Final Year Project

Presentation
Laura Reddy
Visual Alerting system for Emergency
Vehicle Proximity for hear
compromised drivers

The aim of this project is to create a way for people who
are deaf or who are wearing head protection to be more
aware of emergency vehicles that are approaching them.

This involves detecting the sounds and then comparing
them to the specific characteristics of the different sirens
to lower the chances of false alarms going off.
Visual Alerting system for Emergency
Vehicle Proximity for hear
compromised drivers cont.

Once the siren has been detected it should start a led
flashing so that the driver will be aware of the presence
of the emergency vehicle.

The system will have to be battery operated so that it
can easily be placed on the dash board of a car or on a
motorbike.

The system will also have to have a method of turning on
and off so that the user can turn it off while they are not
driving to help with power consumption.
Garda Sirens
The Garda have two different sirens that they use to let
drivers know that they are approaching.

The 1
st
siren is the one that they would use while driving
in busy traffic or trying to get someone to pull over.

The 2
nd
siren is the one that they use if they are
approaching a busy intersection and they need to let
people know that they are trying to get through.

I have taken sample recordings of both of these sirens.
Matlab
I have used Matlab to get graphs of these sirens.

I have saved the results from reading the siren into
Matlab into a text file called outFile.

I have also tried to do some Fourier analysis on both of
these sound waves.

The graphs for the Fourier analysis are not coming out
like sin waves they are coming out like strange circles
when I just use the plot() but if I use the semilog()
command the results come better for the 1
st
siren but
they disappear for the 2
nd
siren.
Graph of 1
st
Garda Siren
Graph of Fourier analysis on the 1
st
Siren
using plot()
Graph of Fourier analysis on the 2
nd
Siren
using plot()
Graph of Fourier analysis on the 1
st
Siren
using semilog()
Graph of 2
nd
Garda Siren
Graph of Fourier analysis on the 2
nd
Siren
using semilog()
Work done
I used two methods to try and get the output of the
Fourier transform to look like a sin wave.

The 1
st
one just plots the result of the fft() function.

The 2
nd
one was the same as the method we were
shown in the tutorial yesterday but that keeps throwing
an error that the vectors are not the same length, they
should be as one value is determining the other. (Ts =
1/Fs)
Work to be done
I have to see why neither method I used to get the
Fourier transform worked correctly.

Once I have the determined the values of frequencies for
each of the sirens I will have to write code to compare an
input signal to these values to establish whether the
input is the same as one of the Garda signals.

I will also have to get sample recordings of the sirens
that the fire brigade & the ambulances use so that they
can be integrated into the system as well.
Matlab code
% Laura Reddy 06608523
% Code to do fourier analysis on Garda sirens

clear all;
close all;
clc

[x fs bits] = wavread('E:\Final Year Project\Laura\GardaSiren1.wav');

outFile = fopen('outFile.txt', 'w');
fprintf(outFile, '%f\n', x);
fclose(outFile);

ts = 1/fs;
t = 0:ts:0.5-ts;

Matlab code cont
X = fft(x);
X1 = semilogx(x);

figure(1);
plot(x);

figure(2);
plot(X);

figure(3);
plot(X1);

abs(x)


Matlab code cont.
%plot(t,x);
%grid on;

% % get fft of the signal
%fftOfSignal1 = fft(x,1024);

% % looking at positive sides only
%fftOneSided1 = fftofSignal1(1:512);

% % getting the frequency axis
%df = (fs/2)/511;
%freqAxis = 0:df:fs/2;

%figure(2);
%plot(freqAxis,abs(fftOneSided1));
%xlabel('frequency (Hz) ');

Matlab code cont.
%ylabel('Amplitude (db) ');
%hold on;
%grid on;

%figure(3);
%plot(freqAxis,20*log10(abs(fftOneSided1)));
%xlabel('frequency (Hz) ');
%ylabel('Amplitude (db) ');
%hold on;
%grid on;

% Siren 2
[y fs bits] = wavread('E:\Final Year Project\Laura\GardaSiren2.wav');

outFile = fopen('outFile2.txt', 'w');

Matlab code cont
fprintf(outFile, '%f\n', y);
fclose(outFile);

Y = fft(y);
Y1 = semilogx(y);
figure(4);
plot(y);

figure(5);
plot(Y);

figure(6);
plot(Y1);

abs(y)


Matlab code cont.
%plot(t,x);
%grid on;

% % get fft of the signal
%fftOfSignal1 = fft(x,1024);

% % looking at positive sides only
%fftOneSided1 = fftofSignal1(1:512);

% % getting the frequency axis
%df = (fs/2)/511;
%freqAxis = 0:df:fs/2;

%figure(5);

Matlab code cont.
%plot(freqAxis,abs(fftOneSided1));
%xlabel('frequency (Hz) ');
%ylabel('Amplitude (db) ');
%hold on;
%grid on;

%figure(6);
%plot(freqAxis,20*log10(abs(fftOneSided1)));
%xlabel('frequency (Hz) ');
%ylabel('Amplitude (db) ');
%hold on;
%grid on;

%[y fs bits] = wavread('E:\Final Year Project\Laura\GardaSiren2.wav')
%X = fft(x)

Matlab code cont.
%X1 = semilogx(x)
%Y = fft(y)
%Y1 = semilogx(y)
%plot(x)
%plot(y)
%abs(x)
%abs(y)

Sample output of Fourier analysis on 1
st

siren
Sample output of Fourier analysis on 2
nd

siren