Joep D / CMSIS_DSP_5

The CMSIS DSP 5 library

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

functions/StatisticsFunctions/arm_rms_q15.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_rms_q15.c
xorjoep 1:24714b45cd1b 4 * Description: Root Mean Square of the elements of a Q15 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 * @addtogroup RMS
xorjoep 1:24714b45cd1b 33 * @{
xorjoep 1:24714b45cd1b 34 */
xorjoep 1:24714b45cd1b 35
xorjoep 1:24714b45cd1b 36 /**
xorjoep 1:24714b45cd1b 37 * @brief Root Mean Square of the elements of a Q15 vector.
xorjoep 1:24714b45cd1b 38 * @param[in] *pSrc points to the input vector
xorjoep 1:24714b45cd1b 39 * @param[in] blockSize length of the input vector
xorjoep 1:24714b45cd1b 40 * @param[out] *pResult rms value returned here
xorjoep 1:24714b45cd1b 41 * @return none.
xorjoep 1:24714b45cd1b 42 *
xorjoep 1:24714b45cd1b 43 * @details
xorjoep 1:24714b45cd1b 44 * <b>Scaling and Overflow Behavior:</b>
xorjoep 1:24714b45cd1b 45 *
xorjoep 1:24714b45cd1b 46 * \par
xorjoep 1:24714b45cd1b 47 * The function is implemented using a 64-bit internal accumulator.
xorjoep 1:24714b45cd1b 48 * The input is represented in 1.15 format.
xorjoep 1:24714b45cd1b 49 * Intermediate multiplication yields a 2.30 format, and this
xorjoep 1:24714b45cd1b 50 * result is added without saturation to a 64-bit accumulator in 34.30 format.
xorjoep 1:24714b45cd1b 51 * With 33 guard bits in the accumulator, there is no risk of overflow, and the
xorjoep 1:24714b45cd1b 52 * full precision of the intermediate multiplication is preserved.
xorjoep 1:24714b45cd1b 53 * Finally, the 34.30 result is truncated to 34.15 format by discarding the lower
xorjoep 1:24714b45cd1b 54 * 15 bits, and then saturated to yield a result in 1.15 format.
xorjoep 1:24714b45cd1b 55 *
xorjoep 1:24714b45cd1b 56 */
xorjoep 1:24714b45cd1b 57
xorjoep 1:24714b45cd1b 58 void arm_rms_q15(
xorjoep 1:24714b45cd1b 59 q15_t * pSrc,
xorjoep 1:24714b45cd1b 60 uint32_t blockSize,
xorjoep 1:24714b45cd1b 61 q15_t * pResult)
xorjoep 1:24714b45cd1b 62 {
xorjoep 1:24714b45cd1b 63 q63_t sum = 0; /* accumulator */
xorjoep 1:24714b45cd1b 64
xorjoep 1:24714b45cd1b 65 #if defined (ARM_MATH_DSP)
xorjoep 1:24714b45cd1b 66 /* Run the below code for Cortex-M4 and Cortex-M3 */
xorjoep 1:24714b45cd1b 67
xorjoep 1:24714b45cd1b 68 q31_t in; /* temporary variable to store the input value */
xorjoep 1:24714b45cd1b 69 q15_t in1; /* temporary variable to store the input value */
xorjoep 1:24714b45cd1b 70 uint32_t blkCnt; /* loop counter */
xorjoep 1:24714b45cd1b 71
xorjoep 1:24714b45cd1b 72 /* loop Unrolling */
xorjoep 1:24714b45cd1b 73 blkCnt = blockSize >> 2U;
xorjoep 1:24714b45cd1b 74
xorjoep 1:24714b45cd1b 75 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
xorjoep 1:24714b45cd1b 76 ** a second loop below computes the remaining 1 to 3 samples. */
xorjoep 1:24714b45cd1b 77 while (blkCnt > 0U)
xorjoep 1:24714b45cd1b 78 {
xorjoep 1:24714b45cd1b 79 /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */
