| 提交 | 用户 | age | ||
| 8b51c7 | 1 | /* ---------------------------------------------------------------------- |
| Q | 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_cmplx_mult_cmplx_f32.c | |
| 9 | * | |
| 10 | * Description: Floating-point complex-by-complex 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 | #include "arm_math.h" | |
| 41 | ||
| 42 | /** | |
| 43 | * @ingroup groupCmplxMath | |
| 44 | */ | |
| 45 | ||
| 46 | /** | |
| 47 | * @defgroup CmplxByCmplxMult Complex-by-Complex Multiplication | |
| 48 | * | |
| 49 | * Multiplies a complex vector by another complex vector and generates a complex result. | |
| 50 | * The data in the complex arrays is stored in an interleaved fashion | |
| 51 | * (real, imag, real, imag, ...). | |
| 52 | * The parameter <code>numSamples</code> represents the number of complex | |
| 53 | * samples processed. The complex arrays have a total of <code>2*numSamples</code> | |
| 54 | * real values. | |
| 55 | * | |
| 56 | * The underlying algorithm is used: | |
| 57 | * | |
| 58 | * <pre> | |
| 59 | * for(n=0; n<numSamples; n++) { | |
| 60 | * pDst[(2*n)+0] = pSrcA[(2*n)+0] * pSrcB[(2*n)+0] - pSrcA[(2*n)+1] * pSrcB[(2*n)+1]; | |
| 61 | * pDst[(2*n)+1] = pSrcA[(2*n)+0] * pSrcB[(2*n)+1] + pSrcA[(2*n)+1] * pSrcB[(2*n)+0]; | |
| 62 | * } | |
| 63 | * </pre> | |
| 64 | * | |
| 65 | * There are separate functions for floating-point, Q15, and Q31 data types. | |
| 66 | */ | |
| 67 | ||
| 68 | /** | |
| 69 | * @addtogroup CmplxByCmplxMult | |
| 70 | * @{ | |
| 71 | */ | |
| 72 | ||
| 73 | ||
| 74 | /** | |
| 75 | * @brief Floating-point complex-by-complex multiplication | |
| 76 | * @param[in] *pSrcA points to the first input vector | |
| 77 | * @param[in] *pSrcB points to the second input vector | |
| 78 | * @param[out] *pDst points to the output vector | |
| 79 | * @param[in] numSamples number of complex samples in each vector | |
| 80 | * @return none. | |
| 81 | */ | |
| 82 | ||
| 83 | void arm_cmplx_mult_cmplx_f32( | |
| 84 | float32_t * pSrcA, | |
| 85 | float32_t * pSrcB, | |
| 86 | float32_t * pDst, | |
| 87 | uint32_t numSamples) | |
| 88 | { | |
| 89 | float32_t a1, b1, c1, d1; /* Temporary variables to store real and imaginary values */ | |
| 90 | uint32_t blkCnt; /* loop counters */ | |
| 91 | ||
| 92 | #ifndef ARM_MATH_CM0_FAMILY | |
| 93 | ||
| 94 | /* Run the below code for Cortex-M4 and Cortex-M3 */ | |
| 95 | float32_t a2, b2, c2, d2; /* Temporary variables to store real and imaginary values */ | |
| 96 | float32_t acc1, acc2, acc3, acc4; | |
| 97 | ||
| 98 | ||
| 99 | /* loop Unrolling */ | |
| 100 | blkCnt = numSamples >> 2u; | |
| 101 | ||
| 102 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. | |
| 103 | ** a second loop below computes the remaining 1 to 3 samples. */ | |
| 104 | while(blkCnt > 0u) | |
| 105 | { | |
| 106 | /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */ | |
| 107 | /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */ | |
| 108 | a1 = *pSrcA; /* A[2 * i] */ | |
| 109 | c1 = *pSrcB; /* B[2 * i] */ | |
| 110 | ||
| 111 | b1 = *(pSrcA + 1); /* A[2 * i + 1] */ | |
| 112 | acc1 = a1 * c1; /* acc1 = A[2 * i] * B[2 * i] */ | |
| 113 | ||
