Go to most recent revision | Details | Last modification | View Log | RSS feed
| Rev | Author | Line No. | Line |
|---|---|---|---|
| 2 | mjames | 1 | /* ---------------------------------------------------------------------- |
| 2 | * Copyright (C) 2010-2014 ARM Limited. All rights reserved. |
||
| 3 | * |
||
| 4 | * $Date: 19. March 2015 |
||
| 5 | * $Revision: V.1.4.5 |
||
| 6 | * |
||
| 7 | * Project: CMSIS DSP Library |
||
| 8 | * Title: arm_std_q15.c |
||
| 9 | * |
||
| 10 | * Description: Standard deviation of an array of Q15 type. |
||
| 11 | * |
||
| 12 | * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0 |
||
| 13 | * |
||
| 14 | * Redistribution and use in source and binary forms, with or without |
||
| 15 | * modification, are permitted provided that the following conditions |
||
| 16 | * are met: |
||
| 17 | * - Redistributions of source code must retain the above copyright |
||
| 18 | * notice, this list of conditions and the following disclaimer. |
||
| 19 | * - Redistributions in binary form must reproduce the above copyright |
||
| 20 | * notice, this list of conditions and the following disclaimer in |
||
| 21 | * the documentation and/or other materials provided with the |
||
| 22 | * distribution. |
||
| 23 | * - Neither the name of ARM LIMITED nor the names of its contributors |
||
| 24 | * may be used to endorse or promote products derived from this |
||
| 25 | * software without specific prior written permission. |
||
| 26 | * |
||
| 27 | * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
||
| 28 | * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
||
| 29 | * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS |
||
| 30 | * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE |
||
| 31 | * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, |
||
| 32 | * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, |
||
| 33 | * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; |
||
| 34 | * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER |
||
| 35 | * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
||
| 36 | * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN |
||
| 37 | * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE |
||
| 38 | * POSSIBILITY OF SUCH DAMAGE. |
||
| 39 | * -------------------------------------------------------------------- */ |
||
| 40 | |||
| 41 | #include "arm_math.h" |
||
| 42 | |||
| 43 | /** |
||
| 44 | * @ingroup groupStats |
||
| 45 | */ |
||
| 46 | |||
| 47 | /** |
||
| 48 | * @addtogroup STD |
||
| 49 | * @{ |
||
| 50 | */ |
||
| 51 | |||
| 52 | /** |
||
| 53 | * @brief Standard deviation of the elements of a Q15 vector. |
||
| 54 | * @param[in] *pSrc points to the input vector |
||
| 55 | * @param[in] blockSize length of the input vector |
||
| 56 | * @param[out] *pResult standard deviation value returned here |
||
| 57 | * @return none. |
||
| 58 | * |
||
| 59 | * @details |
||
| 60 | * <b>Scaling and Overflow Behavior:</b> |
||
| 61 | * |
||
| 62 | * \par |
||
| 63 | * The function is implemented using a 64-bit internal accumulator. |
||
| 64 | * The input is represented in 1.15 format. |
||
| 65 | * Intermediate multiplication yields a 2.30 format, and this |
||
| 66 | * result is added without saturation to a 64-bit accumulator in 34.30 format. |
||
| 67 | * With 33 guard bits in the accumulator, there is no risk of overflow, and the |
||
| 68 | * full precision of the intermediate multiplication is preserved. |
||
| 69 | * Finally, the 34.30 result is truncated to 34.15 format by discarding the lower |
||
| 70 | * 15 bits, and then saturated to yield a result in 1.15 format. |
||
| 71 | */ |
||
| 72 | |||
| 73 | void arm_std_q15( |
||
| 74 | q15_t * pSrc, |
||
| 75 | uint32_t blockSize, |
||
| 76 | q15_t * pResult) |
||
| 77 | { |
||
| 78 | q31_t sum = 0; /* Accumulator */ |
||
| 79 | q31_t meanOfSquares, squareOfMean; /* square of mean and mean of square */ |
||
| 80 | uint32_t blkCnt; /* loop counter */ |
||
| 81 | q63_t sumOfSquares = 0; /* Accumulator */ |
||
| 82 | |||
| 83 | #ifndef ARM_MATH_CM0_FAMILY |
||
| 84 | |||
| 85 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
||
| 86 | |||
| 87 | q31_t in; /* input value */ |
||
