tmk_core/tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/ComplexMathFunctions/arm_cmplx_mult_real_q15.c

   1 /* ----------------------------------------------------------------------
   2 * Copyright (C) 2010-2013 ARM Limited. All rights reserved.
   3 *
   4 * $Date:        17. January 2013
   5 * $Revision:    V1.4.1
   6 *
   7 * Project:          CMSIS DSP Library
   8 * Title:            arm_cmplx_mult_real_q15.c
   9 *
  10 * Description:  Q15 complex by real multiplication
  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 groupCmplxMath
  45  */
  46
  47 /**
  48  * @addtogroup CmplxByRealMult
  49  * @{
  50  */
  51
  52
  53 /**
  54  * @brief  Q15 complex-by-real multiplication
  55  * @param[in]  *pSrcCmplx points to the complex input vector
  56  * @param[in]  *pSrcReal points to the real input vector
  57  * @param[out]  *pCmplxDst points to the complex output vector
  58  * @param[in]  numSamples number of samples in each vector
  59  * @return none.
  60  *
  61  * <b>Scaling and Overflow Behavior:</b>
  62  * \par
  63  * The function uses saturating arithmetic.
  64  * Results outside of the allowable Q15 range [0x8000 0x7FFF] will be saturated.
  65  */
  66
  67 void arm_cmplx_mult_real_q15(
  68   q15_t * pSrcCmplx,
  69   q15_t * pSrcReal,
  70   q15_t * pCmplxDst,
  71   uint32_t numSamples)
  72 {
  73   q15_t in;                                      /* Temporary variable to store input value */
  74
  75 #ifndef ARM_MATH_CM0_FAMILY
  76
  77   /* Run the below code for Cortex-M4 and Cortex-M3 */
  78   uint32_t blkCnt;                               /* loop counters */
  79   q31_t inA1, inA2;                              /* Temporary variables to hold input data */
  80   q31_t inB1;                                    /* Temporary variables to hold input data */
  81   q15_t out1, out2, out3, out4;                  /* Temporary variables to hold output data */
  82   q31_t mul1, mul2, mul3, mul4;                  /* Temporary variables to hold intermediate data */
  83
  84   /* loop Unrolling */
  85   blkCnt = numSamples >> 2u;
  86
  87   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
  88    ** a second loop below computes the remaining 1 to 3 samples. */
  89   while(blkCnt > 0u)
  90   {
  91     /* C[2 * i] = A[2 * i] * B[i].            */
  92     /* C[2 * i + 1] = A[2 * i + 1] * B[i].        */
  93     /* read complex number both real and imaginary from complex input buffer */
  94     inA1 = *__SIMD32(pSrcCmplx)++;
  95     /* read two real values at a time from real input buffer */
  96     inB1 = *__SIMD32(pSrcReal)++;
  97     /* read complex number both real and imaginary from complex input buffer */
  98     inA2 = *__SIMD32(pSrcCmplx)++;
  99
 100     /* multiply complex number with real numbers */
 101 #ifndef ARM_MATH_BIG_ENDIAN
 102
 103     mul1 = (q31_t) ((q15_t) (inA1) * (q15_t) (inB1));
 104     mul2 = (q31_t) ((q15_t) (inA1 >> 16) * (q15_t) (inB1));
 105     mul3 = (q31_t) ((q15_t) (inA2) * (q15_t) (inB1 >> 16));
 106     mul4 = (q31_t) ((q15_t) (inA2 >> 16) * (q15_t) (inB1 >> 16));
 107
 108 #else
 109
 110     mul2 = (q31_t) ((q15_t) (inA1 >> 16) * (q15_t) (inB1 >> 16));
