mirror of
https://github.com/esp8266/Arduino.git
synced 2025-06-20 21:01:25 +03:00
[sam] Adding CMSIS 2.10
This commit is contained in:
Thibaut VIARD
@ -0,0 +1,141 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_conj_f32.c
|
||||
*
|
||||
* Description: Floating-point complex conjugate.
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* ---------------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @defgroup cmplx_conj Complex Conjugate
|
||||
*
|
||||
* Conjugates the elements of a complex data vector.
|
||||
*
|
||||
* The <code>pSrc</code> points to the source data and
|
||||
* <code>pDst</code> points to the where the result should be written.
|
||||
* <code>numSamples</code> specifies the number of complex samples
|
||||
* and the data in each array is stored in an interleaved fashion
|
||||
* (real, imag, real, imag, ...).
|
||||
* Each array has a total of <code>2*numSamples</code> values.
|
||||
* The underlying algorithm is used:
|
||||
*
|
||||
* <pre>
|
||||
* for(n=0; n<numSamples; n++) {
|
||||
* pDst[(2*n)+0)] = pSrc[(2*n)+0]; // real part
|
||||
* pDst[(2*n)+1)] = -pSrc[(2*n)+1]; // imag part
|
||||
* }
|
||||
* </pre>
|
||||
*
|
||||
* There are separate functions for floating-point, Q15, and Q31 data types.
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_conj
|
||||
* @{
|
||||
*/
|
||||
|
||||
/**
|
||||
* @brief Floating-point complex conjugate.
|
||||
* @param *pSrc points to the input vector
|
||||
* @param *pDst points to the output vector
|
||||
* @param numSamples number of complex samples in each vector
|
||||
* @return none.
|
||||
*/
|
||||
|
||||
void arm_cmplx_conj_f32(
|
||||
float32_t * pSrc,
|
||||
float32_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0]+jC[1] = A[0]+ j (-1) A[1] */
|
||||
/* Calculate Complex Conjugate and then store the results in the destination buffer. */
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = -*pSrc++;
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = -*pSrc++;
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = -*pSrc++;
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = -*pSrc++;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0]+jC[1] = A[0]+ j (-1) A[1] */
|
||||
/* Calculate Complex Conjugate and then store the results in the destination buffer. */
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = -*pSrc++;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* realOut + j (imagOut) = realIn + j (-1) imagIn */
|
||||
/* Calculate Complex Conjugate and then store the results in the destination buffer. */
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = -*pSrc++;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_conj group
|
||||
*/
|
||||
@ -0,0 +1,123 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_conj_q15.c
|
||||
*
|
||||
* Description: Q15 complex conjugate.
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* ---------------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_conj
|
||||
* @{
|
||||
*/
|
||||
|
||||
/**
|
||||
* @brief Q15 complex conjugate.
|
||||
* @param *pSrc points to the input vector
|
||||
* @param *pDst points to the output vector
|
||||
* @param numSamples number of complex samples in each vector
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function uses saturating arithmetic.
|
||||
* The Q15 value -1 (0x8000) will be saturated to the maximum allowable positive value 0x7FFF.
|
||||
*/
|
||||
|
||||
void arm_cmplx_conj_q15(
|
||||
q15_t * pSrc,
|
||||
q15_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0]+jC[1] = A[0]+ j (-1) A[1] */
|
||||
/* Calculate Complex Conjugate and then store the results in the destination buffer. */
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = __SSAT(-*pSrc++, 16);
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = __SSAT(-*pSrc++, 16);
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = __SSAT(-*pSrc++, 16);
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = __SSAT(-*pSrc++, 16);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0]+jC[1] = A[0]+ j (-1) A[1] */
|
||||
/* Calculate Complex Conjugate and then store the results in the destination buffer. */
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = __SSAT(-*pSrc++, 16);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* realOut + j (imagOut) = realIn+ j (-1) imagIn */
|
||||
/* Calculate Complex Conjugate and then store the results in the destination buffer. */
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = -*pSrc++;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_conj group
|
||||
*/
|
||||
@ -0,0 +1,131 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_conj_q31.c
|
||||
*
|
||||
* Description: Q31 complex conjugate.
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* ---------------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_conj
|
||||
* @{
|
||||
*/
|
||||
|
||||
/**
|
||||
* @brief Q31 complex conjugate.
|
||||
* @param *pSrc points to the input vector
|
||||
* @param *pDst points to the output vector
|
||||
* @param numSamples number of complex samples in each vector
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function uses saturating arithmetic.
|
||||
* The Q31 value -1 (0x80000000) will be saturated to the maximum allowable positive value 0x7FFFFFFF.