xorjoep 1:24714b45cd1b 80 /* Compute sum of the squares and then store the results in a temporary variable, sum */
xorjoep 1:24714b45cd1b 81 in = *__SIMD32(pSrc)++;
xorjoep 1:24714b45cd1b 82 sum = __SMLALD(in, in, sum);
xorjoep 1:24714b45cd1b 83 in = *__SIMD32(pSrc)++;
xorjoep 1:24714b45cd1b 84 sum = __SMLALD(in, in, sum);
xorjoep 1:24714b45cd1b 85
xorjoep 1:24714b45cd1b 86 /* Decrement the loop counter */
xorjoep 1:24714b45cd1b 87 blkCnt--;
xorjoep 1:24714b45cd1b 88 }
xorjoep 1:24714b45cd1b 89
xorjoep 1:24714b45cd1b 90 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
xorjoep 1:24714b45cd1b 91 ** No loop unrolling is used. */
xorjoep 1:24714b45cd1b 92 blkCnt = blockSize % 0x4U;
xorjoep 1:24714b45cd1b 93
xorjoep 1:24714b45cd1b 94 while (blkCnt > 0U)
xorjoep 1:24714b45cd1b 95 {
xorjoep 1:24714b45cd1b 96 /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */
xorjoep 1:24714b45cd1b 97 /* Compute sum of the squares and then store the results in a temporary variable, sum */
xorjoep 1:24714b45cd1b 98 in1 = *pSrc++;
xorjoep 1:24714b45cd1b 99 sum = __SMLALD(in1, in1, sum);
xorjoep 1:24714b45cd1b 100
xorjoep 1:24714b45cd1b 101 /* Decrement the loop counter */
xorjoep 1:24714b45cd1b 102 blkCnt--;
xorjoep 1:24714b45cd1b 103 }
xorjoep 1:24714b45cd1b 104
xorjoep 1:24714b45cd1b 105 /* Truncating and saturating the accumulator to 1.15 format */
xorjoep 1:24714b45cd1b 106 /* Store the result in the destination */
xorjoep 1:24714b45cd1b 107 arm_sqrt_q15(__SSAT((sum / (q63_t)blockSize) >> 15, 16), pResult);
xorjoep 1:24714b45cd1b 108
xorjoep 1:24714b45cd1b 109 #else
xorjoep 1:24714b45cd1b 110 /* Run the below code for Cortex-M0 */
xorjoep 1:24714b45cd1b 111
xorjoep 1:24714b45cd1b 112 q15_t in; /* temporary variable to store the input value */
xorjoep 1:24714b45cd1b 113 uint32_t blkCnt; /* loop counter */
xorjoep 1:24714b45cd1b 114
xorjoep 1:24714b45cd1b 115 /* Loop over blockSize number of values */
xorjoep 1:24714b45cd1b 116 blkCnt = blockSize;
xorjoep 1:24714b45cd1b 117
xorjoep 1:24714b45cd1b 118 while (blkCnt > 0U)
xorjoep 1:24714b45cd1b 119 {
xorjoep 1:24714b45cd1b 120 /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */
xorjoep 1:24714b45cd1b 121 /* Compute sum of the squares and then store the results in a temporary variable, sum */
xorjoep 1:24714b45cd1b 122 in = *pSrc++;
xorjoep 1:24714b45cd1b 123 sum += ((q31_t) in * in);
xorjoep 1:24714b45cd1b 124
xorjoep 1:24714b45cd1b 125 /* Decrement the loop counter */
xorjoep 1:24714b45cd1b 126 blkCnt--;
xorjoep 1:24714b45cd1b 127 }
xorjoep 1:24714b45cd1b 128
xorjoep 1:24714b45cd1b 129 /* Truncating and saturating the accumulator to 1.15 format */
xorjoep 1:24714b45cd1b 130 /* Store the result in the destination */
xorjoep 1:24714b45cd1b 131 arm_sqrt_q15(__SSAT((sum / (q63_t)blockSize) >> 15, 16), pResult);
xorjoep 1:24714b45cd1b 132
xorjoep 1:24714b45cd1b 133 #endif /* #if defined (ARM_MATH_DSP) */
xorjoep 1:24714b45cd1b 134
xorjoep 1:24714b45cd1b 135 }
xorjoep 1:24714b45cd1b 136
xorjoep 1:24714b45cd1b 137 /**
xorjoep 1:24714b45cd1b 138 * @} end of RMS group
xorjoep 1:24714b45cd1b 139 */