arm_cmplx_dot_prod_f32.c

00001 /* ----------------------------------------------------------------------  
00002 * Copyright (C) 2010 ARM Limited. All rights reserved.  
00003 *  
00004 * $Date:        29. November 2010  
00005 * $Revision:    V1.0.3  
00006 *  
00007 * Project:      CMSIS DSP Library  
00008 * Title:        arm_cmplx_dot_prod_f32.c  
00009 *  
00010 * Description:  Floating-point complex dot product  
00011 *  
00012 * Target Processor: Cortex-M4/Cortex-M3
00013 *  
00014 * Version 1.0.3 2010/11/29 
00015 *    Re-organized the CMSIS folders and updated documentation.  
00016 *   
00017 * Version 1.0.2 2010/11/11  
00018 *    Documentation updated.   
00019 *  
00020 * Version 1.0.1 2010/10/05   
00021 *    Production release and review comments incorporated.  
00022 *  
00023 * Version 1.0.0 2010/09/20   
00024 *    Production release and review comments incorporated.  
00025 * ---------------------------------------------------------------------------- */ 
00026  
00027 #include "arm_math.h" 
00028  
00029 /**  
00030  * @ingroup groupCmplxMath  
00031  */ 
00032  
00033 /**  
00034  * @defgroup cmplx_dot_prod Complex Dot Product  
00035  *  
00036  * Computes the dot product of two complex vectors.  
00037  * The vectors are multiplied element-by-element and then summed.  
00038  * 
00039  * The <code>pSrcA</code> points to the first complex input vector and  
00040  * <code>pSrcB</code> points to the second complex input vector.  
00041  * <code>numSamples</code> specifies the number of complex samples  
00042  * and the data in each array is stored in an interleaved fashion  
00043  * (real, imag, real, imag, ...).  
00044  * Each array has a total of <code>2*numSamples</code> values.  
00045  *  
00046  * The underlying algorithm is used:  
00047  * <pre>  
00048  * realResult=0;  
00049  * imagResult=0;  
00050  * for(n=0; n<numSamples; n++) {  
00051  *     realResult += pSrcA[(2*n)+0]*pSrcB[(2*n)+0] - pSrcA[(2*n)+1]*pSrcB[(2*n)+1];  
00052  *     imagResult += pSrcA[(2*n)+0]*pSrcB[(2*n)+1] + pSrcA[(2*n)+1]*pSrcB[(2*n)+0];  
00053  * }  
00054  * </pre>  
00055  *  
00056  * There are separate functions for floating-point, Q15, and Q31 data types.  
00057  */ 
00058  
00059 /**  
00060  * @addtogroup cmplx_dot_prod  
00061  * @{  
00062  */ 
00063  
00064 /**  
00065  * @brief  Floating-point complex dot product  
00066  * @param  *pSrcA points to the first input vector  
00067  * @param  *pSrcB points to the second input vector  
00068  * @param  numSamples number of complex samples in each vector  
00069  * @param  *realResult real part of the result returned here  
00070  * @param  *imagResult imaginary part of the result returned here  
00071  * @return none.  
00072  */ 
00073  
00074 void arm_cmplx_dot_prod_f32( 
00075   float32_t * pSrcA, 
00076   float32_t * pSrcB, 
00077   uint32_t numSamples, 
00078   float32_t * realResult, 
00079   float32_t * imagResult) 
00080 { 
00081   float32_t real_sum = 0.0f, imag_sum = 0.0f;    /* Temporary result storage */ 
00082   uint32_t blkCnt;                               /* loop counter */ 
00083  
00084   /*loop Unrolling */ 
00085   blkCnt = numSamples >> 2u; 
00086  
00087   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00088    ** a second loop below computes the remaining 1 to 3 samples. */ 
00089   while(blkCnt > 0u) 
00090   { 
00091     /* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */ 
00092     real_sum += (*pSrcA++) * (*pSrcB++); 
00093     /* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */ 
00094     imag_sum += (*pSrcA++) * (*pSrcB++); 
00095  
00096     real_sum += (*pSrcA++) * (*pSrcB++); 
00097     imag_sum += (*pSrcA++) * (*pSrcB++); 
00098  
00099     real_sum += (*pSrcA++) * (*pSrcB++); 
00100     imag_sum += (*pSrcA++) * (*pSrcB++); 
00101  
00102     real_sum += (*pSrcA++) * (*pSrcB++); 
00103     imag_sum += (*pSrcA++) * (*pSrcB++); 
00104  
00105     /* Decrement the loop counter */ 
00106     blkCnt--; 
00107   } 
00108  
00109   /* If the numSamples is not a multiple of 4, compute any remaining output samples here.  
00110    ** No loop unrolling is used. */ 
00111   blkCnt = numSamples % 0x4u; 
00112  
00113   while(blkCnt > 0u) 
00114   { 
00115     /* CReal = A[0]* B[0] + A[2]* B[2] + A[4]* B[4] + .....+ A[numSamples-2]* B[numSamples-2] */ 
00116     real_sum += (*pSrcA++) * (*pSrcB++); 
00117     /* CImag = A[1]* B[1] + A[3]* B[3] + A[5]* B[5] + .....+ A[numSamples-1]* B[numSamples-1] */ 
00118     imag_sum += (*pSrcA++) * (*pSrcB++); 
00119  
00120  
00121     /* Decrement the loop counter */ 
00122     blkCnt--; 
00123   } 
00124  
00125   /* Store the real and imaginary results in the destination buffers */ 
00126   *realResult = real_sum; 
00127   *imagResult = imag_sum; 
00128 } 
00129  
00130 /**  
00131  * @} end of cmplx_dot_prod group  
00132  */