EmbeddedRelated.com

Additive White Gaussian Noise Generator

Dr Cagri Tanriover March 3, 20131 comment Coded in C

This function generates an Additive White Gaussian Noise (AWGN) sample at every call. Its beauty lies in its simplicity! The generated sample set will have zero mean and a standard deviation of 1. Therefore, one can simply scale the output samples by a different standard deviation to generate different noise profiles. To do this scaling, simply multiply the sample with the standard deviation of your choice.

This is a very versatile function that can be used in communication system simulations and random number generation in various applications. I have used it in the past while working with channel encoder and decoder simulators.

 

#include <stdio.h>
#include <stdlib.h>
#include <math.h>

#define PI 3.1415926536

double AWGN_generator()
{/* Generates additive white Gaussian Noise samples with zero mean and a standard deviation of 1. */
 
  double temp1;
  double temp2;
  double result;
  int p;

  p = 1;

  while( p > 0 )
  {
	temp2 = ( rand() / ( (double)RAND_MAX ) ); /*  rand() function generates an
                                                       integer between 0 and  RAND_MAX,
                                                       which is defined in stdlib.h.
                                                   */

    if ( temp2 == 0 )
    {// temp2 is >= (RAND_MAX / 2)
      p = 1;
    }// end if
    else
    {// temp2 is < (RAND_MAX / 2)
       p = -1;
    }// end else

  }// end while()

  temp1 = cos( ( 2.0 * (double)PI ) * rand() / ( (double)RAND_MAX ) );
  result = sqrt( -2.0 * log( temp2 ) ) * temp1;

  return result;	// return the generated random sample to the caller

}// end AWGN_generator()