function [n1,n2]=gngauss(m,sgma)
% [n1 n2]=gngauss(m,sgma)
% [n1 n2]=gngauss(sgma)
% [n1 n2]=gngauss
% GNGAUSS generates two independent gaussian variables with 
% mean m & S.D sgma
% if one of the input argement missing it takes mean as zero
% if neither mean nor variance is given, it generates two 
%standard gaussian random variables

if nargin==0
    m=0;
    sgma=1;
elseif nargin==1
    sgma=m;  
    m=0;
end
u=rand;                         % uniform random variable in (0,1)
z=sgma*(sqrt(2*log(1/(1-u))));   % rayleigh distributed random variable
u=rand;                          % another uniform distributed variable in (0,1)
n1=m+z*cos(2*pi*u);            
n2=m+z*sin(2*pi*u);