lib/main/CMSIS/DSP/Source/ComplexMathFunctions/arm_cmplx_mag_squared_q31.c

   1 /* ----------------------------------------------------------------------
   2  * Project:      CMSIS DSP Library
   3  * Title:        arm_cmplx_mag_squared_q31.c
   4  * Description:  Q31 complex magnitude squared
   5  *
   6  * $Date:        27. January 2017
   7  * $Revision:    V.1.5.1
   8  *
   9  * Target Processor: Cortex-M cores
  10  * -------------------------------------------------------------------- */
  11 /*
  12  * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
  13  *
  14  * SPDX-License-Identifier: Apache-2.0
  15  *
  16  * Licensed under the Apache License, Version 2.0 (the License); you may
  17  * not use this file except in compliance with the License.
  18  * You may obtain a copy of the License at
  19  *
  20  * www.apache.org/licenses/LICENSE-2.0
  21  *
  22  * Unless required by applicable law or agreed to in writing, software
  23  * distributed under the License is distributed on an AS IS BASIS, WITHOUT
  24  * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  25  * See the License for the specific language governing permissions and
  26  * limitations under the License.
  27  */
  28
  29 #include "arm_math.h"
  30
  31 /**
  32  * @ingroup groupCmplxMath
  33  */
  34
  35 /**
  36  * @addtogroup cmplx_mag_squared
  37  * @{
  38  */
  39
  40
  41 /**
  42  * @brief  Q31 complex magnitude squared
  43  * @param  *pSrc points to the complex input vector
  44  * @param  *pDst points to the real output vector
  45  * @param  numSamples number of complex samples in the input vector
  46  * @return none.
  47  *
  48  * <b>Scaling and Overflow Behavior:</b>
  49  * \par
  50  * The function implements 1.31 by 1.31 multiplications and finally output is converted into 3.29 format.
  51  * Input down scaling is not required.
  52  */
  53
  54 void arm_cmplx_mag_squared_q31(
  55   q31_t * pSrc,
  56   q31_t * pDst,
  57   uint32_t numSamples)
  58 {
  59   q31_t real, imag;                              /* Temporary variables to store real and imaginary values */
  60   q31_t acc0, acc1;                              /* Accumulators */
  61
  62 #if defined (ARM_MATH_DSP)
  63
  64   /* Run the below code for Cortex-M4 and Cortex-M3 */
  65   uint32_t blkCnt;                               /* loop counter */
  66
  67   /* loop Unrolling */
  68   blkCnt = numSamples >> 2U;
  69
  70   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
  71    ** a second loop below computes the remaining 1 to 3 samples. */
  72   while (blkCnt > 0U)
  73   {
  74     /* C[0] = (A[0] * A[0] + A[1] * A[1]) */
  75     real = *pSrc++;
  76     imag = *pSrc++;
  77     acc0 = (q31_t) (((q63_t) real * real) >> 33);
  78     acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
  79     /* store the result in 3.29 format in the destination buffer. */
  80     *pDst++ = acc0 + acc1;
  81
  82     real = *pSrc++;
  83     imag = *pSrc++;
  84     acc0 = (q31_t) (((q63_t) real * real) >> 33);
  85     acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
  86     /* store the result in 3.29 format in the destination buffer. */
  87     *pDst++ = acc0 + acc1;
  88
  89     real = *pSrc++;
  90     imag = *pSrc++;
  91     acc0 = (q31_t) (((q63_t) real * real) >> 33);
  92     acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
  93     /* store the result in 3.29 format in the destination buffer. */
  94     *pDst++ = acc0 + acc1;
  95
  96     real = *pSrc++;
  97     imag = *pSrc++;
  98     acc0 = (q31_t) (((q63_t) real * real) >> 33);
  99     acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
 100     /* store the result in 3.29 format in the destination buffer. */
 101     *pDst++ = acc0 + acc1;
 102
 103     /* Decrement the loop counter */
 104     blkCnt--;
 105   }
 106
 107   /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
 108    ** No loop unrolling is used. */
 109   blkCnt = numSamples % 0x4U;
 110
 111   while (blkCnt > 0U)
 112   {
 113     /* C[0] = (A[0] * A[0] + A[1] * A[1]) */
 114     real = *pSrc++;
 115     imag = *pSrc++;
 116     acc0 = (q31_t) (((q63_t) real * real) >> 33);
 117     acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
 118     /* store the result in 3.29 format in the destination buffer. */
 119     *pDst++ = acc0 + acc1;
 120
 121     /* Decrement the loop counter */
 122     blkCnt--;
 123   }
 124
 125 #else
 126
 127   /* Run the below code for Cortex-M0 */
 128
 129   while (numSamples > 0U)
 130   {
 131     /* out = ((real * real) + (imag * imag)) */
 132     real = *pSrc++;
 133     imag = *pSrc++;
 134     acc0 = (q31_t) (((q63_t) real * real) >> 33);
 135     acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
 136     /* store the result in 3.29 format in the destination buffer. */
 137     *pDst++ = acc0 + acc1;
 138
 139     /* Decrement the loop counter */
 140     numSamples--;
 141   }
 142
 143 #endif /* #if defined (ARM_MATH_DSP) */
 144
 145 }
 146
 147 /**
 148  * @} end of cmplx_mag_squared group
 149  */