mbed official / mbed-dsp

CMSIS DSP library

Dependents:   performance_timer Surfboard_ gps2rtty Capstone ... more

Legacy Warning

This is an mbed 2 library. To learn more about mbed OS 5, visit the docs.

cmsis_dsp/ComplexMathFunctions/arm_cmplx_mult_real_q15.c@1:fdd22bb7aa52, 2012-11-28 (annotated)

Committer:
emilmont
Date:
Wed Nov 28 12:30:09 2012 +0000
Revision:
1:fdd22bb7aa52
Child:
2:da51fb522205
DSP library code

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 1:fdd22bb7aa52 5 * $Revision: V1.1.0
emilmont 1:fdd22bb7aa52 6 *
emilmont 1:fdd22bb7aa52 7 * Project: CMSIS DSP Library
emilmont 1:fdd22bb7aa52 8 * Title: arm_cmplx_mult_real_q15.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 1:fdd22bb7aa52 10 * Description: Q15 complex by real multiplication
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 groupCmplxMath
emilmont 1:fdd22bb7aa52 37 */
emilmont 1:fdd22bb7aa52 38
emilmont 1:fdd22bb7aa52 39 /**
emilmont 1:fdd22bb7aa52 40 * @addtogroup CmplxByRealMult
emilmont 1:fdd22bb7aa52 41 * @{
emilmont 1:fdd22bb7aa52 42 */
emilmont 1:fdd22bb7aa52 43
emilmont 1:fdd22bb7aa52 44
emilmont 1:fdd22bb7aa52 45 /**
emilmont 1:fdd22bb7aa52 46 * @brief Q15 complex-by-real multiplication
emilmont 1:fdd22bb7aa52 47 * @param[in] *pSrcCmplx points to the complex input vector
emilmont 1:fdd22bb7aa52 48 * @param[in] *pSrcReal points to the real input vector
emilmont 1:fdd22bb7aa52 49 * @param[out] *pCmplxDst points to the complex output vector
emilmont 1:fdd22bb7aa52 50 * @param[in] numSamples number of samples in each vector
emilmont 1:fdd22bb7aa52 51 * @return none.
emilmont 1:fdd22bb7aa52 52 *
emilmont 1:fdd22bb7aa52 53 * <b>Scaling and Overflow Behavior:</b>
emilmont 1:fdd22bb7aa52 54 * \par
emilmont 1:fdd22bb7aa52 55 * The function uses saturating arithmetic.
emilmont 1:fdd22bb7aa52 56 * Results outside of the allowable Q15 range [0x8000 0x7FFF] will be saturated.
emilmont 1:fdd22bb7aa52 57 */
emilmont 1:fdd22bb7aa52 58
emilmont 1:fdd22bb7aa52 59 void arm_cmplx_mult_real_q15(
emilmont 1:fdd22bb7aa52 60 q15_t * pSrcCmplx,
emilmont 1:fdd22bb7aa52 61 q15_t * pSrcReal,
emilmont 1:fdd22bb7aa52 62 q15_t * pCmplxDst,
emilmont 1:fdd22bb7aa52 63 uint32_t numSamples)
emilmont 1:fdd22bb7aa52 64 {
emilmont 1:fdd22bb7aa52 65 q15_t in; /* Temporary variable to store input value */
emilmont 1:fdd22bb7aa52 66
emilmont 1:fdd22bb7aa52 67 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 68
emilmont 1:fdd22bb7aa52 69 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 70 uint32_t blkCnt; /* loop counters */
emilmont 1:fdd22bb7aa52 71 q31_t inA1, inA2; /* Temporary variables to hold input data */
emilmont 1:fdd22bb7aa52 72 q31_t inB1; /* Temporary variables to hold input data */
emilmont 1:fdd22bb7aa52 73 q15_t out1, out2, out3, out4; /* Temporary variables to hold output data */
emilmont 1:fdd22bb7aa52 74 q31_t mul1, mul2, mul3, mul4; /* Temporary variables to hold intermediate data */
emilmont 1:fdd22bb7aa52 75
emilmont 1:fdd22bb7aa52 76 /* loop Unrolling */
emilmont 1:fdd22bb7aa52 77 blkCnt = numSamples >> 2u;
emilmont 1:fdd22bb7aa52 78
emilmont 1:fdd22bb7aa52 79 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
emilmont 1:fdd22bb7aa52 80 ** a second loop below computes the remaining 1 to 3 samples. */
