arm_rms_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_rms_q31.c  
00009 *  
00010 * Description:  root mean square value of 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  * @addtogroup RMS  
00031  * @{  
00032  */ 
00033  
00034  
00035 /**  
00036  * @brief Root Mean Square of the elements of a Q31 vector.  
00037  * @param[in]       *pSrc points to the input vector  
00038  * @param[in]       blockSize length of the input vector  
00039  * @param[out]      *pResult rms value returned here  
00040  * @return none.  
00041  *  
00042  * @details  
00043  * <b>Scaling and Overflow Behavior:</b>  
00044  *  
00045  *\par  
00046  * The function is implemented using an internal 64-bit accumulator.  
00047  * The input is represented in 1.31 format, and intermediate multiplication  
00048  * yields a 2.62 format.  
00049  * The accumulator maintains full precision of the intermediate multiplication results,   
00050  * but provides only a single guard bit.  
00051  * There is no saturation on intermediate additions.  
00052  * If the accumulator overflows, it wraps around and distorts the result.   
00053  * In order to avoid overflows completely, the input signal must be scaled down by   
00054  * log2(blockSize) bits, as a total of blockSize additions are performed internally.   
00055  * Finally, the 2.62 accumulator is right shifted by 31 bits to yield a 1.31 format value.  
00056  *  
00057  */ 
00058  
00059 void arm_rms_q31( 
00060   q31_t * pSrc, 
00061   uint32_t blockSize, 
00062   q31_t * pResult) 
00063 { 
00064   q31_t *pIn1 = pSrc;                            /* SrcA pointer */ 
00065   q63_t sum = 0;                                 /* accumulator */ 
00066   q31_t in;                                      /* Temporary variable to store the input */ 
00067   uint32_t blkCnt;                               /* loop counter */ 
00068  
00069   /*loop Unrolling */ 
00070   blkCnt = blockSize >> 2u; 
00071  
00072   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00073    ** a second loop below computes the remaining 1 to 3 samples. */ 
00074   while(blkCnt > 0u) 
00075   { 
00076     /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */ 
00077     /* Compute sum of the squares and then store the result in a temporary variable, sum */ 
00078     in = *pIn1++; 
00079     sum += (q63_t) in *in; 
00080     in = *pIn1++; 
00081     sum += (q63_t) in *in; 
00082     in = *pIn1++; 
00083     sum += (q63_t) in *in; 
00084     in = *pIn1++; 
00085     sum += (q63_t) in *in; 
00086  
00087     /* Decrement the loop counter */ 
00088     blkCnt--; 
00089   } 
00090  
00091   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.  
00092    ** No loop unrolling is used. */ 
00093   blkCnt = blockSize % 0x4u; 
00094  
00095   while(blkCnt > 0u) 
00096   { 
00097     /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */ 
00098     /* Compute sum of the squares and then store the results in a temporary variable, sum */ 
00099     in = *pIn1++; 
00100     sum += (q63_t) in *in; 
00101  
00102     /* Decrement the loop counter */ 
00103     blkCnt--; 
00104   } 
00105  
00106   /* Convert data in 2.62 to 1.31 by 31 right shifts */ 
00107   sum = sum >> 31; 
00108  
00109   /* Compute Rms and store the result in the destination vector */ 
00110   arm_sqrt_q31((q31_t) (sum / (int32_t) blockSize), pResult); 
00111 } 
00112  
00113 /**  
00114  * @} end of RMS group  
00115  */