc += mult32x64(Yn2, a1);

      /* acc +=  a2 * y[n-2] */
      acc += mult32x64(Yn1, a2);

      /* The result is converted to 1.63, Yn1 variable is reused */
      Yn1 = acc << shift;

      /* Calc lower part of acc */
      acc_l = acc & 0xffffffff;

      /* Calc upper part of acc */
      acc_h = (acc >> 32) & 0xffffffff;

      /* Apply shift for lower part of acc and upper part of acc */
      acc_h = (uint32_t) acc_l >> lShift | acc_h << uShift;

      /* Read the third input into Xn1, to reuse the value */
      Xn1 = *pIn++;

      /* The result is converted to 1.31 */
      /* Store the output in the destination buffer. */
      *(pOut + 1u) = acc_h;

      /* acc =  b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */

      /* acc =  b0 * x[n] */
      acc = (q63_t) Xn1 *b0;

      /* acc +=  b1 * x[n-1] */
      acc += (q63_t) Xn2 *b1;

      /* acc +=  b[2] * x[n-2] */
      acc += (q63_t) Xn *b2;

      /* acc +=  a1 * y[n-1] */
      acc += mult32x64(Yn1, a1);

      /* acc +=  a2 * y[n-2] */
      acc += mult32x64(Yn2, a2);

      /* The result is converted to 1.63, Yn2 variable is reused  */
      Yn2 = acc << shift;

      /* Calc lower part of acc */
      acc_l = acc & 0xffffffff;

      /* Calc upper part of acc */
      acc_h = (acc >> 32) & 0xffffffff;

      /* Apply shift for lower part of acc and upper part of acc */
      acc_h = (uint32_t) acc_l >> lShift | acc_h << uShift;

      /* Store the output in the destination buffer in 1.31 format. */
      *(pOut + 2u) = acc_h;

      /* Read the fourth input into Xn, to reuse the value */
      Xn = *pIn++;

      /* acc =  b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */
      /* acc =  b0 * x[n] */
      acc = (q63_t) Xn *b0;

      /* acc +=  b1 * x[n-1] */
      acc += (q63_t) Xn1 *b1;

      /* acc +=  b[2] * x[n-2] */
      acc += (q63_t) Xn2 *b2;

      /* acc +=  a1 * y[n-1] */
      acc += mult32x64(Yn2, a1);

      /* acc +=  a2 * y[n-2] */
      acc += mult32x64(Yn1, a2);

      /* The result is converted to 1.63, Yn1 variable is reused  */
      Yn1 = acc << shift;

      /* Calc lower part of acc */
      acc_l = acc & 0xffffffff;

      /* Calc upper part of acc */
      acc_h = (acc >> 32) & 0xffffffff;

      /* Apply shift for lower part of acc and upper part of acc */
      acc_h = (uint32_t) acc_l >> lShift | acc_h << uShift;

      /* Store the output in the destination buffer in 1.31 format. */
      *(pOut + 3u) = acc_h;

      /* Every time after the output is computed state should be updated. */
      /* The states should be updated as:  */
      /* Xn2 = Xn1    */
      /* Xn1 = Xn     */
      /* Yn2 = Yn1    */
      /* Yn1 = acc    */
      Xn2 = Xn1;
      Xn1 = Xn;

      /* update output pointer */
      pOut += 4u;

      /* decrement the loop counter */
      sample--;
    }

    /* If the blockSize is not a multiple of 4, compute any remaining output samples here.    
     ** No loop unrolling is used. */
    sample = (blockSize & 0x3u);

    while(sample > 0u)
    {
      /* Read the input */
      Xn = *pIn++;

      /* acc =  b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */

      /* acc =  b0 * x[n] */
      acc = (q63_t) Xn *b0;
      /* acc +=  b1 * x[n-1] */
      acc += (q63_t) Xn1 *b1;
      /* acc +=  b[2] * x[n-2] */
      acc += (q63_t) Xn2 *b2;
      /* acc +=  a1 * y[n-1] */
      acc += mult32x64(Yn1, a1);
      /* acc +=  a2 * y[n-2] */
      acc += mult32x64(Yn2, a2);

