mbed official / mbed-dsp

CMSIS DSP library

Dependents:   KL25Z_FFT_Demo Hat_Board_v5_1 KL25Z_FFT_Demo_tony KL25Z_FFT_Demo_tony ... more

Fork of mbed-dsp by mbed official

cmsis_dsp/StatisticsFunctions/arm_std_q15.c@2:da51fb522205, 2013-05-30 (annotated)

Committer:
emilmont
Date:
Thu May 30 17:10:11 2013 +0100
Revision:
2:da51fb522205
Parent:
1:fdd22bb7aa52
Child:
3:7a284390b0ce
Keep "cmsis-dsp" module in synch with its source

Who changed what in which revision?

UserRevisionLine numberNew contents of line
emilmont 1:fdd22bb7aa52 1 /* ----------------------------------------------------------------------
emilmont 1:fdd22bb7aa52 2 * Copyright (C) 2010 ARM Limited. All rights reserved.
emilmont 1:fdd22bb7aa52 3 *
emilmont 1:fdd22bb7aa52 4 * $Date: 15. February 2012
emilmont 2:da51fb522205 5 * $Revision: V1.1.0
emilmont 1:fdd22bb7aa52 6 *
emilmont 2:da51fb522205 7 * Project: CMSIS DSP Library
emilmont 2:da51fb522205 8 * Title: arm_std_q15.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 2:da51fb522205 10 * Description: Standard deviation of an array of Q15 type.
emilmont 1:fdd22bb7aa52 11 *
emilmont 1:fdd22bb7aa52 12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
emilmont 1:fdd22bb7aa52 13 *
emilmont 1:fdd22bb7aa52 14 * Version 1.1.0 2012/02/15
emilmont 1:fdd22bb7aa52 15 * Updated with more optimizations, bug fixes and minor API changes.
emilmont 1:fdd22bb7aa52 16 *
emilmont 1:fdd22bb7aa52 17 * Version 1.0.10 2011/7/15
emilmont 1:fdd22bb7aa52 18 * Big Endian support added and Merged M0 and M3/M4 Source code.
emilmont 1:fdd22bb7aa52 19 *
emilmont 1:fdd22bb7aa52 20 * Version 1.0.3 2010/11/29
emilmont 1:fdd22bb7aa52 21 * Re-organized the CMSIS folders and updated documentation.
emilmont 1:fdd22bb7aa52 22 *
emilmont 1:fdd22bb7aa52 23 * Version 1.0.2 2010/11/11
emilmont 1:fdd22bb7aa52 24 * Documentation updated.
emilmont 1:fdd22bb7aa52 25 *
emilmont 1:fdd22bb7aa52 26 * Version 1.0.1 2010/10/05
emilmont 1:fdd22bb7aa52 27 * Production release and review comments incorporated.
emilmont 1:fdd22bb7aa52 28 *
emilmont 1:fdd22bb7aa52 29 * Version 1.0.0 2010/09/20
emilmont 1:fdd22bb7aa52 30 * Production release and review comments incorporated.
emilmont 1:fdd22bb7aa52 31 * -------------------------------------------------------------------- */
emilmont 1:fdd22bb7aa52 32
emilmont 1:fdd22bb7aa52 33 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 34
emilmont 1:fdd22bb7aa52 35 /**
emilmont 1:fdd22bb7aa52 36 * @ingroup groupStats
emilmont 1:fdd22bb7aa52 37 */
emilmont 1:fdd22bb7aa52 38
emilmont 1:fdd22bb7aa52 39 /**
emilmont 1:fdd22bb7aa52 40 * @addtogroup STD
emilmont 1:fdd22bb7aa52 41 * @{
emilmont 1:fdd22bb7aa52 42 */
emilmont 1:fdd22bb7aa52 43
emilmont 1:fdd22bb7aa52 44 /**
emilmont 1:fdd22bb7aa52 45 * @brief Standard deviation of the elements of a Q15 vector.
emilmont 1:fdd22bb7aa52 46 * @param[in] *pSrc points to the input vector