|
||||
*/
|
||||
|
||||
void arm_cmplx_conj_q31(
|
||||
q31_t * pSrc,
|
||||
q31_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
q31_t in; /* Input value */
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0]+jC[1] = A[0]+ j (-1) A[1] */
|
||||
/* Calculate Complex Conjugate and then store the results in the destination buffer. */
|
||||
/* Saturated to 0x7fffffff if the input is -1(0x80000000) */
|
||||
*pDst++ = *pSrc++;
|
||||
in = *pSrc++;
|
||||
*pDst++ = (in == 0x80000000) ? 0x7fffffff : -in;
|
||||
*pDst++ = *pSrc++;
|
||||
in = *pSrc++;
|
||||
*pDst++ = (in == 0x80000000) ? 0x7fffffff : -in;
|
||||
*pDst++ = *pSrc++;
|
||||
in = *pSrc++;
|
||||
*pDst++ = (in == 0x80000000) ? 0x7fffffff : -in;
|
||||
*pDst++ = *pSrc++;
|
||||
in = *pSrc++;
|
||||
*pDst++ = (in == 0x80000000) ? 0x7fffffff : -in;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0]+jC[1] = A[0]+ j (-1) A[1] */
|
||||
/* Calculate Complex Conjugate and then store the results in the destination buffer. */
|
||||
/* Saturated to 0x7fffffff if the input is -1(0x80000000) */
|
||||
*pDst++ = *pSrc++;
|
||||
in = *pSrc++;
|
||||
*pDst++ = (in == 0x80000000) ? 0x7fffffff : -in;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* realOut + j (imagOut) = realIn+ j (-1) imagIn */
|
||||
/* Calculate Complex Conjugate and then store the results in the destination buffer. */
|
||||
*pDst++ = *pSrc++;
|
||||
*pDst++ = -*pSrc++;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_conj group
|
||||
*/
|
||||
@ -0,0 +1,157 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_dot_prod_f32.c
|
||||
*
|
||||
* Description: Floating-point complex dot product
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* ---------------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @defgroup cmplx_dot_prod Complex Dot Product
|
||||
*
|
||||
* Computes the dot product of two complex vectors.
|
||||
* The vectors are multiplied element-by-element and then summed.
|
||||
*
|
||||
* The <code>pSrcA</code> points to the first complex input vector and
|
||||
* <code>pSrcB</code> points to the second complex input vector.
|
||||
* <code>numSamples</code> specifies the number of complex samples
|
||||
* and the data in each array is stored in an interleaved fashion
|
||||
* (real, imag, real, imag, ...).
|
||||
* Each array has a total of <code>2*numSamples</code> values.
|
||||
*
|
||||
* The underlying algorithm is used:
|
||||
* <pre>
|
||||
* realResult=0;
|
||||
* imagResult=0;
|
||||
* for(n=0; n<numSamples; n++) {
|
||||
* realResult += pSrcA[(2*n)+0]*pSrcB[(2*n)+0] - pSrcA[(2*n)+1]*pSrcB[(2*n)+1];
|
||||
* imagResult += pSrcA[(2*n)+0]*pSrcB[(2*n)+1] + pSrcA[(2*n)+1]*pSrcB[(2*n)+0];
|
||||
* }
|
||||
* </pre>
|
||||
*
|
||||
* There are separate functions for floating-point, Q15, and Q31 data types.
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_dot_prod
|
||||
* @{
|
||||
*/
|
||||
|
||||
/**
|
||||
* @brief Floating-point complex dot product
|
||||
* @param *pSrcA points to the first input vector
|
||||
* @param *pSrcB points to the second input vector
|
||||
* @param numSamples number of complex samples in each vector
|
||||
* @param *realResult real part of the result returned here
|
||||
* @param *imagResult imaginary part of the result returned here
|
||||
* @return none.
|
||||
*/
|
||||
|
||||
void arm_cmplx_dot_prod_f32(
|
||||
float32_t * pSrcA,
|
||||
float32_t * pSrcB,
|
||||
uint32_t numSamples,
|
||||
float32_t * realResult,
|
||||
float32_t * imagResult)
|
||||
{
|
||||
float32_t real_sum = 0.0f, imag_sum = 0.0f; /* Temporary result storage */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */
|
||||
real_sum += (*pSrcA++) * (*pSrcB++);
|
||||
/* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */
|
||||
imag_sum += (*pSrcA++) * (*pSrcB++);
|
||||
|
||||
real_sum += (*pSrcA++) * (*pSrcB++);
|
||||
imag_sum += (*pSrcA++) * (*pSrcB++);
|
||||
|
||||
real_sum += (*pSrcA++) * (*pSrcB++);
|
||||
imag_sum += (*pSrcA++) * (*pSrcB++);
|
||||
|
||||
real_sum += (*pSrcA++) * (*pSrcB++);
|
||||
imag_sum += (*pSrcA++) * (*pSrcB++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */
|
||||
real_sum += (*pSrcA++) * (*pSrcB++);
|
||||
/* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */
|
||||
imag_sum += (*pSrcA++) * (*pSrcB++);
|
||||
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */
|
||||
real_sum += (*pSrcA++) * (*pSrcB++);
|
||||
/* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */
|
||||
imag_sum += (*pSrcA++) * (*pSrcB++);
|
||||
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
/* Store the real and imaginary results in the destination buffers */
|
||||
*realResult = real_sum;
|
||||
*imagResult = imag_sum;
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_dot_prod group
|
||||
*/
|
||||
@ -0,0 +1,141 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_dot_prod_q15.c
|
||||
*
|
||||
* Description: Processing function for the Q15 Complex Dot product
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* -------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_dot_prod
|
||||
* @{
|
||||
*/
|
||||
|
||||
/**
|
||||
* @brief Q15 complex dot product
|
||||
* @param *pSrcA points to the first input vector
|
||||
* @param *pSrcB points to the second input vector
|
||||
* @param numSamples number of complex samples in each vector
|
||||
* @param *realResult real part of the result returned here
|
||||
* @param *imagResult imaginary part of the result returned here
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function is implemented using an internal 64-bit accumulator.
|
||||
* The intermediate 1.15 by 1.15 multiplications are performed with full precision and yield a 2.30 result.
|
||||
* These are accumulated in a 64-bit accumulator with 34.30 precision.
|
||||
* As a final step, the accumulators are converted to 8.24 format.
|
||||
* The return results <code>realResult</code> and <code>imagResult</code> are in 8.24 format.
