tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/ComplexMathFunctions/arm_cmplx_dot_prod_q31.c

   1 /* ----------------------------------------------------------------------
   2 * Copyright (C) 2010-2013 ARM Limited. All rights reserved.
   3 *
   4 * $Date:        17. January 2013
   5 * $Revision:    V1.4.1
   6 *
   7 * Project:          CMSIS DSP Library
   8 * Title:                arm_cmplx_dot_prod_q31.c
   9 *
  10 * Description:  Q31 complex dot product
  11 *
  12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
  13 *
  14 * Redistribution and use in source and binary forms, with or without
  15 * modification, are permitted provided that the following conditions
  16 * are met:
  17 *   - Redistributions of source code must retain the above copyright
  18 *     notice, this list of conditions and the following disclaimer.
  19 *   - Redistributions in binary form must reproduce the above copyright
  20 *     notice, this list of conditions and the following disclaimer in
  21 *     the documentation and/or other materials provided with the
  22 *     distribution.
  23 *   - Neither the name of ARM LIMITED nor the names of its contributors
  24 *     may be used to endorse or promote products derived from this
  25 *     software without specific prior written permission.
  26 *
  27 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
  28 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  29 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
  30 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
  31 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
  32 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
  33 * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
  34 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
  35 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  36 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
  37 * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
  38 * POSSIBILITY OF SUCH DAMAGE.
  39 * -------------------------------------------------------------------- */
  40
  41 #include "arm_math.h"
  42
  43 /**
  44  * @ingroup groupCmplxMath
  45  */
  46
  47 /**
  48  * @addtogroup cmplx_dot_prod
  49  * @{
  50  */
  51
  52 /**
  53  * @brief  Q31 complex dot product
  54  * @param  *pSrcA points to the first input vector
  55  * @param  *pSrcB points to the second input vector
  56  * @param  numSamples number of complex samples in each vector
  57  * @param  *realResult real part of the result returned here
  58  * @param  *imagResult imaginary part of the result returned here
  59  * @return none.
  60  *
  61  * <b>Scaling and Overflow Behavior:</b>
  62  * \par
  63  * The function is implemented using an internal 64-bit accumulator.
  64  * The intermediate 1.31 by 1.31 multiplications are performed with 64-bit precision and then shifted to 16.48 format.
  65  * The internal real and imaginary accumulators are in 16.48 format and provide 15 guard bits.
  66  * Additions are nonsaturating and no overflow will occur as long as <code>numSamples</code> is less than 32768.
  67  * The return results <code>realResult</code> and <code>imagResult</code> are in 16.48 format.
  68  * Input down scaling is not required.
  69  */
  70
  71 void arm_cmplx_dot_prod_q31(
  72   q31_t * pSrcA,
  73   q31_t * pSrcB,
  74   uint32_t numSamples,
  75   q63_t * realResult,
  76   q63_t * imagResult)
  77 {
  78   q63_t real_sum = 0, imag_sum = 0;              /* Temporary result storage */
  79
  80 #ifndef ARM_MATH_CM0_FAMILY
  81
  82   /* Run the below code for Cortex-M4 and Cortex-M3 */
  83   uint32_t blkCnt;                               /* loop counter */
  84
  85
  86   /*loop Unrolling */
  87   blkCnt = numSamples >> 2u;
  88
  89   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
  90    ** a second loop below computes the remaining 1 to 3 samples. */
  91   while(blkCnt > 0u)
  92   {
  93     /* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */
  94     /* Convert real data in 2.62 to 16.48 by 14 right shifts */
  95     real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
  96     /* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */
  97     /* Convert imag data in 2.62 to 16.48 by 14 right shifts */
  98     imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
  99
 100     real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
 101     imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
 102
 103     real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
 104     imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
 105
 106     real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
 107     imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
 108
 109
 110     /* Decrement the loop counter */
 111     blkCnt--;
 112   }
 113
 114   /* If the numSamples  is not a multiple of 4, compute any remaining output samples here.
 115    ** No loop unrolling is used. */
 116   blkCnt = numSamples % 0x4u;
 117
 118   while(blkCnt > 0u)
 119   {
 120     /* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */
 121     real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
 122     /* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */
 123     imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
 124
 125     /* Decrement the loop counter */
 126     blkCnt--;
 127   }
 128
 129 #else
 130
 131   /* Run the below code for Cortex-M0 */
 132
 133   while(numSamples > 0u)
 134   {
 135     /* outReal = realA[0]* realB[0] + realA[2]* realB[2] + realA[4]* realB[4] + .....+ realA[numSamples-2]* realB[numSamples-2] */
 136     real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
 137     /* outImag = imagA[1]* imagB[1] + imagA[3]* imagB[3] + imagA[5]* imagB[5] + .....+ imagA[numSamples-1]* imagB[numSamples-1] */
 138     imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
 139
 140     /* Decrement the loop counter */
 141     numSamples--;
 142   }
 143
 144 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
 145
 146   /* Store the real and imaginary results in 16.48 format  */
 147   *realResult = real_sum;
 148   *imagResult = imag_sum;
 149 }
 150
 151 /**
 152  * @} end of cmplx_dot_prod group
 153  */