| 114 | a2 = *(pSrcA + 2); /* A[2 * i + 2] */ | |
| 115 | acc2 = (b1 * c1); /* acc2 = A[2 * i + 1] * B[2 * i] */ | |
| 116 | ||
| 117 | d1 = *(pSrcB + 1); /* B[2 * i + 1] */ | |
| 118 | c2 = *(pSrcB + 2); /* B[2 * i + 2] */ | |
| 119 | acc1 -= b1 * d1; /* acc1 = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1] */ | |
| 120 | ||
| 121 | d2 = *(pSrcB + 3); /* B[2 * i + 3] */ | |
| 122 | acc3 = a2 * c2; /* acc3 = A[2 * i + 2] * B[2 * i + 2] */ | |
| 123 | ||
| 124 | b2 = *(pSrcA + 3); /* A[2 * i + 3] */ | |
| 125 | acc2 += (a1 * d1); /* acc2 = A[2 * i + 1] * B[2 * i] + A[2 * i] * B[2 * i + 1] */ | |
| 126 | ||
| 127 | a1 = *(pSrcA + 4); /* A[2 * i + 4] */ | |
| 128 | acc4 = (a2 * d2); /* acc4 = A[2 * i + 2] * B[2 * i + 3] */ | |
| 129 | ||
| 130 | c1 = *(pSrcB + 4); /* B[2 * i + 4] */ | |
| 131 | acc3 -= (b2 * d2); /* acc3 = A[2 * i + 2] * B[2 * i + 2] - A[2 * i + 3] * B[2 * i + 3] */ | |
| 132 | *pDst = acc1; /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1] */ | |
| 133 | ||
| 134 | b1 = *(pSrcA + 5); /* A[2 * i + 5] */ | |
| 135 | acc4 += b2 * c2; /* acc4 = A[2 * i + 2] * B[2 * i + 3] + A[2 * i + 3] * B[2 * i + 2] */ | |
| 136 | ||
| 137 | *(pDst + 1) = acc2; /* C[2 * i + 1] = A[2 * i + 1] * B[2 * i] + A[2 * i] * B[2 * i + 1] */ | |
| 138 | acc1 = (a1 * c1); | |
| 139 | ||
| 140 | d1 = *(pSrcB + 5); | |
| 141 | acc2 = (b1 * c1); | |
| 142 | ||
| 143 | *(pDst + 2) = acc3; | |
| 144 | *(pDst + 3) = acc4; | |
| 145 | ||
| 146 | a2 = *(pSrcA + 6); | |
| 147 | acc1 -= (b1 * d1); | |
| 148 | ||
| 149 | c2 = *(pSrcB + 6); | |
| 150 | acc2 += (a1 * d1); | |
| 151 | ||
| 152 | b2 = *(pSrcA + 7); | |
| 153 | acc3 = (a2 * c2); | |
| 154 | ||
| 155 | d2 = *(pSrcB + 7); | |
| 156 | acc4 = (b2 * c2); | |
| 157 | ||
| 158 | *(pDst + 4) = acc1; | |
| 159 | pSrcA += 8u; | |
| 160 | ||
| 161 | acc3 -= (b2 * d2); | |
| 162 | acc4 += (a2 * d2); | |
| 163 | ||
| 164 | *(pDst + 5) = acc2; | |
| 165 | pSrcB += 8u; | |
| 166 | ||
| 167 | *(pDst + 6) = acc3; | |
| 168 | *(pDst + 7) = acc4; | |
| 169 | ||
| 170 | pDst += 8u; | |
| 171 | ||
| 172 | /* Decrement the numSamples loop counter */ | |
| 173 | blkCnt--; | |
| 174 | } | |
| 175 | ||
| 176 | /* If the numSamples is not a multiple of 4, compute any remaining output samples here. | |
| 177 | ** No loop unrolling is used. */ | |
| 178 | blkCnt = numSamples % 0x4u; | |
| 179 | ||
| 180 | #else | |
| 181 | ||
| 182 | /* Run the below code for Cortex-M0 */ | |
| 183 | blkCnt = numSamples; | |
| 184 | ||
| 185 | #endif /* #ifndef ARM_MATH_CM0_FAMILY */ | |
| 186 | ||
| 187 | while(blkCnt > 0u) | |
| 188 | { | |
| 189 | /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */ | |
| 190 | /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */ | |
| 191 | a1 = *pSrcA++; | |
| 192 | b1 = *pSrcA++; | |
| 193 | c1 = *pSrcB++; | |
| 194 | d1 = *pSrcB++; | |
| 195 | ||
| 196 | /* store the result in the destination buffer. */ | |
| 197 | *pDst++ = (a1 * c1) - (b1 * d1); | |
| 198 | *pDst++ = (a1 * d1) + (b1 * c1); | |
| 199 | ||
| 200 | /* Decrement the numSamples loop counter */ | |
| 201 | blkCnt--; | |
| 202 | } | |
| 203 | } | |
| 204 | ||
| 205 | /** | |
| 206 | * @} end of CmplxByCmplxMult group | |
| 207 | */ | |
