arm_rms_q15.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_rms_q15.c  
00009 *  
00010 * Description:  Processing function for the Q15 RMS  
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  * @addtogroup RMS  
00031  * @{  
00032  */ 
00033  
00034 /**  
00035  * @brief Root Mean Square of the elements of a Q15 vector.  
00036  * @param[in]       *pSrc points to the input vector  
00037  * @param[in]       blockSize length of the input vector  
00038  * @param[out]      *pResult rms value returned here  
00039  * @return none.  
00040  *  
00041  * @details  
00042  * <b>Scaling and Overflow Behavior:</b>  
00043  *  
00044  * \par  
00045  * The function is implemented using a 64-bit internal accumulator.  
00046  * The input is represented in 1.15 format.  
00047  * Intermediate multiplication yields a 2.30 format, and this  
00048  * result is added without saturation to a 64-bit accumulator in 34.30 format.  
00049  * With 33 guard bits in the accumulator, there is no risk of overflow, and the  
00050  * full precision of the intermediate multiplication is preserved.  
00051  * Finally, the 34.30 result is truncated to 34.15 format by discarding the lower   
00052  * 15 bits, and then saturated to yield a result in 1.15 format.  
00053  *  
00054  */ 
00055  
00056 void arm_rms_q15( 
00057   q15_t * pSrc, 
00058   uint32_t blockSize, 
00059   q15_t * pResult) 
00060 { 
00061   q63_t sum = 0;                                 /* accumulator */ 
00062   q31_t in;                                      /* temporary variable to store the input value */ 
00063   q15_t in1;                                     /* temporary variable to store the input value */ 
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[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */ 
00074     /* Compute sum of the squares and then store the results in a temporary variable, sum */ 
00075     in = *__SIMD32(pSrc)++; 
00076     sum = __SMLALD(in, in, sum); 
00077     in = *__SIMD32(pSrc)++; 
00078     sum = __SMLALD(in, in, sum); 
00079  
00080     /* Decrement the loop counter */ 
00081     blkCnt--; 
00082   } 
00083  
00084   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.  
00085    ** No loop unrolling is used. */ 
00086   blkCnt = blockSize % 0x4u; 
00087  
00088   while(blkCnt > 0u) 
00089   { 
00090     /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */ 
00091     /* Compute sum of the squares and then store the results in a temporary variable, sum */ 
00092     in1 = *pSrc++; 
00093     sum = __SMLALD(in1, in1, sum); 
00094  
00095     /* Decrement the loop counter */ 
00096     blkCnt--; 
00097   } 
00098  
00099   /* Truncating and saturating the accumulator to 1.15 format */ 
00100   sum = __SSAT((q31_t) (sum >> 15), 16); 
00101  
00102   in1 = (q15_t) (sum / blockSize); 
00103  
00104   /* Store the result in the destination */ 
00105   arm_sqrt_q15(in1, pResult); 
00106 } 
00107  
00108 /**  
00109  * @} end of RMS group  
00110  */