arm_mean_q31.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_mean_q31.c  
00009 *  
00010 * Description:  Mean value of two Q31 arrays.  
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 groupStats  
00031  */ 
00032  
00033 /**  
00034  * @addtogroup mean  
00035  * @{  
00036  */ 
00037  
00038 /**  
00039  * @brief Mean value of a Q31 vector.  
00040  * @param[in]       *pSrc points to the input vector  
00041  * @param[in]       blockSize length of the input vector  
00042  * @param[out]      *pResult mean value returned here  
00043  * @return none.  
00044  *  
00045  * @details  
00046  * <b>Scaling and Overflow Behavior:</b>  
00047  *\par  
00048  * The function is implemented using a 64-bit internal accumulator.  
00049  * The input is represented in 1.31 format and is accumulated in a 64-bit  
00050  * accumulator in 33.31 format.  
00051  * There is no risk of internal overflow with this approach, and the   
00052  * full precision of intermediate result is preserved.   
00053  * Finally, the accumulator is truncated to yield a result of 1.31 format.  
00054  *  
00055  */ 
00056  
00057  
00058 void arm_mean_q31( 
00059   q31_t * pSrc, 
00060   uint32_t blockSize, 
00061   q31_t * pResult) 
00062 { 
00063   q63_t sum = 0;                                 /* Temporary result storage */ 
00064   uint32_t blkCnt;                               /* loop counter */ 
00065  
00066   /*loop Unrolling */ 
00067   blkCnt = blockSize >> 2u; 
00068  
00069   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00070    ** a second loop below computes the remaining 1 to 3 samples. */ 
00071   while(blkCnt > 0u) 
00072   { 
00073     /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) */ 
00074     sum += *pSrc++; 
00075     sum += *pSrc++; 
00076     sum += *pSrc++; 
00077     sum += *pSrc++; 
00078  
00079     /* Decrement the loop counter */ 
00080     blkCnt--; 
00081   } 
00082  
00083   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.  
00084    ** No loop unrolling is used. */ 
00085   blkCnt = blockSize % 0x4u; 
00086  
00087   while(blkCnt > 0u) 
00088   { 
00089     /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) */ 
00090     sum += *pSrc++; 
00091  
00092     /* Decrement the loop counter */ 
00093     blkCnt--; 
00094   } 
00095  
00096   /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) / blockSize  */ 
00097   /* Store the result to the destination */ 
00098   *pResult = (q31_t) (sum / (int32_t) blockSize); 
00099 } 
00100  
00101 /**  
00102  * @} end of mean group  
00103  */