|
||||
*/
|
||||
|
||||
void arm_cmplx_dot_prod_q15(
|
||||
q15_t * pSrcA,
|
||||
q15_t * pSrcB,
|
||||
uint32_t numSamples,
|
||||
q31_t * realResult,
|
||||
q31_t * imagResult)
|
||||
{
|
||||
q63_t real_sum = 0, imag_sum = 0; /* Temporary result storage */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */
|
||||
real_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
|
||||
/* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */
|
||||
imag_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
|
||||
real_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
imag_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
|
||||
real_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
imag_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
|
||||
real_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
imag_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */
|
||||
real_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
/* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */
|
||||
imag_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */
|
||||
real_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
/* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */
|
||||
imag_sum += ((q31_t) * pSrcA++ * *pSrcB++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
/* Store the real and imaginary results in 8.24 format */
|
||||
/* Convert real data in 34.30 to 8.24 by 6 right shifts */
|
||||
*realResult = (q31_t) (real_sum) >> 6;
|
||||
/* Convert imaginary data in 34.30 to 8.24 by 6 right shifts */
|
||||
*imagResult = (q31_t) (imag_sum) >> 6;
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_dot_prod group
|
||||
*/
|
||||
@ -0,0 +1,142 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_dot_prod_q31.c
|
||||
*
|
||||
* Description: Q31 complex dot product
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* -------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_dot_prod
|
||||
* @{
|
||||
*/
|
||||
|
||||
/**
|
||||
* @brief Q31 complex dot product
|
||||
* @param *pSrcA points to the first input vector
|
||||
* @param *pSrcB points to the second input vector
|
||||
* @param numSamples number of complex samples in each vector
|
||||
* @param *realResult real part of the result returned here
|
||||
* @param *imagResult imaginary part of the result returned here
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function is implemented using an internal 64-bit accumulator.
|
||||
* The intermediate 1.31 by 1.31 multiplications are performed with 64-bit precision and then shifted to 16.48 format.
|
||||
* The internal real and imaginary accumulators are in 16.48 format and provide 15 guard bits.
|
||||
* Additions are nonsaturating and no overflow will occur as long as <code>numSamples</code> is less than 32768.
|
||||
* The return results <code>realResult</code> and <code>imagResult</code> are in 16.48 format.
|
||||
* Input down scaling is not required.
|
||||
*/
|
||||
|
||||
void arm_cmplx_dot_prod_q31(
|
||||
q31_t * pSrcA,
|
||||
q31_t * pSrcB,
|
||||
uint32_t numSamples,
|
||||
q63_t * realResult,
|
||||
q63_t * imagResult)
|
||||
{
|
||||
q63_t real_sum = 0, imag_sum = 0; /* Temporary result storage */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */
|
||||
/* Convert real data in 2.62 to 16.48 by 14 right shifts */
|
||||
real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
/* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */
|
||||
/* Convert imag data in 2.62 to 16.48 by 14 right shifts */
|
||||
imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
|
||||
real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
|
||||
real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
|
||||
real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */
|
||||
real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
/* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */
|
||||
imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* outReal = realA[0]* realB[0] + realA[2]* realB[2] + realA[4]* realB[4] + .....+ realA[numSamples-2]* realB[numSamples-2] */
|
||||
real_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
/* outImag = imagA[1]* imagB[1] + imagA[3]* imagB[3] + imagA[5]* imagB[5] + .....+ imagA[numSamples-1]* imagB[numSamples-1] */
|
||||
imag_sum += (q63_t) * pSrcA++ * (*pSrcB++) >> 14;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
/* Store the real and imaginary results in 16.48 format */
|
||||
*realResult = real_sum;
|
||||
*imagResult = imag_sum;
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_dot_prod group
|
||||
*/
|
||||
@ -0,0 +1,154 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mag_f32.c
|
||||
*
|
||||
* Description: Floating-point complex magnitude.
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* ---------------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @defgroup cmplx_mag Complex Magnitude
|
||||
*
|
||||
* Computes the magnitude of the elements of a complex data vector.
|
||||
*
|
||||
* The <code>pSrc</code> points to the source data and
|
||||
* <code>pDst</code> points to the where the result should be written.
|
||||
* <code>numSamples</code> specifies the number of complex samples
|
||||
* in the input array and the data is stored in an interleaved fashion
|
||||
* (real, imag, real, imag, ...).
|
||||
* The input array has a total of <code>2*numSamples</code> values;
|
||||
* the output array has a total of <code>numSamples</code> values.
|
||||
* The underlying algorithm is used:
|
||||
*
|
||||
* <pre>
|
||||
* for(n=0; n<numSamples; n++) {
|
||||
* pDst[n] = sqrt(pSrc[(2*n)+0]^2 + pSrc[(2*n)+1]^2);
|
||||
* }
|
||||
* </pre>
|
||||
*
|
||||
* There are separate functions for floating-point, Q15, and Q31 data types.
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_mag
|
||||
* @{
|
||||
*/
|
||||
/**
|
||||
* @brief Floating-point complex magnitude.
|
||||
* @param[in] *pSrc points to complex input buffer
|
||||
* @param[out] *pDst points to real output buffer
|
||||
* @param[in] numSamples number of complex samples in the input vector
|
||||
* @return none.