emilmont 1:fdd22bb7aa52 47 * @param[in] blockSize length of the input vector
emilmont 1:fdd22bb7aa52 48 * @param[out] *pResult standard deviation value returned here
emilmont 1:fdd22bb7aa52 49 * @return none.
emilmont 1:fdd22bb7aa52 50 *
emilmont 1:fdd22bb7aa52 51 * @details
emilmont 1:fdd22bb7aa52 52 * <b>Scaling and Overflow Behavior:</b>
emilmont 1:fdd22bb7aa52 53 *
emilmont 1:fdd22bb7aa52 54 * \par
emilmont 1:fdd22bb7aa52 55 * The function is implemented using a 64-bit internal accumulator.
emilmont 1:fdd22bb7aa52 56 * The input is represented in 1.15 format.
emilmont 1:fdd22bb7aa52 57 * Intermediate multiplication yields a 2.30 format, and this
emilmont 1:fdd22bb7aa52 58 * result is added without saturation to a 64-bit accumulator in 34.30 format.
emilmont 1:fdd22bb7aa52 59 * With 33 guard bits in the accumulator, there is no risk of overflow, and the
emilmont 1:fdd22bb7aa52 60 * full precision of the intermediate multiplication is preserved.
emilmont 1:fdd22bb7aa52 61 * Finally, the 34.30 result is truncated to 34.15 format by discarding the lower
emilmont 1:fdd22bb7aa52 62 * 15 bits, and then saturated to yield a result in 1.15 format.
emilmont 1:fdd22bb7aa52 63 */
emilmont 1:fdd22bb7aa52 64
emilmont 1:fdd22bb7aa52 65 void arm_std_q15(
emilmont 1:fdd22bb7aa52 66 q15_t * pSrc,
emilmont 1:fdd22bb7aa52 67 uint32_t blockSize,
emilmont 1:fdd22bb7aa52 68 q15_t * pResult)
emilmont 1:fdd22bb7aa52 69 {
emilmont 1:fdd22bb7aa52 70 q31_t sum = 0; /* Accumulator */
emilmont 1:fdd22bb7aa52 71 q31_t meanOfSquares, squareOfMean; /* square of mean and mean of square */
emilmont 1:fdd22bb7aa52 72 q15_t mean; /* mean */
emilmont 1:fdd22bb7aa52 73 uint32_t blkCnt; /* loop counter */
emilmont 1:fdd22bb7aa52 74 q15_t t; /* Temporary variable */
emilmont 1:fdd22bb7aa52 75 q63_t sumOfSquares = 0; /* Accumulator */
emilmont 1:fdd22bb7aa52 76
emilmont 1:fdd22bb7aa52 77 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 78
emilmont 1:fdd22bb7aa52 79 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 80
emilmont 1:fdd22bb7aa52 81 q31_t in; /* input value */
emilmont 1:fdd22bb7aa52 82 q15_t in1; /* input value */
emilmont 1:fdd22bb7aa52 83
emilmont 1:fdd22bb7aa52 84 /*loop Unrolling */
emilmont 1:fdd22bb7aa52 85 blkCnt = blockSize >> 2u;
emilmont 1:fdd22bb7aa52 86
emilmont 1:fdd22bb7aa52 87 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
emilmont 1:fdd22bb7aa52 88 ** a second loop below computes the remaining 1 to 3 samples. */
emilmont 1:fdd22bb7aa52 89 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 90 {
emilmont 1:fdd22bb7aa52 91 /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */
emilmont 1:fdd22bb7aa52 92 /* Compute Sum of squares of the input samples
emilmont 1:fdd22bb7aa52 93 * and then store the result in a temporary variable, sum. */
emilmont 1:fdd22bb7aa52 94 in = *__SIMD32(pSrc)++;
