%
% FUNCTION 3.4 : "cp0303_PPM_random"
%
% Generation of a PPM-UWB signal in the case of
% a random modulating signal and rectangular pulses
% The modulating signal is characterized by
% a normal distribution
%
% Transmitted power is fixed at 'Pow'
% The signal is sampled with frequency 'fc'
% 'np' is the number of generated pulses
% 'Ts' is the average pulse repetition period
% Each rectangular pulse has time duration 'Tr'
% The random modulating signal is characterized
% by standard deviation 'sigma'
%
% The function returns the generated signal 'Stx',
% the corresponding sampling frequency 'fc',
% and vector 'M0' of all the PPM timeshifts
%
% Programmed by Guerino Giancola
%

function [Stx,fc,M0]=cp0303_PPM_random;

% ----------------------------
% Step Zero - Input parameters
% ----------------------------

Pow = -30;    % average transmitted power (dBm)

fc = 1e11;    % sampling frequency

np = 10000;   % number of pulses

Tr = 0.5e-9;  % time duration of the rectangular pulse [s]

Ts = 2e-9;    % average pulse repetition period [s]

sigma = 0.1e-9;  % standard deviation of the modulating signal

% ----------------------------------------
% Step One - Simulating transmission chain
% ----------------------------------------

dt = 1 / fc;            % sampling period
sTs = floor(Ts/dt);     % number of samples per frame 
sTot = sTs * np;        % total number of samples
Stx = zeros(1,sTot);    % output vector

% PPM
j = (0:1:np-1);
M0 = max(zeros(1,np), min((Ts -Tr).*...
   ones(1,np),((Ts/2)+sigma.*randn(1,np))));
M1 = j.*Ts;
Mtot = M0 + M1;
for k = 1 : np
  Stx(1+floor(Mtot(k)/dt))=1;
end

% shaping filter
sP = floor(Tr/dt);             % number of samples per pulse
                               
p0 = (1/sqrt(Tr)).*ones(1,sP); % energy normalized rect
power = (10^(Pow/10))/1000;    % average transmitted power
                               %  (watt)
Ex = power * Ts;        % energy per pulse
ptx= p0 .* sqrt(Ex);    % pulse waveform

Stx = conv(Stx,ptx);
Stx = Stx(1:sTot);