|
||||
*
|
||||
*/
|
||||
|
||||
|
||||
void arm_cmplx_mag_f32(
|
||||
float32_t * pSrc,
|
||||
float32_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
float32_t realIn, imagIn; /* Temporary variables to hold input values */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
|
||||
/* C[0] = sqrt(A[0] * A[0] + A[1] * A[1]) */
|
||||
realIn = *pSrc++;
|
||||
imagIn = *pSrc++;
|
||||
/* store the result in the destination buffer. */
|
||||
arm_sqrt_f32((realIn * realIn) + (imagIn * imagIn), pDst++);
|
||||
|
||||
realIn = *pSrc++;
|
||||
imagIn = *pSrc++;
|
||||
arm_sqrt_f32((realIn * realIn) + (imagIn * imagIn), pDst++);
|
||||
|
||||
realIn = *pSrc++;
|
||||
imagIn = *pSrc++;
|
||||
arm_sqrt_f32((realIn * realIn) + (imagIn * imagIn), pDst++);
|
||||
|
||||
realIn = *pSrc++;
|
||||
imagIn = *pSrc++;
|
||||
arm_sqrt_f32((realIn * realIn) + (imagIn * imagIn), pDst++);
|
||||
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0] = sqrt(A[0] * A[0] + A[1] * A[1]) */
|
||||
realIn = *pSrc++;
|
||||
imagIn = *pSrc++;
|
||||
/* store the result in the destination buffer. */
|
||||
arm_sqrt_f32((realIn * realIn) + (imagIn * imagIn), pDst++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* out = sqrt((real * real) + (imag * imag)) */
|
||||
realIn = *pSrc++;
|
||||
imagIn = *pSrc++;
|
||||
/* store the result in the destination buffer. */
|
||||
arm_sqrt_f32((realIn * realIn) + (imagIn * imagIn), pDst++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_mag group
|
||||
*/
|
||||
@ -0,0 +1,153 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mag_q15.c
|
||||
*
|
||||
* Description: Q15 complex magnitude.
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* ---------------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_mag
|
||||
* @{
|
||||
*/
|
||||
|
||||
|
||||
/**
|
||||
* @brief Q15 complex magnitude
|
||||
* @param *pSrc points to the complex input vector
|
||||
* @param *pDst points to the real output vector
|
||||
* @param numSamples number of complex samples in the input vector
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function implements 1.15 by 1.15 multiplications and finally output is converted into 2.14 format.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mag_q15(
|
||||
q15_t * pSrc,
|
||||
q15_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
q15_t real, imag; /* Temporary variables to hold input values */
|
||||
q31_t acc0, acc1; /* Accumulators */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
|
||||
/* C[0] = sqrt(A[0] * A[0] + A[1] * A[1]) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = __SMUAD(real, real);
|
||||
acc1 = __SMUAD(imag, imag);
|
||||
/* store the result in 2.14 format in the destination buffer. */
|
||||
arm_sqrt_q15((q15_t) (((q63_t) acc0 + acc1) >> 17), pDst++);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = __SMUAD(real, real);
|
||||
acc1 = __SMUAD(imag, imag);
|
||||
/* store the result in 2.14 format in the destination buffer. */
|
||||
arm_sqrt_q15((q15_t) (((q63_t) acc0 + acc1) >> 17), pDst++);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = __SMUAD(real, real);
|
||||
acc1 = __SMUAD(imag, imag);
|
||||
/* store the result in 2.14 format in the destination buffer. */
|
||||
arm_sqrt_q15((q15_t) (((q63_t) acc0 + acc1) >> 17), pDst++);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = __SMUAD(real, real);
|
||||
acc1 = __SMUAD(imag, imag);
|
||||
/* store the result in 2.14 format in the destination buffer. */
|
||||
arm_sqrt_q15((q15_t) (((q63_t) acc0 + acc1) >> 17), pDst++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0] = sqrt(A[0] * A[0] + A[1] * A[1]) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = __SMUAD(real, real);
|
||||
acc1 = __SMUAD(imag, imag);
|
||||
/* store the result in 2.14 format in the destination buffer. */
|
||||
arm_sqrt_q15((q15_t) (((q63_t) acc0 + acc1) >> 17), pDst++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* out = sqrt(real * real + imag * imag) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
|
||||
acc0 = (real * real);
|
||||
acc1 = (imag * imag);
|
||||
|
||||
/* store the result in 2.14 format in the destination buffer. */
|
||||
arm_sqrt_q15((q15_t) (((q63_t) acc0 + acc1) >> 17), pDst++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_mag group
|
||||
*/
|
||||
@ -0,0 +1,151 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mag_q31.c
|
||||
*
|
||||
* Description: Q31 complex magnitude
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* ---------------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_mag
|
||||
* @{
|
||||
*/
|
||||
|
||||
/**
|
||||
* @brief Q31 complex magnitude
|
||||
* @param *pSrc points to the complex input vector
|
||||
* @param *pDst points to the real output vector
|
||||
* @param numSamples number of complex samples in the input vector
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function implements 1.31 by 1.31 multiplications and finally output is converted into 2.30 format.
|
||||
* Input down scaling is not required.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mag_q31(
|
||||
q31_t * pSrc,
|
||||
q31_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
q31_t real, imag; /* Temporary variables to hold input values */
|
||||
q31_t acc0, acc1; /* Accumulators */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
|
||||
/* C[0] = sqrt(A[0] * A[0] + A[1] * A[1]) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 2.30 format in the destination buffer. */
|
||||
arm_sqrt_q31(acc0 + acc1, pDst++);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 2.30 format in the destination buffer. */
|
||||
arm_sqrt_q31(acc0 + acc1, pDst++);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 2.30 format in the destination buffer. */
|
||||
arm_sqrt_q31(acc0 + acc1, pDst++);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 2.30 format in the destination buffer. */
|
||||
arm_sqrt_q31(acc0 + acc1, pDst++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0] = sqrt(A[0] * A[0] + A[1] * A[1]) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 2.30 format in the destination buffer. */
|
||||
arm_sqrt_q31(acc0 + acc1, pDst++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* out = sqrt((real * real) + (imag * imag)) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 2.30 format in the destination buffer. */
|
||||
arm_sqrt_q31(acc0 + acc1, pDst++);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_mag group
|
||||
*/
|
||||
@ -0,0 +1,155 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mag_squared_f32.c
|
||||
*
|
||||
* Description: Floating-point complex magnitude squared.