emilmont 1:fdd22bb7aa52 95 sum += ((in << 16) >> 16);
emilmont 1:fdd22bb7aa52 96 sum += (in >> 16);
emilmont 1:fdd22bb7aa52 97 sumOfSquares = __SMLALD(in, in, sumOfSquares);
emilmont 1:fdd22bb7aa52 98 in = *__SIMD32(pSrc)++;
emilmont 1:fdd22bb7aa52 99 sum += ((in << 16) >> 16);
emilmont 1:fdd22bb7aa52 100 sum += (in >> 16);
emilmont 1:fdd22bb7aa52 101 sumOfSquares = __SMLALD(in, in, sumOfSquares);
emilmont 1:fdd22bb7aa52 102
emilmont 1:fdd22bb7aa52 103 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 104 blkCnt--;
emilmont 1:fdd22bb7aa52 105 }
emilmont 1:fdd22bb7aa52 106
emilmont 1:fdd22bb7aa52 107 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
emilmont 1:fdd22bb7aa52 108 ** No loop unrolling is used. */
emilmont 1:fdd22bb7aa52 109 blkCnt = blockSize % 0x4u;
emilmont 1:fdd22bb7aa52 110
emilmont 1:fdd22bb7aa52 111 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 112 {
emilmont 1:fdd22bb7aa52 113 /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */
emilmont 1:fdd22bb7aa52 114 /* Compute Sum of squares of the input samples
emilmont 1:fdd22bb7aa52 115 * and then store the result in a temporary variable, sum. */
emilmont 1:fdd22bb7aa52 116 in1 = *pSrc++;
emilmont 1:fdd22bb7aa52 117 sumOfSquares = __SMLALD(in1, in1, sumOfSquares);
emilmont 1:fdd22bb7aa52 118 sum += in1;
emilmont 1:fdd22bb7aa52 119
emilmont 1:fdd22bb7aa52 120 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 121 blkCnt--;
emilmont 1:fdd22bb7aa52 122 }
emilmont 1:fdd22bb7aa52 123
emilmont 1:fdd22bb7aa52 124 /* Compute Mean of squares of the input samples
emilmont 1:fdd22bb7aa52 125 * and then store the result in a temporary variable, meanOfSquares. */
emilmont 1:fdd22bb7aa52 126 t = (q15_t) ((1.0 / (blockSize - 1)) * 16384LL);
emilmont 1:fdd22bb7aa52 127 sumOfSquares = __SSAT((sumOfSquares >> 15u), 16u);
emilmont 1:fdd22bb7aa52 128
emilmont 1:fdd22bb7aa52 129 meanOfSquares = (q31_t) ((sumOfSquares * t) >> 14u);
emilmont 1:fdd22bb7aa52 130
emilmont 1:fdd22bb7aa52 131 /* Compute mean of all input values */
emilmont 1:fdd22bb7aa52 132 t = (q15_t) ((1.0 / (blockSize * (blockSize - 1))) * 32768LL);
emilmont 1:fdd22bb7aa52 133 mean = (q15_t) __SSAT(sum, 16u);
emilmont 1:fdd22bb7aa52 134
emilmont 1:fdd22bb7aa52 135 /* Compute square of mean */
emilmont 1:fdd22bb7aa52 136 squareOfMean = ((q31_t) mean * mean) >> 15;
emilmont 1:fdd22bb7aa52 137 squareOfMean = (q31_t) (((q63_t) squareOfMean * t) >> 15);
emilmont 1:fdd22bb7aa52 138
emilmont 1:fdd22bb7aa52 139 /* mean of the squares minus the square of the mean. */
emilmont 1:fdd22bb7aa52 140 in1 = (q15_t) (meanOfSquares - squareOfMean);
emilmont 1:fdd22bb7aa52 141
emilmont 1:fdd22bb7aa52 142 /* Compute standard deviation and store the result to the destination */
emilmont 1:fdd22bb7aa52 143 arm_sqrt_q15(in1, pResult);
emilmont 1:fdd22bb7aa52 144
emilmont 1:fdd22bb7aa52 145 #else
emilmont 1:fdd22bb7aa52 146
