1 /* ----------------------------------------------------------------------
2 * Project: CMSIS DSP Library
3 * Title: arm_rms_f16.c
4 * Description: Root mean square value of the elements of a floating-point vector
5 *
6 * $Date: 23 April 2021
7 * $Revision: V1.9.0
8 *
9 * Target Processor: Cortex-M and Cortex-A cores
10 * -------------------------------------------------------------------- */
11 /*
12 * Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
13 *
14 * SPDX-License-Identifier: Apache-2.0
15 *
16 * Licensed under the Apache License, Version 2.0 (the License); you may
17 * not use this file except in compliance with the License.
18 * You may obtain a copy of the License at
19 *
20 * www.apache.org/licenses/LICENSE-2.0
21 *
22 * Unless required by applicable law or agreed to in writing, software
23 * distributed under the License is distributed on an AS IS BASIS, WITHOUT
24 * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
25 * See the License for the specific language governing permissions and
26 * limitations under the License.
27 */
28
29 #include "dsp/statistics_functions_f16.h"
30
31 #if defined(ARM_FLOAT16_SUPPORTED)
32
33
34 /**
35 @ingroup groupStats
36 */
37
38
39
40 /**
41 @addtogroup RMS
42 @{
43 */
44
45 /**
46 @brief Root Mean Square of the elements of a floating-point vector.
47 @param[in] pSrc points to the input vector
48 @param[in] blockSize number of samples in input vector
49 @param[out] pResult root mean square value returned here
50 @return none
51 */
52
53 #if defined(ARM_MATH_MVE_FLOAT16) && !defined(ARM_MATH_AUTOVECTORIZE)
54
arm_rms_f16(const float16_t * pSrc,uint32_t blockSize,float16_t * pResult)55 void arm_rms_f16(
56 const float16_t * pSrc,
57 uint32_t blockSize,
58 float16_t * pResult)
59 {
60 float16_t pow = 0.0f;
61
62 arm_power_f16(pSrc, blockSize, &pow);
63
64 /* Compute Rms and store the result in the destination */
65 arm_sqrt_f16((_Float16)pow / (_Float16) blockSize, pResult);
66 }
67 #else
68
arm_rms_f16(const float16_t * pSrc,uint32_t blockSize,float16_t * pResult)69 void arm_rms_f16(
70 const float16_t * pSrc,
71 uint32_t blockSize,
72 float16_t * pResult)
73 {
74 uint32_t blkCnt; /* Loop counter */
75 _Float16 sum = 0.0f16; /* Temporary result storage */
76 _Float16 in; /* Temporary variable to store input value */
77
78 #if defined (ARM_MATH_LOOPUNROLL) && !defined(ARM_MATH_AUTOVECTORIZE)
79
80 /* Loop unrolling: Compute 4 outputs at a time */
81 blkCnt = blockSize >> 2U;
82
83 while (blkCnt > 0U)
84 {
85 /* C = A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1] */
86
87 in = *pSrc++;
88 /* Compute sum of squares and store result in a temporary variable, sum. */
89 sum += in * in;
90
91 in = *pSrc++;
92 sum += in * in;
93
94 in = *pSrc++;
95 sum += in * in;
96
97 in = *pSrc++;
98 sum += in * in;
99
100 /* Decrement loop counter */
101 blkCnt--;
102 }
103
104 /* Loop unrolling: Compute remaining outputs */
105 blkCnt = blockSize % 0x4U;
106
107 #else
108
109 /* Initialize blkCnt with number of samples */
110 blkCnt = blockSize;
111
112 #endif /* #if defined (ARM_MATH_LOOPUNROLL) */
113
114 while (blkCnt > 0U)
115 {
116 /* C = A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1] */
117
118 in = *pSrc++;
119 /* Compute sum of squares and store result in a temporary variable. */
120 sum += ( in * in);
121
122 /* Decrement loop counter */
123 blkCnt--;
124 }
125
126 /* Compute Rms and store result in destination */
127 arm_sqrt_f16((_Float16)sum / (_Float16) blockSize, pResult);
128 }
129 #endif /* defined(ARM_MATH_MVEF) && !defined(ARM_MATH_AUTOVECTORIZE) */
130
131 /**
132 @} end of RMS group
133 */
134
135 #endif /* #if defined(ARM_FLOAT16_SUPPORTED) */
136
137