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* ---------------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @defgroup cmplx_mag_squared Complex Magnitude Squared
|
||||
*
|
||||
* Computes the magnitude squared of the elements of a complex data vector.
|
||||
*
|
||||
* The <code>pSrc</code> points to the source data and
|
||||
* <code>pDst</code> points to the where the result should be written.
|
||||
* <code>numSamples</code> specifies the number of complex samples
|
||||
* in the input array and the data is stored in an interleaved fashion
|
||||
* (real, imag, real, imag, ...).
|
||||
* The input array has a total of <code>2*numSamples</code> values;
|
||||
* the output array has a total of <code>numSamples</code> values.
|
||||
*
|
||||
* The underlying algorithm is used:
|
||||
*
|
||||
* <pre>
|
||||
* for(n=0; n<numSamples; n++) {
|
||||
* pDst[n] = pSrc[(2*n)+0]^2 + pSrc[(2*n)+1]^2;
|
||||
* }
|
||||
* </pre>
|
||||
*
|
||||
* There are separate functions for floating-point, Q15, and Q31 data types.
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_mag_squared
|
||||
* @{
|
||||
*/
|
||||
|
||||
|
||||
/**
|
||||
* @brief Floating-point complex magnitude squared
|
||||
* @param[in] *pSrc points to the complex input vector
|
||||
* @param[out] *pDst points to the real output vector
|
||||
* @param[in] numSamples number of complex samples in the input vector
|
||||
* @return none.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mag_squared_f32(
|
||||
float32_t * pSrc,
|
||||
float32_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
float32_t real, imag; /* Temporary variables to store real and imaginary values */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0] = (A[0] * A[0] + A[1] * A[1]) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pDst++ = (real * real) + (imag * imag);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
*pDst++ = (real * real) + (imag * imag);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
*pDst++ = (real * real) + (imag * imag);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
*pDst++ = (real * real) + (imag * imag);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0] = (A[0] * A[0] + A[1] * A[1]) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pDst++ = (real * real) + (imag * imag);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* reading real and imaginary values */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
|
||||
/* out = (real * real) + (imag * imag) */
|
||||
/* store the result in the destination buffer. */
|
||||
*pDst++ = (real * real) + (imag * imag);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_mag_squared group
|
||||
*/
|
||||
@ -0,0 +1,148 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mag_squared_q15.c
|
||||
*
|
||||
* Description: Q15 complex magnitude squared.
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* ---------------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_mag_squared
|
||||
* @{
|
||||
*/
|
||||
|
||||
/**
|
||||
* @brief Q15 complex magnitude squared
|
||||
* @param *pSrc points to the complex input vector
|
||||
* @param *pDst points to the real output vector
|
||||
* @param numSamples number of complex samples in the input vector
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function implements 1.15 by 1.15 multiplications and finally output is converted into 3.13 format.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mag_squared_q15(
|
||||
q15_t * pSrc,
|
||||
q15_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
q15_t real, imag; /* Temporary variables to store real and imaginary values */
|
||||
q31_t acc0, acc1; /* Accumulators */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
/*loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0] = (A[0] * A[0] + A[1] * A[1]) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = __SMUAD(real, real);
|
||||
acc1 = __SMUAD(imag, imag);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ = (q15_t) (((q63_t) acc0 + acc1) >> 17);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = __SMUAD(real, real);
|
||||
acc1 = __SMUAD(imag, imag);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ = (q15_t) (((q63_t) acc0 + acc1) >> 17);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = __SMUAD(real, real);
|
||||
acc1 = __SMUAD(imag, imag);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ = (q15_t) (((q63_t) acc0 + acc1) >> 17);
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = __SMUAD(real, real);
|
||||
acc1 = __SMUAD(imag, imag);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ = (q15_t) (((q63_t) acc0 + acc1) >> 17);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0] = (A[0] * A[0] + A[1] * A[1]) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = __SMUAD(real, real);
|
||||
acc1 = __SMUAD(imag, imag);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ = (q15_t) (((q63_t) acc0 + acc1) >> 17);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* out = ((real * real) + (imag * imag)) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (real * real);
|
||||
acc1 = (imag * imag);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ = (q15_t) (((q63_t) acc0 + acc1) >> 17);
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_mag_squared group
|
||||
*/
|
||||
@ -0,0 +1,150 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mag_squared_q31.c
|
||||
*
|
||||
* Description: Q31 complex magnitude squared.
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* ---------------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup cmplx_mag_squared
|
||||
* @{
|
||||
*/
|
||||
|
||||
|
||||
/**
|
||||
* @brief Q31 complex magnitude squared
|
||||
* @param *pSrc points to the complex input vector
|
||||
* @param *pDst points to the real output vector
|
||||
* @param numSamples number of complex samples in the input vector
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function implements 1.31 by 1.31 multiplications and finally output is converted into 3.29 format.