emilmont 1:fdd22bb7aa52 81 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 82 {
emilmont 1:fdd22bb7aa52 83 /* C[2 * i] = A[2 * i] * B[i]. */
emilmont 1:fdd22bb7aa52 84 /* C[2 * i + 1] = A[2 * i + 1] * B[i]. */
emilmont 1:fdd22bb7aa52 85 /* read complex number both real and imaginary from complex input buffer */
emilmont 1:fdd22bb7aa52 86 inA1 = *__SIMD32(pSrcCmplx)++;
emilmont 1:fdd22bb7aa52 87 /* read two real values at a time from real input buffer */
emilmont 1:fdd22bb7aa52 88 inB1 = *__SIMD32(pSrcReal)++;
emilmont 1:fdd22bb7aa52 89 /* read complex number both real and imaginary from complex input buffer */
emilmont 1:fdd22bb7aa52 90 inA2 = *__SIMD32(pSrcCmplx)++;
emilmont 1:fdd22bb7aa52 91
emilmont 1:fdd22bb7aa52 92 /* multiply complex number with real numbers */
emilmont 1:fdd22bb7aa52 93 #ifndef ARM_MATH_BIG_ENDIAN
emilmont 1:fdd22bb7aa52 94
emilmont 1:fdd22bb7aa52 95 mul1 = (q31_t) ((q15_t) (inA1) * (q15_t) (inB1));
emilmont 1:fdd22bb7aa52 96 mul2 = (q31_t) ((q15_t) (inA1 >> 16) * (q15_t) (inB1));
emilmont 1:fdd22bb7aa52 97 mul3 = (q31_t) ((q15_t) (inA2) * (q15_t) (inB1 >> 16));
emilmont 1:fdd22bb7aa52 98 mul4 = (q31_t) ((q15_t) (inA2 >> 16) * (q15_t) (inB1 >> 16));
emilmont 1:fdd22bb7aa52 99
emilmont 1:fdd22bb7aa52 100 #else
emilmont 1:fdd22bb7aa52 101
emilmont 1:fdd22bb7aa52 102 mul2 = (q31_t) ((q15_t) (inA1 >> 16) * (q15_t) (inB1 >> 16));
emilmont 1:fdd22bb7aa52 103 mul1 = (q31_t) ((q15_t) inA1 * (q15_t) (inB1 >> 16));
emilmont 1:fdd22bb7aa52 104 mul4 = (q31_t) ((q15_t) (inA2 >> 16) * (q15_t) inB1);
emilmont 1:fdd22bb7aa52 105 mul3 = (q31_t) ((q15_t) inA2 * (q15_t) inB1);
emilmont 1:fdd22bb7aa52 106
emilmont 1:fdd22bb7aa52 107 #endif // #ifndef ARM_MATH_BIG_ENDIAN
emilmont 1:fdd22bb7aa52 108
emilmont 1:fdd22bb7aa52 109 /* saturate the result */
emilmont 1:fdd22bb7aa52 110 out1 = (q15_t) __SSAT(mul1 >> 15u, 16);
emilmont 1:fdd22bb7aa52 111 out2 = (q15_t) __SSAT(mul2 >> 15u, 16);
emilmont 1:fdd22bb7aa52 112 out3 = (q15_t) __SSAT(mul3 >> 15u, 16);
emilmont 1:fdd22bb7aa52 113 out4 = (q15_t) __SSAT(mul4 >> 15u, 16);
emilmont 1:fdd22bb7aa52 114
emilmont 1:fdd22bb7aa52 115 /* pack real and imaginary outputs and store them to destination */
emilmont 1:fdd22bb7aa52 116 *__SIMD32(pCmplxDst)++ = __PKHBT(out1, out2, 16);
emilmont 1:fdd22bb7aa52 117 *__SIMD32(pCmplxDst)++ = __PKHBT(out3, out4, 16);
emilmont 1:fdd22bb7aa52 118
emilmont 1:fdd22bb7aa52 119 inA1 = *__SIMD32(pSrcCmplx)++;
emilmont 1:fdd22bb7aa52 120 inB1 = *__SIMD32(pSrcReal)++;
emilmont 1:fdd22bb7aa52 121 inA2 = *__SIMD32(pSrcCmplx)++;
emilmont 1:fdd22bb7aa52 122
emilmont 1:fdd22bb7aa52 123 #ifndef ARM_MATH_BIG_ENDIAN
emilmont 1:fdd22bb7aa52 124
emilmont 1:fdd22bb7aa52 125 mul1 = (q31_t) ((q15_t) (inA1) * (q15_t) (inB1));
emilmont 1:fdd22bb7aa52 126 mul2 = (q31_t) ((q15_t) (inA1 >> 16) * (q15_t) (inB1));
emilmont 1:fdd22bb7aa52 127 mul3 = (q31_t) ((q15_t) (inA2) * (q15_t) (inB1 >> 16));
emilmont 1:fdd22bb7aa52 128 mul4 = (q31_t) ((q15_t) (inA2 >> 16) * (q15_t) (inB1 >> 16));
emilmont 1:fdd22bb7aa52 129
emilmont 1:fdd22bb7aa52 130 #else
emilmont 1:fdd22bb7aa52 131
emilmont 1:fdd22bb7aa52 132 mul2 = (q31_t) ((q15_t) (inA1 >> 16) * (q15_t) (inB1 >> 16));
