function u = gaussian(t,t0,FWHM,P0,m,C)
  
% This function computes a gaussian or supergaussian pulse with
% the specified parameters:
%
%   sqrt(P0) * exp{ -[(1+iC)/2] * [(t-t0)/T]^(2m) }
%      where T = FWHM/2*sqrt(log(2))
% 
% USAGE:
% 
% u = gaussian (t);
% u = gaussian (t,t0);
% u = gaussian (t,t0,FWHM);
% u = gaussian (t,t0,FWHM,P0);
% u = gaussian (t,t0,FWHM,P0,m);
% u = gaussian (t,t0,FWHM,P0,m,C);
% 
% INPUT:
% 
% t     vector of times at which to compute u
% t0    center of pulse (default = 0)
% FWHM  full-width at half-intensity of pulse (default = 1)
% P0    peak intensity (|u|^2 at t=t0) of pulse (default = 1)
% m     Gaussian order (default = 1)
% C     chirp parameter (default = 0)
% 
% OUTPUT:
% 
% u     vector of the same size as t, representing pulse
%       amplitude
%
% NOTES:
%
% In order to produce a standard gaussian pulse of the form
% exp(-t.^2/2), you can use:  
%
% u = gaussian(t,0,2*sqrt(log(2)));
  
if (nargin<6)
  C = 0;
end
if (nargin<5)
  m = 1;
end
if (nargin<4)
  P0 = 1;
end
if (nargin<3)
  FWHM = 1;
end
if (nargin<2)
  t0 = 0;
end

u = sqrt(P0)*pow2(-((1+i*C)/2)*(2*(t-t0)/FWHM).^(2*m));