/*This program generates Additive White Gaussian Noise (AWGN) based on the Box-Muller Transformation*/
/*given by S= [-2*loge r1]1/2 * cos (2*PI*r2) where r1 and r2 are two functions used to generate*/
/*random numbers between 0.0 and 1.0.*/
#include<ctype.h>
#include<math.h>
#include<stdlib.h>
#include<stdio.h>
#define e 2.71828
#define PI 3.142
#define SEED 100 // Used together with library function srand so
// that same sequence of AWGN is obtained each time program is executed.
// Refer to Scahume's programming with c
void main()
{
double r1,r2,s,mid_value;
int count,item,num;
int freq[10]={0};
FILE *noise_freq; // Excel data file to store noise and corresponding frequency of occurence for each noise interval.
noise_freq=fopen("AWGN_frequency.xls","w"); //To be able to plot required graph
srand(SEED); // Used to initialise random number generator.
printf("\n\t Generation of AWGN based on the Box-Muller transformation.");
printf("\n\nEnter number of noise signal to be generated (preferably a very large value):\n\n");
//The larger the number inout, the larger the standard deviation of the noise distribution
//Try with no. of noise signal to be generated: 50000 or even greater!!!
scanf("%d",&num);
count=0;
printf("\n\n\t\t\t\tCount Noise voltage (V)\n");
while(count<num)
{
r1= rand()/ 32768.0;
r2= rand()/32768.0;
s= sqrt(-2* log(r1)/log(e)) * cos(2*PI*r2);//Box-Muller Transformation
printf("\n\t\t\t\t%d \t %0.2f",count+1,s);
for(item=-5;item<=4;++item) // Finding frequency of occurence of noise voltage lying in a particular the interval.
{
if((s>item)&&(s<=(item+1))) //Interval N to N+1
{
++freq[item+5];//freq[0] to freq[9] because of 10 different noice ranges
break;
}
}
++count;
}
printf("\n\n\n\n\t\t\t Noise(N) \t Frequency \n");// Displaying frequency of noise in each voltage interval.
fprintf(noise_freq,"Noise (midpoint) \t Frequency\n");// Displaying frequency for each noise interval so that graph can be plotted.
for(item=0;item<=9.0;++item)
{
mid_value=((2.0*item)-9.0)/2.0;
printf(" \n\t\t\t %d < N <= %d \t %d",item-5,item-4,freq[item]);
fprintf(noise_freq,"%0.2f \t %d \n",mid_value,freq[item]);
}
fclose(noise_freq);
printf("\n\n\t\t\t");
}