| 88 | q15_t in1; /* input value */ |
||
| 89 | |||
| 90 | if(blockSize == 1) |
||
| 91 | { |
||
| 92 | *pResult = 0; |
||
| 93 | return; |
||
| 94 | } |
||
| 95 | |||
| 96 | /*loop Unrolling */ |
||
| 97 | blkCnt = blockSize >> 2u; |
||
| 98 | |||
| 99 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
||
| 100 | ** a second loop below computes the remaining 1 to 3 samples. */ |
||
| 101 | while(blkCnt > 0u) |
||
| 102 | { |
||
| 103 | /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */ |
||
| 104 | /* Compute Sum of squares of the input samples |
||
| 105 | * and then store the result in a temporary variable, sum. */ |
||
| 106 | in = *__SIMD32(pSrc)++; |
||
| 107 | sum += ((in << 16) >> 16); |
||
| 108 | sum += (in >> 16); |
||
| 109 | sumOfSquares = __SMLALD(in, in, sumOfSquares); |
||
| 110 | in = *__SIMD32(pSrc)++; |
||
| 111 | sum += ((in << 16) >> 16); |
||
| 112 | sum += (in >> 16); |
||
| 113 | sumOfSquares = __SMLALD(in, in, sumOfSquares); |
||
| 114 | |||
| 115 | /* Decrement the loop counter */ |
||
| 116 | blkCnt--; |
||
| 117 | } |
||
| 118 | |||
| 119 | /* If the blockSize is not a multiple of 4, compute any remaining output samples here. |
||
| 120 | ** No loop unrolling is used. */ |
||
| 121 | blkCnt = blockSize % 0x4u; |
||
| 122 | |||
| 123 | while(blkCnt > 0u) |
||
| 124 | { |
||
| 125 | /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */ |
||
| 126 | /* Compute Sum of squares of the input samples |
||
| 127 | * and then store the result in a temporary variable, sum. */ |
||
| 128 | in1 = *pSrc++; |
||
| 129 | sumOfSquares = __SMLALD(in1, in1, sumOfSquares); |
||
| 130 | sum += in1; |
||
| 131 | |||
| 132 | /* Decrement the loop counter */ |
||
| 133 | blkCnt--; |
||
| 134 | } |
||
| 135 | |||
| 136 | /* Compute Mean of squares of the input samples |
||
| 137 | * and then store the result in a temporary variable, meanOfSquares. */ |
||
| 138 | meanOfSquares = (q31_t)(sumOfSquares / (q63_t)(blockSize - 1)); |
||
| 139 | |||
| 140 | /* Compute square of mean */ |
||
| 141 | squareOfMean = (q31_t) ((q63_t)sum * sum / (q63_t)(blockSize * (blockSize - 1))); |
||
| 142 | |||
| 143 | /* mean of the squares minus the square of the mean. */ |
||
| 144 | /* Compute standard deviation and store the result to the destination */ |
||
| 145 | arm_sqrt_q15(__SSAT((meanOfSquares - squareOfMean) >> 15, 16u), pResult); |
||
| 146 | |||
| 147 | #else |
||
| 148 | |||
| 149 | /* Run the below code for Cortex-M0 */ |
||
| 150 | q15_t in; /* input value */ |
||
| 151 | |||
| 152 | if(blockSize == 1) |
||
| 153 | { |
||
| 154 | *pResult = 0; |
||
| 155 | return; |
||
| 156 | } |
||
| 157 | |||
| 158 | /* Loop over blockSize number of values */ |
||
| 159 | blkCnt = blockSize; |
||
| 160 | |||
| 161 | while(blkCnt > 0u) |
||
| 162 | { |
||
| 163 | /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */ |
||
| 164 | /* Compute Sum of squares of the input samples |
||
| 165 | * and then store the result in a temporary variable, sumOfSquares. */ |
||
| 166 | in = *pSrc++; |
||
| 167 | sumOfSquares += (in * in); |
||
| 168 | |||
| 169 | /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) */ |
||
| 170 | /* Compute sum of all input values and then store the result in a temporary variable, sum. */ |
||
| 171 | sum += in; |
||
| 172 | |||
| 173 | /* Decrement the loop counter */ |
||
| 174 | blkCnt--; |
||
| 175 | } |
||
| 176 | |||
| 177 | /* Compute Mean of squares of the input samples |
||
| 178 | * and then store the result in a temporary variable, meanOfSquares. */ |
||
| 179 | meanOfSquares = (q31_t)(sumOfSquares / (q63_t)(blockSize - 1)); |
||
| 180 | |||
| 181 | /* Compute square of mean */ |
||
| 182 | squareOfMean = (q31_t) ((q63_t)sum * sum / (q63_t)(blockSize * (blockSize - 1))); |
||
| 183 | |||
| 184 | /* mean of the squares minus the square of the mean. */ |
||
| 185 | /* Compute standard deviation and store the result to the destination */ |
||
| 186 | arm_sqrt_q15(__SSAT((meanOfSquares - squareOfMean) >> 15, 16u), pResult); |
||
| 187 | |||
| 188 | #endif /* #ifndef ARM_MATH_CM0_FAMILY */ |
||
| 189 | |||
| 190 | |||
| 191 | } |
||
| 192 | |||
| 193 | /** |
||
| 194 | * @} end of STD group |
||
| 195 | */ |