      /* Every time after the output is computed state should be updated. */
      /* The states should be updated as:  */
      /* Xn2 = Xn1    */
      /* Xn1 = Xn     */
      /* Yn2 = Yn1    */
      /* Yn1 = acc    */
      Xn2 = Xn1;
      Xn1 = Xn;
      Yn2 = Yn1;
      /* The result is converted to 1.63, Yn1 variable is reused  */
      Yn1 = acc << shift;

      /* Calc lower part of acc */
      acc_l = acc & 0xffffffff;

      /* Calc upper part of acc */
      acc_h = (acc >> 32) & 0xffffffff;

      /* Apply shift for lower part of acc and upper part of acc */
      acc_h = (uint32_t) acc_l >> lShift | acc_h << uShift;

      /* Store the output in the destination buffer in 1.31 format. */
      *pOut++ = acc_h;
      /* Yn1 = acc << shift; */

      /* Store the output in the destination buffer in 1.31 format. */
/*      *pOut++ = (q31_t) (acc >> (32 - shift));  */

      /* decrement the loop counter */
      sample--;
    }

    /*  The first stage output is given as input to the second stage. */
    pIn = pDst;

    /* Reset to destination buffer working pointer */
    pOut = pDst;

    /*  Store the updated state variables back into the pState array */
    /*  Store the updated state variables back into the pState array */
    *pState++ = (q63_t) Xn1;
    *pState++ = (q63_t) Xn2;
    *pState++ = Yn1;
    *pState++ = Yn2;

  } while(--stage);

#else

  /* Run the below code for Cortex-M0 */

  do
  {
    /* Reading the coefficients */
    b0 = *pCoeffs++;
    b1 = *pCoeffs++;
    b2 = *pCoeffs++;
    a1 = *pCoeffs++;
    a2 = *pCoeffs++;

    /* Reading the state values */
    Xn1 = pState[0];
    Xn2 = pState[1];
    Yn1 = pState[2];
    Yn2 = pState[3];

    /* The variable acc hold output value that is being computed and        
     * stored in the destination buffer            
     * acc =  b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2]            
     */

    sample = blockSize;

    while(sample > 0u)
    {
      /* Read the input */
      Xn = *pIn++;

      /* acc =  b0 * x[n] + b1 * x[n-1] + b2 * x[n-2] + a1 * y[n-1] + a2 * y[n-2] */
      /* acc =  b0 * x[n] */
      acc = (q63_t) Xn *b0;
      /* acc +=  b1 * x[n-1] */
      acc += (q63_t) Xn1 *b1;
      /* acc +=  b[2] * x[n-2] */
      acc += (q63_t) Xn2 *b2;
      /* acc +=  a1 * y[n-1] */
      acc += mult32x64(Yn1, a1);
      /* acc +=  a2 * y[n-2] */
      acc += mult32x64(Yn2, a2);

      /* Every time after the output is computed state should be updated. */
      /* The states should be updated as:  */
      /* Xn2 = Xn1    */
      /* Xn1 = Xn     */
      /* Yn2 = Yn1    */
      /* Yn1 = acc    */
      Xn2 = Xn1;
      Xn1 = Xn;
      Yn2 = Yn1;

      /* The result is converted to 1.63, Yn1 variable is reused  */
      Yn1 = acc << shift;

      /* Calc lower part of acc */
      acc_l = acc & 0xffffffff;

      /* Calc upper part of acc */
      acc_h = (acc >> 32) & 0xffffffff;

      /* Apply shift for lower part of acc and upper part of acc */
      acc_h = (uint32_t) acc_l >> lShift | acc_h << uShift;

      /* Store the output in the destination buffer in 1.31 format. */
      *pOut++ = acc_h;

      /* Yn1 = acc << shift; */

      /* Store the output in the destination buffer in 1.31 format. */
      /* *pOut++ = (q31_t) (acc >> (32 - shift)); */

      /* decrement the loop counter */
      sample--;
    }

    /*  The first stage output is given as input to the second stage. */
    pIn = pDst;

    /* Reset to destination buffer working pointer */
    pOut = pDst;

    /*  Store the updated state variables back into the pState array */
    *pState++ = (q63_t) Xn1;
    *pState++ = (q63_t) Xn2;
    *pState++ = Yn1;
    *pState++ = Yn2;

  } while(--stage);

#endif /*    #ifndef ARM_MATH_CM0_FAMILY     */
}