arm_power_f32.c

00001 /* ----------------------------------------------------------------------
00002  * Project:      CMSIS DSP Library
00003  * Title:        arm_power_f32.c
00004  * Description:  Sum of the squares of the elements of a floating-point vector
00005  *
00006  * $Date:        27. January 2017
00007  * $Revision:    V.1.5.1
00008  *
00009  * Target Processor: Cortex-M cores
00010  * -------------------------------------------------------------------- */
00011 /*
00012  * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
00013  *
00014  * SPDX-License-Identifier: Apache-2.0
00015  *
00016  * Licensed under the Apache License, Version 2.0 (the License); you may
00017  * not use this file except in compliance with the License.
00018  * You may obtain a copy of the License at
00019  *
00020  * www.apache.org/licenses/LICENSE-2.0
00021  *
00022  * Unless required by applicable law or agreed to in writing, software
00023  * distributed under the License is distributed on an AS IS BASIS, WITHOUT
00024  * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
00025  * See the License for the specific language governing permissions and
00026  * limitations under the License.
00027  */
00028 
00029 #include "arm_math.h"
00030 
00031 /**
00032  * @ingroup groupStats
00033  */
00034 
00035 /**
00036  * @defgroup power Power
00037  *
00038  * Calculates the sum of the squares of the elements in the input vector.
00039  * The underlying algorithm is used:
00040  *
00041  * <pre>
00042  *  Result = pSrc[0] * pSrc[0] + pSrc[1] * pSrc[1] + pSrc[2] * pSrc[2] + ... + pSrc[blockSize-1] * pSrc[blockSize-1];
00043  * </pre>
00044  *
00045  * There are separate functions for floating point, Q31, Q15, and Q7 data types.
00046  */
00047 
00048 /**
00049  * @addtogroup power
00050  * @{
00051  */
00052 
00053 
00054 /**
00055  * @brief Sum of the squares of the elements of a floating-point vector.
00056  * @param[in]       *pSrc points to the input vector
00057  * @param[in]       blockSize length of the input vector
00058  * @param[out]      *pResult sum of the squares value returned here
00059  * @return none.
00060  *
00061  */
00062 
00063 
00064 void arm_power_f32(
00065   float32_t * pSrc,
00066   uint32_t blockSize,
00067   float32_t * pResult)
00068 {
00069   float32_t sum = 0.0f;                          /* accumulator */
00070   float32_t in;                                  /* Temporary variable to store input value */
00071   uint32_t blkCnt;                               /* loop counter */
00072 
00073 #if defined (ARM_MATH_DSP)
00074   /* Run the below code for Cortex-M4 and Cortex-M3 */
00075 
00076   /*loop Unrolling */
00077   blkCnt = blockSize >> 2U;
00078 
00079   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
00080    ** a second loop below computes the remaining 1 to 3 samples. */
00081   while (blkCnt > 0U)
00082   {
00083     /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */
00084     /* Compute Power and then store the result in a temporary variable, sum. */
00085     in = *pSrc++;
00086     sum += in * in;
00087     in = *pSrc++;
00088     sum += in * in;
00089     in = *pSrc++;
00090     sum += in * in;
00091     in = *pSrc++;
00092     sum += in * in;
00093 
00094     /* Decrement the loop counter */
00095     blkCnt--;
00096   }
00097 
00098   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
00099    ** No loop unrolling is used. */
00100   blkCnt = blockSize % 0x4U;
00101 
00102 
00103 #else
00104   /* Run the below code for Cortex-M0 */
00105 
00106   /* Loop over blockSize number of values */
00107   blkCnt = blockSize;
00108 
00109 #endif /* #if defined (ARM_MATH_DSP) */
00110 
00111 
00112   while (blkCnt > 0U)
00113   {
00114     /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */
00115     /* compute power and then store the result in a temporary variable, sum. */
00116     in = *pSrc++;
00117     sum += in * in;
00118 
00119     /* Decrement the loop counter */
00120     blkCnt--;
00121   }
00122 
00123   /* Store the result to the destination */
00124   *pResult = sum;
00125 }
00126 
00127 /**
00128  * @} end of power group
00129  */
00130