arm_power_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_power_q31.c  
00009 *  
00010 * Description:  sum of the square of the elements in an array of Q31 type  
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 power  
00035  * @{  
00036  */ 
00037  
00038 /**  
00039  * @brief Sum of the squares of the elements 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 sum of the squares value returned here  
00043  * @return none.  
00044  *  
00045  * @details  
00046  * <b>Scaling and Overflow Behavior:</b>  
00047  *  
00048  * \par  
00049  * The function is implemented using a 64-bit internal accumulator.  
00050  * The input is represented in 1.31 format.  
00051  * Intermediate multiplication yields a 2.62 format, and this  
00052  * result is truncated to 2.48 format by discarding the lower 14 bits.  
00053  * The 2.48 result is then added without saturation to a 64-bit accumulator in 16.48 format.  
00054  * With 15 guard bits in the accumulator, there is no risk of overflow, and the  
00055  * full precision of the intermediate multiplication is preserved.  
00056  * Finally, the return result is in 16.48 format.   
00057  *  
00058  */ 
00059  
00060 void arm_power_q31( 
00061   q31_t * pSrc, 
00062   uint32_t blockSize, 
00063   q63_t * pResult) 
00064 { 
00065   q63_t sum = 0;                                 /* Temporary result storage */ 
00066   q31_t in; 
00067   uint32_t blkCnt;                               /* loop counter */ 
00068  
00069  
00070   /*loop Unrolling */ 
00071   blkCnt = blockSize >> 2u; 
00072  
00073   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00074    ** a second loop below computes the remaining 1 to 3 samples. */ 
00075   while(blkCnt > 0u) 
00076   { 
00077     /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */ 
00078     /* Compute Power then shift intermediate results by 14 bits to maintain 16.48 format and then store the result in a temporary variable sum, providing 15 guard bits. */ 
00079     in = *pSrc++; 
00080     sum += ((q63_t) in * in) >> 14u; 
00081  
00082     in = *pSrc++; 
00083     sum += ((q63_t) in * in) >> 14u; 
00084  
00085     in = *pSrc++; 
00086     sum += ((q63_t) in * in) >> 14u; 
00087  
00088     in = *pSrc++; 
00089     sum += ((q63_t) in * in) >> 14u; 
00090  
00091     /* Decrement the loop counter */ 
00092     blkCnt--; 
00093   } 
00094  
00095   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.  
00096    ** No loop unrolling is used. */ 
00097   blkCnt = blockSize % 0x4u; 
00098  
00099   while(blkCnt > 0u) 
00100   { 
00101     /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */ 
00102     /* Compute Power and then store the result in a temporary variable, sum. */ 
00103     in = *pSrc++; 
00104     sum += ((q63_t) in * in) >> 14u; 
00105  
00106     /* Decrement the loop counter */ 
00107     blkCnt--; 
00108   } 
00109  
00110   /* Store the results in 16.48 format  */ 
00111   *pResult = sum; 
00112 } 
00113  
00114 /**  
00115  * @} end of power group  
00116  */