arm_lms_f32.c

Go to the documentation of this file.

/* ----------------------------------------------------------------------    
* Copyright (C) 2010-2014 ARM Limited. All rights reserved.    
*    
* $Date:        19. March 2015
* $Revision:    V.1.4.5
*    
* Project:      CMSIS DSP Library    
* Title:        arm_lms_f32.c    
*    
* Description:  Processing function for the floating-point LMS filter.    
*    
* 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_lms_f32(
  const arm_lms_instance_f32 * S,
  float32_t * pSrc,
  float32_t * pRef,
  float32_t * pOut,
  float32_t * pErr,
  uint32_t blockSize)
{
  float32_t *pState = S->pState;                 /* State pointer */
  float32_t *pCoeffs = S->pCoeffs;               /* Coefficient pointer */
  float32_t *pStateCurnt;                        /* Points to the current sample of the state */
  float32_t *px, *pb;                            /* Temporary pointers for state and coefficient buffers */
  float32_t mu = S->mu;                          /* Adaptive factor */
  uint32_t numTaps = S->numTaps;                 /* Number of filter coefficients in the filter */
  uint32_t tapCnt, blkCnt;                       /* Loop counters */
  float32_t sum, e, d;                           /* accumulator, error, reference data sample */
  float32_t w = 0.0f;                            /* weight factor */

  e = 0.0f;
  d = 0.0f;

  /* S->pState points to state array which contains previous frame (numTaps - 1) samples */
  /* pStateCurnt points to the location where the new input data should be written */
  pStateCurnt = &(S->pState[(numTaps - 1u)]);

  blkCnt = blockSize;


#ifndef ARM_MATH_CM0_FAMILY

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

  while(blkCnt > 0u)
  {
    /* Copy the new input sample into the state buffer */
    *pStateCurnt++ = *pSrc++;

    /* Initialize pState pointer */
    px = pState;

    /* Initialize coeff pointer */
    pb = (pCoeffs);

    /* Set the accumulator to zero */
    sum = 0.0f;

    /* Loop unrolling.  Process 4 taps at a time. */
    tapCnt = numTaps >> 2;

    while(tapCnt > 0u)
    {
      /* Perform the multiply-accumulate */
      sum += (*px++) * (*pb++);
      sum += (*px++) * (*pb++);
      sum += (*px++) * (*pb++);
      sum += (*px++) * (*pb++);

      /* Decrement the loop counter */
      tapCnt--;
    }

    /* If the filter length is not a multiple of 4, compute the remaining filter taps */
    tapCnt = numTaps % 0x4u;

    while(tapCnt > 0u)
    {
      /* Perform the multiply-accumulate */
      sum += (*px++) * (*pb++);

      /* Decrement the loop counter */
      tapCnt--;
    }

    /* The result in the accumulator, store in the destination buffer. */
    *pOut++ = sum;

    /* Compute and store error */
    d = (float32_t) (*pRef++);
    e = d - sum;
    *pErr++ = e;

    /* Calculation of Weighting factor for the updating filter coefficients */
    w = e * mu;

    /* Initialize pState pointer */
    px = pState;

    /* Initialize coeff pointer */
    pb = (pCoeffs);

    /* Loop unrolling.  Process 4 taps at a time. */
    tapCnt = numTaps >> 2;

    /* Update filter coefficients */
    while(tapCnt > 0u)
    {
      /* Perform the multiply-accumulate */
      *pb = *pb + (w * (*px++));
      pb++;

      *pb = *pb + (w * (*px++));
      pb++;

      *pb = *pb + (w * (*px++));
      pb++;

      *pb = *pb + (w * (*px++));
      pb++;

      /* Decrement the loop counter */
      tapCnt--;
    }

    /* If the filter length is not a multiple of 4, compute the remaining filter taps */
    tapCnt = numTaps % 0x4u;

    while(tapCnt > 0u)
    {
      /* Perform the multiply-accumulate */
      *pb = *pb + (w * (*px++));
      pb++;

      /* Decrement the loop counter */
      tapCnt--;
    }

    /* Advance state pointer by 1 for the next sample */
    pState = pState + 1;

    /* Decrement the loop counter */
    blkCnt--;
  }


  /* Processing is complete. Now copy the last numTaps - 1 samples to the    
     satrt of the state buffer. This prepares the state buffer for the    
     next function call. */

  /* Points to the start of the pState buffer */
  pStateCurnt = S->pState;

  /* Loop unrolling for (numTaps - 1u) samples copy */
  tapCnt = (numTaps - 1u) >> 2u;

  /* copy data */
  while(tapCnt > 0u)
  {
    *pStateCurnt++ = *pState++;
    *pStateCurnt++ = *pState++;
    *pStateCurnt++ = *pState++;
    *pStateCurnt++ = *pState++;

    /* Decrement the loop counter */
    tapCnt--;
  }

  /* Calculate remaining number of copies */
  tapCnt = (numTaps - 1u) % 0x4u;

  /* Copy the remaining q31_t data */
  while(tapCnt > 0u)
  {
    *pStateCurnt++ = *pState++;

    /* Decrement the loop counter */
    tapCnt--;
  }

#else

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

  while(blkCnt > 0u)
  {
    /* Copy the new input sample into the state buffer */
    *pStateCurnt++ = *pSrc++;

    /* Initialize pState pointer */
    px = pState;

    /* Initialize pCoeffs pointer */
    pb = pCoeffs;

    /* Set the accumulator to zero */
    sum = 0.0f;

    /* Loop over numTaps number of values */
    tapCnt = numTaps;

    while(tapCnt > 0u)
    {
      /* Perform the multiply-accumulate */
      sum += (*px++) * (*pb++);

      /* Decrement the loop counter */
      tapCnt--;
    }

    /* The result is stored in the destination buffer. */
    *pOut++ = sum;

    /* Compute and store error */
    d = (float32_t) (*pRef++);
    e = d - sum;
    *pErr++ = e;

    /* Weighting factor for the LMS version */
    w = e * mu;

    /* Initialize pState pointer */
    px = pState;

    /* Initialize pCoeffs pointer */
    pb = pCoeffs;

    /* Loop over numTaps number of values */
    tapCnt = numTaps;

    while(tapCnt > 0u)
    {
      /* Perform the multiply-accumulate */
      *pb = *pb + (w * (*px++));
      pb++;

      /* Decrement the loop counter */
      tapCnt--;
    }

    /* Advance state pointer by 1 for the next sample */
    pState = pState + 1;

    /* Decrement the loop counter */
    blkCnt--;
  }


  /* Processing is complete. Now copy the last numTaps - 1 samples to the        
   * start of the state buffer. This prepares the state buffer for the        
   * next function call. */

  /* Points to the start of the pState buffer */
  pStateCurnt = S->pState;

  /*  Copy (numTaps - 1u) samples  */
  tapCnt = (numTaps - 1u);

  /* Copy the data */
  while(tapCnt > 0u)
  {
    *pStateCurnt++ = *pState++;

    /* Decrement the loop counter */
    tapCnt--;
  }

#endif /*   #ifndef ARM_MATH_CM0_FAMILY */

}