arm_power_q7.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_q7.c  
00009 *  
00010 * Description:  sum of the square of the elements in an array of Q7 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 Q7 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 32-bit internal accumulator.   
00050  * The input is represented in 1.7 format. 
00051  * Intermediate multiplication yields a 2.14 format, and this  
00052  * result is added without saturation to an accumulator in 18.14 format.  
00053  * With 17 guard bits in the accumulator, there is no risk of overflow, and the  
00054  * full precision of the intermediate multiplication is preserved.  
00055  * Finally, the return result is in 18.14 format.   
00056  *  
00057  */ 
00058  
00059 void arm_power_q7( 
00060   q7_t * pSrc, 
00061   uint32_t blockSize, 
00062   q31_t * pResult) 
00063 { 
00064   q31_t sum = 0;                                 /* Temporary result storage */ 
00065   q31_t input1;                                  /* Temporary variable to store packed input */ 
00066   q15_t in1, in2;                                /* Temporary variables to store input */ 
00067   q7_t in;                                       /* Temporary variable to store input */ 
00068   uint32_t blkCnt;                               /* loop counter */ 
00069  
00070  
00071   /*loop Unrolling */ 
00072   blkCnt = blockSize >> 2u; 
00073  
00074   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00075    ** a second loop below computes the remaining 1 to 3 samples. */ 
00076   while(blkCnt > 0u) 
00077   { 
00078     /* Reading two inputs of pSrc vector and packing */ 
00079     in1 = (q15_t) * pSrc++; 
00080     in2 = (q15_t) * pSrc++; 
00081     input1 = ((q31_t) in1 & 0x0000FFFF) | ((q31_t) in2 << 16); 
00082  
00083     /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */ 
00084     /* Compute Power and then store the result in a temporary variable, sum. */ 
00085     sum = __SMLAD(input1, input1, sum); 
00086  
00087     /* Reading two inputs of pSrc vector and packing */ 
00088     in1 = (q15_t) * pSrc++; 
00089     in2 = (q15_t) * pSrc++; 
00090     input1 = ((q31_t) in1 & 0x0000FFFF) | ((q31_t) in2 << 16); 
00091  
00092     /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */ 
00093     /* Compute Power and then store the result in a temporary variable, sum. */ 
00094     sum = __SMLAD(input1, input1, sum); 
00095  
00096     /* Decrement the loop counter */ 
00097     blkCnt--; 
00098   } 
00099  
00100   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.  
00101    ** No loop unrolling is used. */ 
00102   blkCnt = blockSize % 0x4u; 
00103  
00104   while(blkCnt > 0u) 
00105   { 
00106     /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */ 
00107     /* Compute Power and then store the result in a temporary variable, sum. */ 
00108     in = *pSrc++; 
00109     sum += ((q15_t) in * in); 
00110  
00111     /* Decrement the loop counter */ 
00112     blkCnt--; 
00113   } 
00114  
00115   /* Store the result in 18.14 format  */ 
00116   *pResult = sum; 
00117 } 
00118  
00119 /**  
00120  * @} end of power group  
00121  */