Rectangular Modulation of Noise

Hi everyone,

I'd like to generate a white gaussian noise signal which is modulated by a square signal. I thought the simplest way is just by multiplying in the time domain, but an Error comes out saying "Out of memory" which I can understand because my noise signal is a one column array and the rectangular function is a huge matrix. I think I need to unify both...

However, I thought there might be a modulation-function which can be used for any signal like the ammod(x,f,fs) function from the Communications Toolbox (which is a sine modulation). Is there any chance for a rectangular modulation function?

This is my code:

close all
clear all

fs = 44100;     % Samplefequency in Hz
dt = 1/fs;      % Seconds per Sample
duration = 4;   % duration in seconds
t = (0:dt:duration-dt); 

L = 176400;     % Number of samples after 4 seconds for fs=44,1kHz
sigma = 0;      % standard deviation
mu = 0;

x = sigma+randn(L,1)+mu;        % Gaussian Noise Signal
s = 4*square(2*pi*1*t+1.5);     % 1 Hz Rectangular Signal

y = ammod(x,1,fs);              % 1 Hz Sine Modulation of Noise x

figure(1);

subplot(2,1,1);
plot(t,x,t,s);
axis([0 duration -5 5]);
title('White Noise and 1 Hz Rectangular Function');
xlabel('Seconds');
ylabel('Sample Values');
grid on;

subplot(2,1,2);
plot(t,y);
axis([0 duration -5 5]);
title('White Noise 1 Hz Sine Modulated');
xlabel('Seconds');
ylabel('Sample Values');
grid on;

What I am trying to achieve is a noise signal modulated by a square wave with modulation depth of 100% and a duty cycle of 50% like the red function in the upper plot. Thank you so much, I appreciate any help!

Greetings

[sonicycle]

Alright, solved the problem in the following way:

fs = 44100;     % Sample fequency in Hz
L = 176400;     % Number of samples after 4 seconds for fs=44,1kHz
sigma = 0;      % standard derviation
mu = 0;

fmod = 4;       % Modulation frequency
cycles = L/fs*fmod;

x = sigma+randn(L,1)+mu;            % Gaussian Noise Signal

for j = 1:cycles
    x(fs*1/fmod*j-(fs*1/(2*fmod)):fs*1/fmod*j)=0;
end

A for-loop now mutes the noise-signal in periodic intervals. The modulation frequency fmod and the sample frequency fs can be determined freely.

Looks like this now (e.g. 4 Hz modulated):

Kind Regards :)

sonicycle