emilmont 1:fdd22bb7aa52 133 mul1 = (q31_t) ((q15_t) inA1 * (q15_t) (inB1 >> 16));
emilmont 1:fdd22bb7aa52 134 mul4 = (q31_t) ((q15_t) (inA2 >> 16) * (q15_t) inB1);
emilmont 1:fdd22bb7aa52 135 mul3 = (q31_t) ((q15_t) inA2 * (q15_t) inB1);
emilmont 1:fdd22bb7aa52 136
emilmont 1:fdd22bb7aa52 137 #endif // #ifndef ARM_MATH_BIG_ENDIAN
emilmont 1:fdd22bb7aa52 138
emilmont 1:fdd22bb7aa52 139 out1 = (q15_t) __SSAT(mul1 >> 15u, 16);
emilmont 1:fdd22bb7aa52 140 out2 = (q15_t) __SSAT(mul2 >> 15u, 16);
emilmont 1:fdd22bb7aa52 141 out3 = (q15_t) __SSAT(mul3 >> 15u, 16);
emilmont 1:fdd22bb7aa52 142 out4 = (q15_t) __SSAT(mul4 >> 15u, 16);
emilmont 1:fdd22bb7aa52 143
emilmont 1:fdd22bb7aa52 144 *__SIMD32(pCmplxDst)++ = __PKHBT(out1, out2, 16);
emilmont 1:fdd22bb7aa52 145 *__SIMD32(pCmplxDst)++ = __PKHBT(out3, out4, 16);
emilmont 1:fdd22bb7aa52 146
emilmont 1:fdd22bb7aa52 147 /* Decrement the numSamples loop counter */
emilmont 1:fdd22bb7aa52 148 blkCnt--;
emilmont 1:fdd22bb7aa52 149 }
emilmont 1:fdd22bb7aa52 150
emilmont 1:fdd22bb7aa52 151 /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
emilmont 1:fdd22bb7aa52 152 ** No loop unrolling is used. */
emilmont 1:fdd22bb7aa52 153 blkCnt = numSamples % 0x4u;
emilmont 1:fdd22bb7aa52 154
emilmont 1:fdd22bb7aa52 155 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 156 {
emilmont 1:fdd22bb7aa52 157 /* C[2 * i] = A[2 * i] * B[i]. */
emilmont 1:fdd22bb7aa52 158 /* C[2 * i + 1] = A[2 * i + 1] * B[i]. */
emilmont 1:fdd22bb7aa52 159 in = *pSrcReal++;
emilmont 1:fdd22bb7aa52 160 /* store the result in the destination buffer. */
emilmont 1:fdd22bb7aa52 161 *pCmplxDst++ =
emilmont 1:fdd22bb7aa52 162 (q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
emilmont 1:fdd22bb7aa52 163 *pCmplxDst++ =
emilmont 1:fdd22bb7aa52 164 (q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
emilmont 1:fdd22bb7aa52 165
emilmont 1:fdd22bb7aa52 166 /* Decrement the numSamples loop counter */
emilmont 1:fdd22bb7aa52 167 blkCnt--;
emilmont 1:fdd22bb7aa52 168 }
emilmont 1:fdd22bb7aa52 169
emilmont 1:fdd22bb7aa52 170 #else
emilmont 1:fdd22bb7aa52 171
emilmont 1:fdd22bb7aa52 172 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 173
emilmont 1:fdd22bb7aa52 174 while(numSamples > 0u)
emilmont 1:fdd22bb7aa52 175 {
emilmont 1:fdd22bb7aa52 176 /* realOut = realA * realB. */
emilmont 1:fdd22bb7aa52 177 /* imagOut = imagA * realB. */
emilmont 1:fdd22bb7aa52 178 in = *pSrcReal++;
emilmont 1:fdd22bb7aa52 179 /* store the result in the destination buffer. */
emilmont 1:fdd22bb7aa52 180 *pCmplxDst++ =
emilmont 1:fdd22bb7aa52 181 (q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
emilmont 1:fdd22bb7aa52 182 *pCmplxDst++ =
emilmont 1:fdd22bb7aa52 183 (q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
emilmont 1:fdd22bb7aa52 184
emilmont 1:fdd22bb7aa52 185 /* Decrement the numSamples loop counter */
emilmont 1:fdd22bb7aa52 186 numSamples--;
emilmont 1:fdd22bb7aa52 187 }
emilmont 1:fdd22bb7aa52 188
emilmont 1:fdd22bb7aa52 189 #endif /* #ifndef ARM_MATH_CM0 */
emilmont 1:fdd22bb7aa52 190
emilmont 1:fdd22bb7aa52 191 }
emilmont 1:fdd22bb7aa52 192
emilmont 1:fdd22bb7aa52 193 /**
emilmont 1:fdd22bb7aa52 194 * @} end of CmplxByRealMult group
emilmont 1:fdd22bb7aa52 195 */