lib/main/CMSIS/DSP/Source/ComplexMathFunctions/arm_cmplx_mag_q15.c

   1 /* ----------------------------------------------------------------------
   2  * Project:      CMSIS DSP Library
   3  * Title:        arm_cmplx_mag_q15.c
   4  * Description:  Q15 complex magnitude
   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
  37  * @{
  38  */
  39
  40
  41 /**
  42  * @brief  Q15 complex magnitude
  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 1.15 by 1.15 multiplications and finally output is converted into 2.14 format.
  51  */
  52
  53 void arm_cmplx_mag_q15(
  54   q15_t * pSrc,
  55   q15_t * pDst,
  56   uint32_t numSamples)
  57 {
  58   q31_t acc0, acc1;                              /* Accumulators */
  59
  60 #if defined (ARM_MATH_DSP)
  61
  62   /* Run the below code for Cortex-M4 and Cortex-M3 */
  63   uint32_t blkCnt;                               /* loop counter */
  64   q31_t in1, in2, in3, in4;
  65   q31_t acc2, acc3;
  66
  67
  68   /*loop Unrolling */
  69   blkCnt = numSamples >> 2U;
  70
  71   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
  72    ** a second loop below computes the remaining 1 to 3 samples. */
  73   while (blkCnt > 0U)
  74   {
  75
  76     /* C[0] = sqrt(A[0] * A[0] + A[1] * A[1]) */
  77     in1 = *__SIMD32(pSrc)++;
  78     in2 = *__SIMD32(pSrc)++;
  79     in3 = *__SIMD32(pSrc)++;
  80     in4 = *__SIMD32(pSrc)++;
  81
  82     acc0 = __SMUAD(in1, in1);
  83     acc1 = __SMUAD(in2, in2);
  84     acc2 = __SMUAD(in3, in3);
  85     acc3 = __SMUAD(in4, in4);
  86
  87     /* store the result in 2.14 format in the destination buffer. */
  88     arm_sqrt_q15((q15_t) ((acc0) >> 17), pDst++);
  89     arm_sqrt_q15((q15_t) ((acc1) >> 17), pDst++);
  90     arm_sqrt_q15((q15_t) ((acc2) >> 17), pDst++);
  91     arm_sqrt_q15((q15_t) ((acc3) >> 17), pDst++);
  92
  93     /* Decrement the loop counter */
  94     blkCnt--;
  95   }
  96
  97   /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
  98    ** No loop unrolling is used. */
  99   blkCnt = numSamples % 0x4U;
 100
 101   while (blkCnt > 0U)
 102   {
 103     /* C[0] = sqrt(A[0] * A[0] + A[1] * A[1]) */
 104     in1 = *__SIMD32(pSrc)++;
 105     acc0 = __SMUAD(in1, in1);
 106
 107     /* store the result in 2.14 format in the destination buffer. */
 108     arm_sqrt_q15((q15_t) (acc0 >> 17), pDst++);
 109
 110     /* Decrement the loop counter */
 111     blkCnt--;
 112   }
 113
 114 #else
 115
 116   /* Run the below code for Cortex-M0 */
 117   q15_t real, imag;                              /* Temporary variables to hold input values */
 118
 119   while (numSamples > 0U)
 120   {
 121     /* out = sqrt(real * real + imag * imag) */
 122     real = *pSrc++;
 123     imag = *pSrc++;
 124
 125     acc0 = (real * real);
 126     acc1 = (imag * imag);
 127
 128     /* store the result in 2.14 format in the destination buffer. */
 129     arm_sqrt_q15((q15_t) (((q63_t) acc0 + acc1) >> 17), pDst++);
 130
 131     /* Decrement the loop counter */
 132     numSamples--;
 133   }
 134
 135 #endif /* #if defined (ARM_MATH_DSP) */
 136
 137 }
 138
 139 /**
 140  * @} end of cmplx_mag group
 141  */