 111     mul1 = (q31_t) ((q15_t) inA1 * (q15_t) (inB1 >> 16));
 112     mul4 = (q31_t) ((q15_t) (inA2 >> 16) * (q15_t) inB1);
 113     mul3 = (q31_t) ((q15_t) inA2 * (q15_t) inB1);
 114
 115 #endif //      #ifndef ARM_MATH_BIG_ENDIAN
 116
 117     /* saturate the result */
 118     out1 = (q15_t) __SSAT(mul1 >> 15u, 16);
 119     out2 = (q15_t) __SSAT(mul2 >> 15u, 16);
 120     out3 = (q15_t) __SSAT(mul3 >> 15u, 16);
 121     out4 = (q15_t) __SSAT(mul4 >> 15u, 16);
 122
 123     /* pack real and imaginary outputs and store them to destination */
 124     *__SIMD32(pCmplxDst)++ = __PKHBT(out1, out2, 16);
 125     *__SIMD32(pCmplxDst)++ = __PKHBT(out3, out4, 16);
 126
 127     inA1 = *__SIMD32(pSrcCmplx)++;
 128     inB1 = *__SIMD32(pSrcReal)++;
 129     inA2 = *__SIMD32(pSrcCmplx)++;
 130
 131 #ifndef ARM_MATH_BIG_ENDIAN
 132
 133     mul1 = (q31_t) ((q15_t) (inA1) * (q15_t) (inB1));
 134     mul2 = (q31_t) ((q15_t) (inA1 >> 16) * (q15_t) (inB1));
 135     mul3 = (q31_t) ((q15_t) (inA2) * (q15_t) (inB1 >> 16));
 136     mul4 = (q31_t) ((q15_t) (inA2 >> 16) * (q15_t) (inB1 >> 16));
 137
 138 #else
 139
 140     mul2 = (q31_t) ((q15_t) (inA1 >> 16) * (q15_t) (inB1 >> 16));
 141     mul1 = (q31_t) ((q15_t) inA1 * (q15_t) (inB1 >> 16));
 142     mul4 = (q31_t) ((q15_t) (inA2 >> 16) * (q15_t) inB1);
 143     mul3 = (q31_t) ((q15_t) inA2 * (q15_t) inB1);
 144
 145 #endif //      #ifndef ARM_MATH_BIG_ENDIAN
 146
 147     out1 = (q15_t) __SSAT(mul1 >> 15u, 16);
 148     out2 = (q15_t) __SSAT(mul2 >> 15u, 16);
 149     out3 = (q15_t) __SSAT(mul3 >> 15u, 16);
 150     out4 = (q15_t) __SSAT(mul4 >> 15u, 16);
 151
 152     *__SIMD32(pCmplxDst)++ = __PKHBT(out1, out2, 16);
 153     *__SIMD32(pCmplxDst)++ = __PKHBT(out3, out4, 16);
 154
 155     /* Decrement the numSamples loop counter */
 156     blkCnt--;
 157   }
 158
 159   /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
 160    ** No loop unrolling is used. */
 161   blkCnt = numSamples % 0x4u;
 162
 163   while(blkCnt > 0u)
 164   {
 165     /* C[2 * i] = A[2 * i] * B[i].            */
 166     /* C[2 * i + 1] = A[2 * i + 1] * B[i].        */
 167     in = *pSrcReal++;
 168     /* store the result in the destination buffer. */
 169     *pCmplxDst++ =
 170       (q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
 171     *pCmplxDst++ =
 172       (q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
 173
 174     /* Decrement the numSamples loop counter */
 175     blkCnt--;
 176   }
 177
 178 #else
 179
 180   /* Run the below code for Cortex-M0 */
 181
 182   while(numSamples > 0u)
 183   {
 184     /* realOut = realA * realB.            */
 185     /* imagOut = imagA * realB.                */
 186     in = *pSrcReal++;
 187     /* store the result in the destination buffer. */
 188     *pCmplxDst++ =
 189       (q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
 190     *pCmplxDst++ =
 191       (q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
 192
 193     /* Decrement the numSamples loop counter */
 194     numSamples--;
 195   }
 196
 197 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
 198
 199 }
 200
 201 /**
 202  * @} end of CmplxByRealMult group
 203  */