|
||||
* Input down scaling is not required.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mag_squared_q31(
|
||||
q31_t * pSrc,
|
||||
q31_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
q31_t real, imag; /* Temporary variables to store real and imaginary values */
|
||||
q31_t acc0, acc1; /* Accumulators */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counter */
|
||||
|
||||
/* loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0] = (A[0] * A[0] + A[1] * A[1]) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = acc0 + acc1;
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = acc0 + acc1;
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = acc0 + acc1;
|
||||
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = acc0 + acc1;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[0] = (A[0] * A[0] + A[1] * A[1]) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = acc0 + acc1;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* out = ((real * real) + (imag * imag)) */
|
||||
real = *pSrc++;
|
||||
imag = *pSrc++;
|
||||
acc0 = (q31_t) (((q63_t) real * real) >> 33);
|
||||
acc1 = (q31_t) (((q63_t) imag * imag) >> 33);
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = acc0 + acc1;
|
||||
|
||||
/* Decrement the loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of cmplx_mag_squared group
|
||||
*/
|
||||
@ -0,0 +1,180 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mult_cmplx_f32.c
|
||||
*
|
||||
* Description: Floating-point complex-by-complex multiplication
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* -------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @defgroup CmplxByCmplxMult Complex-by-Complex Multiplication
|
||||
*
|
||||
* Multiplies a complex vector by another complex vector and generates a complex result.
|
||||
* The data in the complex arrays is stored in an interleaved fashion
|
||||
* (real, imag, real, imag, ...).
|
||||
* The parameter <code>numSamples</code> represents the number of complex
|
||||
* samples processed. The complex arrays have a total of <code>2*numSamples</code>
|
||||
* real values.
|
||||
*
|
||||
* The underlying algorithm is used:
|
||||
*
|
||||
* <pre>
|
||||
* for(n=0; n<numSamples; n++) {
|
||||
* pDst[(2*n)+0] = pSrcA[(2*n)+0] * pSrcB[(2*n)+0] - pSrcA[(2*n)+1] * pSrcB[(2*n)+1];
|
||||
* pDst[(2*n)+1] = pSrcA[(2*n)+0] * pSrcB[(2*n)+1] + pSrcA[(2*n)+1] * pSrcB[(2*n)+0];
|
||||
* }
|
||||
* </pre>
|
||||
*
|
||||
* There are separate functions for floating-point, Q15, and Q31 data types.
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup CmplxByCmplxMult
|
||||
* @{
|
||||
*/
|
||||
|
||||
|
||||
/**
|
||||
* @brief Floating-point complex-by-complex multiplication
|
||||
* @param[in] *pSrcA points to the first input vector
|
||||
* @param[in] *pSrcB points to the second input vector
|
||||
* @param[out] *pDst points to the output vector
|
||||
* @param[in] numSamples number of complex samples in each vector
|
||||
* @return none.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mult_cmplx_f32(
|
||||
float32_t * pSrcA,
|
||||
float32_t * pSrcB,
|
||||
float32_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
float32_t a, b, c, d; /* Temporary variables to store real and imaginary values */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counters */
|
||||
|
||||
/* loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
|
||||
/* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in the destination buffer. */
|
||||
*pDst++ = (a * c) - (b * d);
|
||||
*pDst++ = (a * d) + (b * c);
|
||||
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
*pDst++ = (a * c) - (b * d);
|
||||
*pDst++ = (a * d) + (b * c);
|
||||
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
*pDst++ = (a * c) - (b * d);
|
||||
*pDst++ = (a * d) + (b * c);
|
||||
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
*pDst++ = (a * c) - (b * d);
|
||||
*pDst++ = (a * d) + (b * c);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
|
||||
/* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in the destination buffer. */
|
||||
*pDst++ = (a * c) - (b * d);
|
||||
*pDst++ = (a * d) + (b * c);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
|
||||
/* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in the destination buffer. */
|
||||
*pDst++ = (a * c) - (b * d);
|
||||
*pDst++ = (a * d) + (b * c);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of CmplxByCmplxMult group
|
||||
*/
|
||||
@ -0,0 +1,182 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mult_cmplx_q15.c
|
||||
*
|
||||
* Description: Q15 complex-by-complex multiplication
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* -------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup CmplxByCmplxMult
|
||||
* @{
|
||||
*/
|
||||
|
||||
/**
|
||||
* @brief Q15 complex-by-complex multiplication
|
||||
* @param[in] *pSrcA points to the first input vector
|
||||
* @param[in] *pSrcB points to the second input vector
|
||||
* @param[out] *pDst points to the output vector
|
||||
* @param[in] numSamples number of complex samples in each vector
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function implements 1.15 by 1.15 multiplications and finally output is converted into 3.13 format.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mult_cmplx_q15(
|
||||
q15_t * pSrcA,
|
||||
q15_t * pSrcB,
|
||||
q15_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
q15_t a, b, c, d; /* Temporary variables to store real and imaginary values */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counters */
|
||||
|
||||
/* loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
|
||||
/* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * c) >> 17) - (((q31_t) b * d) >> 17);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * d) >> 17) + (((q31_t) b * c) >> 17);
|
||||
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * c) >> 17) - (((q31_t) b * d) >> 17);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * d) >> 17) + (((q31_t) b * c) >> 17);
|
||||
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * c) >> 17) - (((q31_t) b * d) >> 17);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * d) >> 17) + (((q31_t) b * c) >> 17);
|
||||
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * c) >> 17) - (((q31_t) b * d) >> 17);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * d) >> 17) + (((q31_t) b * c) >> 17);
|
||||
|
||||
/* Decrement the blockSize loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the blockSize is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
|
||||
/* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * c) >> 17) - (((q31_t) b * d) >> 17);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * d) >> 17) + (((q31_t) b * c) >> 17);
|
||||
|
||||
/* Decrement the blockSize loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
|
||||
/* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * c) >> 17) - (((q31_t) b * d) >> 17);
|
||||
/* store the result in 3.13 format in the destination buffer. */
|
||||
*pDst++ =
|
||||
(q15_t) (q31_t) (((q31_t) a * d) >> 17) + (((q31_t) b * c) >> 17);
|
||||
|
||||
/* Decrement the blockSize loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of CmplxByCmplxMult group
|
||||
*/
|
||||
@ -0,0 +1,209 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mult_cmplx_q31.c
|
||||
*
|
||||
* Description: Q31 complex-by-complex multiplication
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* -------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup CmplxByCmplxMult
|
||||
* @{
|
||||
*/
|
||||
|
||||
|
||||
/**
|
||||
* @brief Q31 complex-by-complex multiplication
|
||||
* @param[in] *pSrcA points to the first input vector
|
||||
* @param[in] *pSrcB points to the second input vector
|
||||
* @param[out] *pDst points to the output vector
|
||||
* @param[in] numSamples number of complex samples in each vector
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function implements 1.31 by 1.31 multiplications and finally output is converted into 3.29 format.
