Joep D / CMSIS_DSP_5

The CMSIS DSP 5 library

Dependents:   Nucleo-Heart-Rate ejercicioVrms2 PROYECTOFINAL ejercicioVrms ... more

functions/StatisticsFunctions/arm_power_q31.c@3:4098b9d3d571, 2018-06-21 (annotated)

Committer:
xorjoep
Date:
Thu Jun 21 11:56:27 2018 +0000
Revision:
3:4098b9d3d571
Parent:
1:24714b45cd1b 
headers is a folder not a library

Who changed what in which revision?

UserRevisionLine numberNew contents of line
xorjoep 1:24714b45cd1b 1 /* ----------------------------------------------------------------------
xorjoep 1:24714b45cd1b 2 * Project: CMSIS DSP Library
xorjoep 1:24714b45cd1b 3 * Title: arm_power_q31.c
xorjoep 1:24714b45cd1b 4 * Description: Sum of the squares of the elements of a Q31 vector
xorjoep 1:24714b45cd1b 5 *
xorjoep 1:24714b45cd1b 6 * $Date: 27. January 2017
xorjoep 1:24714b45cd1b 7 * $Revision: V.1.5.1
xorjoep 1:24714b45cd1b 8 *
xorjoep 1:24714b45cd1b 9 * Target Processor: Cortex-M cores
xorjoep 1:24714b45cd1b 10 * -------------------------------------------------------------------- */
xorjoep 1:24714b45cd1b 11 /*
xorjoep 1:24714b45cd1b 12 * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
xorjoep 1:24714b45cd1b 13 *
xorjoep 1:24714b45cd1b 14 * SPDX-License-Identifier: Apache-2.0
xorjoep 1:24714b45cd1b 15 *
xorjoep 1:24714b45cd1b 16 * Licensed under the Apache License, Version 2.0 (the License); you may
xorjoep 1:24714b45cd1b 17 * not use this file except in compliance with the License.
xorjoep 1:24714b45cd1b 18 * You may obtain a copy of the License at
xorjoep 1:24714b45cd1b 19 *
xorjoep 1:24714b45cd1b 20 * www.apache.org/licenses/LICENSE-2.0
xorjoep 1:24714b45cd1b 21 *
xorjoep 1:24714b45cd1b 22 * Unless required by applicable law or agreed to in writing, software
xorjoep 1:24714b45cd1b 23 * distributed under the License is distributed on an AS IS BASIS, WITHOUT
xorjoep 1:24714b45cd1b 24 * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
xorjoep 1:24714b45cd1b 25 * See the License for the specific language governing permissions and
xorjoep 1:24714b45cd1b 26 * limitations under the License.
xorjoep 1:24714b45cd1b 27 */
xorjoep 1:24714b45cd1b 28
xorjoep 1:24714b45cd1b 29 #include "arm_math.h"
xorjoep 1:24714b45cd1b 30
xorjoep 1:24714b45cd1b 31 /**
xorjoep 1:24714b45cd1b 32 * @ingroup groupStats
xorjoep 1:24714b45cd1b 33 */
xorjoep 1:24714b45cd1b 34
xorjoep 1:24714b45cd1b 35 /**
xorjoep 1:24714b45cd1b 36 * @addtogroup power
xorjoep 1:24714b45cd1b 37 * @{
xorjoep 1:24714b45cd1b 38 */
xorjoep 1:24714b45cd1b 39
xorjoep 1:24714b45cd1b 40 /**
xorjoep 1:24714b45cd1b 41 * @brief Sum of the squares of the elements of a Q31 vector.
xorjoep 1:24714b45cd1b 42 * @param[in] *pSrc points to the input vector
xorjoep 1:24714b45cd1b 43 * @param[in] blockSize length of the input vector
xorjoep 1:24714b45cd1b 44 * @param[out] *pResult sum of the squares value returned here
xorjoep 1:24714b45cd1b 45 * @return none.
xorjoep 1:24714b45cd1b 46 *
xorjoep 1:24714b45cd1b 47 * @details
xorjoep 1:24714b45cd1b 48 * <b>Scaling and Overflow Behavior:</b>
xorjoep 1:24714b45cd1b 49 *
xorjoep 1:24714b45cd1b 50 * \par
xorjoep 1:24714b45cd1b 51 * The function is implemented using a 64-bit internal accumulator.
xorjoep 1:24714b45cd1b 52 * The input is represented in 1.31 format.
xorjoep 1:24714b45cd1b 53 * Intermediate multiplication yields a 2.62 format, and this
xorjoep 1:24714b45cd1b 54 * result is truncated to 2.48 format by discarding the lower 14 bits.
xorjoep 1:24714b45cd1b 55 * The 2.48 result is then added without saturation to a 64-bit accumulator in 16.48 format.
