1 /* ----------------------------------------------------------------------
2 * Copyright (C) 2010-2013 ARM Limited. All rights reserved.
4 * $Date: 17. January 2013
7 * Project: CMSIS DSP Library
10 * Description: Root Mean Square of the elements of a Q15 vector.
12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
14 * Redistribution and use in source and binary forms, with or without
15 * modification, are permitted provided that the following conditions
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
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.
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 * ---------------------------------------------------------------------------- */
49 * @brief Root Mean Square of the elements of a Q15 vector.
50 * @param[in] *pSrc points to the input vector
51 * @param[in] blockSize length of the input vector
52 * @param[out] *pResult rms value returned here
56 * <b>Scaling and Overflow Behavior:</b>
59 * The function is implemented using a 64-bit internal accumulator.
60 * The input is represented in 1.15 format.
61 * Intermediate multiplication yields a 2.30 format, and this
62 * result is added without saturation to a 64-bit accumulator in 34.30 format.
63 * With 33 guard bits in the accumulator, there is no risk of overflow, and the
64 * full precision of the intermediate multiplication is preserved.
65 * Finally, the 34.30 result is truncated to 34.15 format by discarding the lower
66 * 15 bits, and then saturated to yield a result in 1.15 format.
75 q63_t sum = 0; /* accumulator */
77 #ifndef ARM_MATH_CM0_FAMILY
79 /* Run the below code for Cortex-M4 and Cortex-M3 */
81 q31_t in; /* temporary variable to store the input value */
82 q15_t in1; /* temporary variable to store the input value */
83 uint32_t blkCnt; /* loop counter */
86 blkCnt = blockSize >> 2u;
88 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
89 ** a second loop below computes the remaining 1 to 3 samples. */
92 /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */
93 /* Compute sum of the squares and then store the results in a temporary variable, sum */
94 in = *__SIMD32(pSrc)++;
95 sum = __SMLALD(in, in, sum);
96 in = *__SIMD32(pSrc)++;
97 sum = __SMLALD(in, in, sum);
99 /* Decrement the loop counter */
103 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
104 ** No loop unrolling is used. */
105 blkCnt = blockSize % 0x4u;
109 /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */
110 /* Compute sum of the squares and then store the results in a temporary variable, sum */
112 sum = __SMLALD(in1, in1, sum);
114 /* Decrement the loop counter */
118 /* Truncating and saturating the accumulator to 1.15 format */
119 in = (q31_t)(sum >> 15);
121 in1 = __SSAT(in / blockSize, 16);
123 /* Store the result in the destination */
124 arm_sqrt_q15(in1, pResult);
128 /* Run the below code for Cortex-M0 */
130 q15_t in; /* temporary variable to store the input value */
131 q31_t tmp; /* temporary variable to store the input value */
132 uint32_t blkCnt; /* loop counter */
134 /* Loop over blockSize number of values */
139 /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */
140 /* Compute sum of the squares and then store the results in a temporary variable, sum */
142 sum += ((q31_t) in * in);
144 /* Decrement the loop counter */
148 /* Truncating and saturating the accumulator to 1.15 format */
149 tmp = (q31_t)(sum >> 15);
151 in = __SSAT(tmp / blockSize, 16);
153 /* Store the result in the destination */
154 arm_sqrt_q15(in, pResult);
156 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
161 * @} end of RMS group