|
||||
* Input down scaling is not required.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mult_cmplx_q31(
|
||||
q31_t * pSrcA,
|
||||
q31_t * pSrcB,
|
||||
q31_t * pDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
q31_t a, b, c, d; /* Temporary variables to store real and imaginary values */
|
||||
uint32_t blkCnt; /* loop counters */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
|
||||
/* loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
|
||||
/* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the real result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33));
|
||||
/* store the imag result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33));
|
||||
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33));
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33));
|
||||
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33));
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33));
|
||||
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33));
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33));
|
||||
|
||||
/* Decrement the blockSize loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the blockSize is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
|
||||
/* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33));
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33));
|
||||
|
||||
/* Decrement the blockSize loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
/* loop Unrolling */
|
||||
blkCnt = numSamples >> 1u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 2 outputs at a time.
|
||||
** a second loop below computes the remaining 1 sample. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
|
||||
/* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the real result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33));
|
||||
/* store the imag result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33));
|
||||
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33));
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33));
|
||||
|
||||
/* Decrement the blockSize loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the blockSize is not a multiple of 2, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x2u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
|
||||
/* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
|
||||
a = *pSrcA++;
|
||||
b = *pSrcA++;
|
||||
c = *pSrcB++;
|
||||
d = *pSrcB++;
|
||||
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33));
|
||||
/* store the result in 3.29 format in the destination buffer. */
|
||||
*pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33));
|
||||
|
||||
/* Decrement the blockSize loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of CmplxByCmplxMult group
|
||||
*/
|
||||
@ -0,0 +1,157 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mult_real_f32.c
|
||||
*
|
||||
* Description: Floating-point complex by real multiplication
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* -------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @defgroup CmplxByRealMult Complex-by-Real Multiplication
|
||||
*
|
||||
* Multiplies a complex vector by a real vector and generates a complex result.
|
||||
* The data in the complex arrays is stored in an interleaved fashion
|
||||
* (real, imag, real, imag, ...).
|
||||
* The parameter <code>numSamples</code> represents the number of complex
|
||||
* samples processed. The complex arrays have a total of <code>2*numSamples</code>
|
||||
* real values while the real array has a total of <code>numSamples</code>
|
||||
* real values.
|
||||
*
|
||||
* The underlying algorithm is used:
|
||||
*
|
||||
* <pre>
|
||||
* for(n=0; n<numSamples; n++) {
|
||||
* pCmplxDst[(2*n)+0] = pSrcCmplx[(2*n)+0] * pSrcReal[n];
|
||||
* pCmplxDst[(2*n)+1] = pSrcCmplx[(2*n)+1] * pSrcReal[n];
|
||||
* }
|
||||
* </pre>
|
||||
*
|
||||
* There are separate functions for floating-point, Q15, and Q31 data types.
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup CmplxByRealMult
|
||||
* @{
|
||||
*/
|
||||
|
||||
|
||||
/**
|
||||
* @brief Floating-point complex-by-real multiplication
|
||||
* @param[in] *pSrcCmplx points to the complex input vector
|
||||
* @param[in] *pSrcReal points to the real input vector
|
||||
* @param[out] *pCmplxDst points to the complex output vector
|
||||
* @param[in] numSamples number of samples in each vector
|
||||
* @return none.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mult_real_f32(
|
||||
float32_t * pSrcCmplx,
|
||||
float32_t * pSrcReal,
|
||||
float32_t * pCmplxDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
float32_t in; /* Temporary variable to store input value */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counters */
|
||||
|
||||
/* loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[i]. */
|
||||
/* C[2 * i + 1] = A[2 * i + 1] * B[i]. */
|
||||
in = *pSrcReal++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
|
||||
in = *pSrcReal++;
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
|
||||
in = *pSrcReal++;
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
|
||||
in = *pSrcReal++;
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[i]. */
|
||||
/* C[2 * i + 1] = A[2 * i + 1] * B[i]. */
|
||||
in = *pSrcReal++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* realOut = realA * realB. */
|
||||
/* imagOut = imagA * realB. */
|
||||
in = *pSrcReal++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
*pCmplxDst++ = (*pSrcCmplx++) * (in);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of CmplxByRealMult group
|
||||
*/
|
||||
@ -0,0 +1,151 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mult_real_q15.c
|
||||
*
|
||||
* Description: Q15 complex by real multiplication
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* -------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup CmplxByRealMult
|
||||
* @{
|
||||
*/
|
||||
|
||||
|
||||
/**
|
||||
* @brief Q15 complex-by-real multiplication
|
||||
* @param[in] *pSrcCmplx points to the complex input vector
|
||||
* @param[in] *pSrcReal points to the real input vector
|
||||
* @param[out] *pCmplxDst points to the complex output vector
|
||||
* @param[in] numSamples number of samples in each vector
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function uses saturating arithmetic.
