How can i remove the background noise of my signal?

I have created a sine sweep wave and recorded it using four microphones to the file s1m1(960000:4). I have also recorded the background noise from the room without the sine sweep playing as backgroundroom2(960000:4) using the same four microphones. Both of these files have been attached. I have tried to remove the background noise from my file s1m1 using the answer to a similar problem: http://www.mathworks.com/matlabcentral/answers/106510#answer_115590 but this did not work. Does anyone else have any suggestions?

Thanks, Rory

Answer 1

[Image Analyst]

See the Mathworks official recommendation: <http://www.mathworks.com/matlabcentral/answers/103916-how-can-i-smooth-a-signal-while-preserving-peaks-in-matlab-r2013a. Basically use a Savitky-Golay filter. Why would that work so well in your situation? Well you know from the Taylor series that a sine wave is like a0 + a1*x + a3*x^3 + fifth and higher odd orders. So a Siviztky Golay filter with a 3rd order would be about perfect. Here's a demo for you:

clc;    % Clear the command window.
close all;  % Close all figures (except those of imtool.)
clear;  % Erase all existing variables. Or clearvars if you want.
workspace;  % Make sure the workspace panel is showing.
format long g;
format compact;
fontSize = 25;

% Make a noisy sine wave signal
x = 1 : 300;
period = 50
y = sin(2*pi*x/period);
noiseAmplitude = 0.8;
y = y + noiseAmplitude * rand(size(y));
subplot(2,1,1);
plot(x, y, 'b-', 'LineWidth', 2);
title('Noisy Signal', 'FontSize', fontSize);
% Enlarge figure to full screen.
set(gcf, 'Units', 'Normalized', 'OuterPosition', [0 0 1 1]);
% Give a name to the title bar.
set(gcf, 'Name', 'Demo by ImageAnalyst', 'NumberTitle', 'Off') 

% Now smooth with a Savitzky-Golay sliding polynomial filter
windowWidth = 27
polynomialOrder = 3
smoothY = sgolayfilt(y, polynomialOrder, windowWidth);
subplot(2,1,2);
plot(x, smoothY, 'b-', 'LineWidth', 2);
title('Smoothed Signal', 'FontSize', fontSize);

Answer 2

[Youssef Khmou]

hi, One of the ways to handle this problem is to treat the signal in frequency domain, after applying the Fourier transform try to delete the unwanted frequencies and/or smooth the wideband spectrum of the sweep, i realized that there are two types of peaks , wide and narrow , try to start with these tests, on first signal, once the operation is good , apply it to three left signals :

f1=(fft(s1m1(:,1)));
t=f1;
t(abs(t)<30)=0;
z=ifft(t);
sound(z)
figure; plot(abs(t))
figure; plot(real(t))
figure; plot(imag(t))