xorjoep 1:24714b45cd1b 56 * With 15 guard bits in the accumulator, there is no risk of overflow, and the
xorjoep 1:24714b45cd1b 57 * full precision of the intermediate multiplication is preserved.
xorjoep 1:24714b45cd1b 58 * Finally, the return result is in 16.48 format.
xorjoep 1:24714b45cd1b 59 *
xorjoep 1:24714b45cd1b 60 */
xorjoep 1:24714b45cd1b 61
xorjoep 1:24714b45cd1b 62 void arm_power_q31(
xorjoep 1:24714b45cd1b 63 q31_t * pSrc,
xorjoep 1:24714b45cd1b 64 uint32_t blockSize,
xorjoep 1:24714b45cd1b 65 q63_t * pResult)
xorjoep 1:24714b45cd1b 66 {
xorjoep 1:24714b45cd1b 67 q63_t sum = 0; /* Temporary result storage */
xorjoep 1:24714b45cd1b 68 q31_t in;
xorjoep 1:24714b45cd1b 69 uint32_t blkCnt; /* loop counter */
xorjoep 1:24714b45cd1b 70
xorjoep 1:24714b45cd1b 71
xorjoep 1:24714b45cd1b 72 #if defined (ARM_MATH_DSP)
xorjoep 1:24714b45cd1b 73 /* Run the below code for Cortex-M4 and Cortex-M3 */
xorjoep 1:24714b45cd1b 74
xorjoep 1:24714b45cd1b 75 /*loop Unrolling */
xorjoep 1:24714b45cd1b 76 blkCnt = blockSize >> 2U;
xorjoep 1:24714b45cd1b 77
xorjoep 1:24714b45cd1b 78 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
xorjoep 1:24714b45cd1b 79 ** a second loop below computes the remaining 1 to 3 samples. */
xorjoep 1:24714b45cd1b 80 while (blkCnt > 0U)
xorjoep 1:24714b45cd1b 81 {
xorjoep 1:24714b45cd1b 82 /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */
xorjoep 1:24714b45cd1b 83 /* Compute Power then shift intermediate results by 14 bits to maintain 16.48 format and then store the result in a temporary variable sum, providing 15 guard bits. */
xorjoep 1:24714b45cd1b 84 in = *pSrc++;
xorjoep 1:24714b45cd1b 85 sum += ((q63_t) in * in) >> 14U;
xorjoep 1:24714b45cd1b 86
xorjoep 1:24714b45cd1b 87 in = *pSrc++;
xorjoep 1:24714b45cd1b 88 sum += ((q63_t) in * in) >> 14U;
xorjoep 1:24714b45cd1b 89
xorjoep 1:24714b45cd1b 90 in = *pSrc++;
xorjoep 1:24714b45cd1b 91 sum += ((q63_t) in * in) >> 14U;
xorjoep 1:24714b45cd1b 92
xorjoep 1:24714b45cd1b 93 in = *pSrc++;
xorjoep 1:24714b45cd1b 94 sum += ((q63_t) in * in) >> 14U;
xorjoep 1:24714b45cd1b 95
xorjoep 1:24714b45cd1b 96 /* Decrement the loop counter */
xorjoep 1:24714b45cd1b 97 blkCnt--;
xorjoep 1:24714b45cd1b 98 }
xorjoep 1:24714b45cd1b 99
xorjoep 1:24714b45cd1b 100 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
xorjoep 1:24714b45cd1b 101 ** No loop unrolling is used. */
xorjoep 1:24714b45cd1b 102 blkCnt = blockSize % 0x4U;
xorjoep 1:24714b45cd1b 103
xorjoep 1:24714b45cd1b 104 #else
xorjoep 1:24714b45cd1b 105 /* Run the below code for Cortex-M0 */
xorjoep 1:24714b45cd1b 106
xorjoep 1:24714b45cd1b 107 /* Loop over blockSize number of values */
xorjoep 1:24714b45cd1b 108 blkCnt = blockSize;
xorjoep 1:24714b45cd1b 109
xorjoep 1:24714b45cd1b 110 #endif /* #if defined (ARM_MATH_DSP) */
xorjoep 1:24714b45cd1b 111
xorjoep 1:24714b45cd1b 112 while (blkCnt > 0U)
xorjoep 1:24714b45cd1b 113 {
xorjoep 1:24714b45cd1b 114 /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */
xorjoep 1:24714b45cd1b 115 /* Compute Power and then store the result in a temporary variable, sum. */
xorjoep 1:24714b45cd1b 116 in = *pSrc++;
xorjoep 1:24714b45cd1b 117 sum += ((q63_t) in * in) >> 14U;
xorjoep 1:24714b45cd1b 118
xorjoep 1:24714b45cd1b 119 /* Decrement the loop counter */
xorjoep 1:24714b45cd1b 120 blkCnt--;
xorjoep 1:24714b45cd1b 121 }
xorjoep 1:24714b45cd1b 122
xorjoep 1:24714b45cd1b 123 /* Store the results in 16.48 format */
xorjoep 1:24714b45cd1b 124 *pResult = sum;
xorjoep 1:24714b45cd1b 125 }
xorjoep 1:24714b45cd1b 126
xorjoep 1:24714b45cd1b 127 /**
xorjoep 1:24714b45cd1b 128 * @} end of power group
xorjoep 1:24714b45cd1b 129 */