|
||||
* Results outside of the allowable Q15 range [0x8000 0x7FFF] will be saturated.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mult_real_q15(
|
||||
q15_t * pSrcCmplx,
|
||||
q15_t * pSrcReal,
|
||||
q15_t * pCmplxDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
q15_t in; /* Temporary variable to store input value */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counters */
|
||||
|
||||
/* loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[i]. */
|
||||
/* C[2 * i + 1] = A[2 * i + 1] * B[i]. */
|
||||
in = *pSrcReal++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
|
||||
in = *pSrcReal++;
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
|
||||
in = *pSrcReal++;
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
|
||||
in = *pSrcReal++;
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[i]. */
|
||||
/* C[2 * i + 1] = A[2 * i + 1] * B[i]. */
|
||||
in = *pSrcReal++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* realOut = realA * realB. */
|
||||
/* imagOut = imagA * realB. */
|
||||
in = *pSrcReal++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
*pCmplxDst++ =
|
||||
(q15_t) __SSAT((((q31_t) (*pSrcCmplx++) * (in)) >> 15), 16);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of CmplxByRealMult group
|
||||
*/
|
||||
@ -0,0 +1,151 @@
|
||||
/* ----------------------------------------------------------------------
|
||||
* Copyright (C) 2010 ARM Limited. All rights reserved.
|
||||
*
|
||||
* $Date: 15. July 2011
|
||||
* $Revision: V1.0.10
|
||||
*
|
||||
* Project: CMSIS DSP Library
|
||||
* Title: arm_cmplx_mult_real_q31.c
|
||||
*
|
||||
* Description: Q31 complex by real multiplication
|
||||
*
|
||||
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
||||
*
|
||||
* Version 1.0.10 2011/7/15
|
||||
* Big Endian support added and Merged M0 and M3/M4 Source code.
|
||||
*
|
||||
* Version 1.0.3 2010/11/29
|
||||
* Re-organized the CMSIS folders and updated documentation.
|
||||
*
|
||||
* Version 1.0.2 2010/11/11
|
||||
* Documentation updated.
|
||||
*
|
||||
* Version 1.0.1 2010/10/05
|
||||
* Production release and review comments incorporated.
|
||||
*
|
||||
* Version 1.0.0 2010/09/20
|
||||
* Production release and review comments incorporated.
|
||||
* -------------------------------------------------------------------- */
|
||||
|
||||
#include "arm_math.h"
|
||||
|
||||
/**
|
||||
* @ingroup groupCmplxMath
|
||||
*/
|
||||
|
||||
/**
|
||||
* @addtogroup CmplxByRealMult
|
||||
* @{
|
||||
*/
|
||||
|
||||
|
||||
/**
|
||||
* @brief Q31 complex-by-real multiplication
|
||||
* @param[in] *pSrcCmplx points to the complex input vector
|
||||
* @param[in] *pSrcReal points to the real input vector
|
||||
* @param[out] *pCmplxDst points to the complex output vector
|
||||
* @param[in] numSamples number of samples in each vector
|
||||
* @return none.
|
||||
*
|
||||
* <b>Scaling and Overflow Behavior:</b>
|
||||
* \par
|
||||
* The function uses saturating arithmetic.
|
||||
* Results outside of the allowable Q31 range[0x80000000 0x7FFFFFFF] will be saturated.
|
||||
*/
|
||||
|
||||
void arm_cmplx_mult_real_q31(
|
||||
q31_t * pSrcCmplx,
|
||||
q31_t * pSrcReal,
|
||||
q31_t * pCmplxDst,
|
||||
uint32_t numSamples)
|
||||
{
|
||||
q31_t in; /* Temporary variable to store input value */
|
||||
|
||||
#ifndef ARM_MATH_CM0
|
||||
|
||||
/* Run the below code for Cortex-M4 and Cortex-M3 */
|
||||
uint32_t blkCnt; /* loop counters */
|
||||
|
||||
/* loop Unrolling */
|
||||
blkCnt = numSamples >> 2u;
|
||||
|
||||
/* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
||||
** a second loop below computes the remaining 1 to 3 samples. */
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[i]. */
|
||||
/* C[2 * i + 1] = A[2 * i + 1] * B[i]. */
|
||||
in = *pSrcReal++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
|
||||
in = *pSrcReal++;
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
|
||||
in = *pSrcReal++;
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
|
||||
in = *pSrcReal++;
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
/* If the numSamples is not a multiple of 4, compute any remaining output samples here.
|
||||
** No loop unrolling is used. */
|
||||
blkCnt = numSamples % 0x4u;
|
||||
|
||||
while(blkCnt > 0u)
|
||||
{
|
||||
/* C[2 * i] = A[2 * i] * B[i]. */
|
||||
/* C[2 * i + 1] = A[2 * i + 1] * B[i]. */
|
||||
in = *pSrcReal++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
blkCnt--;
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
/* Run the below code for Cortex-M0 */
|
||||
|
||||
while(numSamples > 0u)
|
||||
{
|
||||
/* realOut = realA * realB. */
|
||||
/* imagReal = imagA * realB. */
|
||||
in = *pSrcReal++;
|
||||
/* store the result in the destination buffer. */
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
*pCmplxDst++ =
|
||||
(q31_t) clip_q63_to_q31(((q63_t) * pSrcCmplx++ * in) >> 31);
|
||||
|
||||
/* Decrement the numSamples loop counter */
|
||||
numSamples--;
|
||||
}
|
||||
|
||||
#endif /* #ifndef ARM_MATH_CM0 */
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* @} end of CmplxByRealMult group
|
||||
*/
|
||||
Reference in New Issue
Block a user