emilmont 1:fdd22bb7aa52 147 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 148 q15_t in; /* input value */
emilmont 1:fdd22bb7aa52 149
emilmont 1:fdd22bb7aa52 150 /* Loop over blockSize number of values */
emilmont 1:fdd22bb7aa52 151 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 152
emilmont 1:fdd22bb7aa52 153 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 154 {
emilmont 1:fdd22bb7aa52 155 /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */
emilmont 1:fdd22bb7aa52 156 /* Compute Sum of squares of the input samples
emilmont 1:fdd22bb7aa52 157 * and then store the result in a temporary variable, sumOfSquares. */
emilmont 1:fdd22bb7aa52 158 in = *pSrc++;
emilmont 1:fdd22bb7aa52 159 sumOfSquares += (in * in);
emilmont 1:fdd22bb7aa52 160
emilmont 1:fdd22bb7aa52 161 /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) */
emilmont 1:fdd22bb7aa52 162 /* Compute sum of all input values and then store the result in a temporary variable, sum. */
emilmont 1:fdd22bb7aa52 163 sum += in;
emilmont 1:fdd22bb7aa52 164
emilmont 1:fdd22bb7aa52 165 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 166 blkCnt--;
emilmont 1:fdd22bb7aa52 167 }
emilmont 1:fdd22bb7aa52 168
emilmont 1:fdd22bb7aa52 169 /* Compute Mean of squares of the input samples
emilmont 1:fdd22bb7aa52 170 * and then store the result in a temporary variable, meanOfSquares. */
emilmont 1:fdd22bb7aa52 171 t = (q15_t) ((1.0 / (blockSize - 1)) * 16384LL);
emilmont 1:fdd22bb7aa52 172 sumOfSquares = __SSAT((sumOfSquares >> 15u), 16u);
emilmont 1:fdd22bb7aa52 173 meanOfSquares = (q31_t) ((sumOfSquares * t) >> 14u);
emilmont 1:fdd22bb7aa52 174
emilmont 1:fdd22bb7aa52 175 /* Compute mean of all input values */
emilmont 1:fdd22bb7aa52 176 mean = (q15_t) __SSAT(sum, 16u);
emilmont 1:fdd22bb7aa52 177
emilmont 1:fdd22bb7aa52 178 /* Compute square of mean of the input samples
emilmont 1:fdd22bb7aa52 179 * and then store the result in a temporary variable, squareOfMean.*/
emilmont 1:fdd22bb7aa52 180 t = (q15_t) ((1.0 / (blockSize * (blockSize - 1))) * 32768LL);
emilmont 1:fdd22bb7aa52 181 squareOfMean = ((q31_t) mean * mean) >> 15;
emilmont 1:fdd22bb7aa52 182 squareOfMean = (q31_t) (((q63_t) squareOfMean * t) >> 15);
emilmont 1:fdd22bb7aa52 183
emilmont 1:fdd22bb7aa52 184 /* mean of the squares minus the square of the mean. */
emilmont 1:fdd22bb7aa52 185 in = (q15_t) (meanOfSquares - squareOfMean);
emilmont 1:fdd22bb7aa52 186
emilmont 1:fdd22bb7aa52 187 /* Compute standard deviation and store the result to the destination */
emilmont 1:fdd22bb7aa52 188 arm_sqrt_q15(in, pResult);
emilmont 1:fdd22bb7aa52 189
emilmont 1:fdd22bb7aa52 190 #endif /* #ifndef ARM_MATH_CM0 */
emilmont 1:fdd22bb7aa52 191
emilmont 1:fdd22bb7aa52 192
emilmont 1:fdd22bb7aa52 193 }
emilmont 1:fdd22bb7aa52 194
emilmont 1:fdd22bb7aa52 195 /**
emilmont 1:fdd22bb7aa52 196 * @} end of STD group
emilmont 1:fdd22bb7aa52 197 */