International Journal of Advanced Engineering Research and Technology (IJAERT) 64

Volume 4 Issue 4, April 2016, ISSN No.: 2348 8190

Sound Localization of mobile Camera

Sumna Khan1, Simrat Singh Gill2, Sanyam Saxena3, Shivani Rajput4, Shubham Kumar5
( Assistant Professor, Electronics & Communication Department, MIT Moradabad, U.P.- 244001)
1,3,4,5
( Electronics & Communication Engineering, MIT Moradabad, U.P. - 244001 )

Abstract
A In this paper, we discuss how we designed a
platform for camera to capture pictures while clapping.
The servo motor is used to rotate the camera steering,
from where the sound comes of the desired range. The
three microphones are mounted on the cardboard on
the different vertices and servo motor is placed
between the three microphones on the platform and it
will not interfere in the localizing the direction of the
sound source. The output shaft of a servo does not
rotate freely, but rather is made to seek a particular
angular position under electronic control. The DC
geared motor is driven on the direct power supply and
has the range of movement of 360 degree and no noise
of gears.
Keywords: Microphones, Servo Motor, DC Geared
Motor, Camera, Microcontroller.

I. INTRODUCTION
The camera platform is designed in this project can
turn its direction to the face wherever a nearby hand
clapping or other similar sharp impulse comes from.
Therefore we came up with the platform that uses
microphones to detect clap direction and
ATmega1284p microcontroller with servo motor to
control the camera. This design makes selfie much
easier and convenient since all you need is to choose
your favorites scene and clap the hands. In order to
guarantee the accuracy of turning angle, we need to
use a motor whose turning angle and direction are
controllable, so we use servo motor to control the
turning direction of the platform [1].

three vertices. The servo motor and the camera mount

are located at the centre of the platform. Since the
noise of servo motor is really low, so the motor
mounted in the center of three microphones will not
interference the judgment of sound source direction.

II. SERVO MOTOR

A servomotor is a rotary actuator that allows for
precise control of angular position. As the name
suggests, a servomotor is a servomechanism. More
specifically, it is a closed-loop servomechanism that
uses position feedback to control its motion and final
position [9]. The input to its control is some signal,
either analogue or digital, representing the position
commanded for the output shaft. They typically have a
movement range of 180 degree. The motor is paired
with some type of encoder to provide position and
speed feedback. In the simplest case, only the position
is measured. The measured position of the output is
compared to the command position, the external input
to the controller. If the output position differs from that
required, an error signal is generated which then
causes the motor to rotate in either direction, as needed
to bring the output shaft to the appropriate position. As
the positions approach, the error signal reduces to zero
and the motor stops. The very simplest servomotors
use position-only sensing via potentiometer and bangbang control of their motor; the motor always rotates at
full speed (or is stopped). This type of servomotor is
not widely used in industrial motion control, but they
form the basis of the simple and cheap servos used
for radio-controlled models [9]-[10].

Fig 1: Block Diagram

Figure1 illustrates the basic building block diagram of the model.

The mechanical part of the system is constructed with

the cardboard; the three microphones are mounted on

Fig 2: Servo Motor

Figure2 illustrates the Servo motor configuration which is used for

angular positioning.

www.ijaert.org

International Journal of Advanced Engineering Research and Technology (IJAERT) 65

Volume 4 Issue 4, April 2016, ISSN No.: 2348 8190

More sophisticated servomotors measure both the

position and also the speed of the output shaft. They
may also control the speed of their motor, rather than
always running at full speed. Both of these
enhancements, usually in combination with a PID
control algorithm, allow the servomotor to be brought
to its commanded position more quickly and more
precisely, with less overshooting.
The servo motor is controlled by the pulse signal with
the width ranging from 1ms to 20ms. The minimum
pulse width, 1ms is corresponding to 0 degree, while
the maximum pulse width is corresponding to 180
degree as shown in figure.

Figure 3: Motor Rotation Mechanism

Figure 3 illustrates that if the pulse is shorter than 1.5 ms, then the
motor will turn the shaft to closer to 0 degrees. If the pulse is
longer than 1.5ms, the shaft turns closer to 180 degrees.

The constraint of the pulse width is very important, if

the pulse width is out of range it could destroy the
inner structure of servo motor.

III. SOUND SOURCE LOCALIZATION

Sound Source Localization (SSL) is the process of
determining the spatial location of a sound source
based on multiple observations of the emitted sound
signal [2]. The existing strategies of SSL may broadly
be divided into two main classes: indirect and direct
approaches [3]. Indirect approaches to source
localization are usually two-step methods: first, the
relative time delays for the various microphone pairs
are evaluated and then the source location is found as
the intersection of a pair of a set of half-hyperboloids
centered on the different microphone pairs. Each halfhyperboloid determines the possible location of a
sound source based on the measure of the time
difference of arrival between the two microphones. On
the other hand, direct approaches generally scan a set
of candidate source positions and pick the most likely
candidate as an estimate of the sound source location,
thus performing the localization in a single step.
For both approaches, techniques such as the
Generalized Cross-Correlation (GCC) method,
proposed by Knapp and Carter in 1976, are widely
used [4]. The Time Delay Estimation (TDE) between
signals from any pair of microphones can be
performed by computing the cross-correlation function
of the two signals after applying a suitable weighting

step. The lag at which the cross correlation function

has its maximum is taken as the time delay between
them. The type of weighting used with GCC is crucial
to localization performance. Among several types of
weighting, the phase transform (PHAT) is the most
commonly used pre-filter for the GCC because it is
more robust against reverberation. The GCC with the
phase transform (GCC-PHAT) approach has been
shown to perform well in a mild reverberant
environment. Unfortunately, in the presence of even
moderate reverberation levels, the algorithm is
seriously hampered, due to the presence of spurious
peaks.
When the phase transform filter is incorporated with
the steered-beam former method, the resulting
algorithm (SRP-PHAT) is superior in combating the
adverse effects of background noise and reverberation
compared to the conventional steered-beam former
method and the pair wise method, GCC-PHAT
[4].However, the computational requirements of the
method are large and this makes real-time
implementation difficult. Since the SRP-PHAT
method was proposed, there have been several
attempts to reduce the computational requirements of
the intrinsic SRP search process [5],[6].
Time Delay Of Arrival estimation (TDOA) are
reasonably effective in moderately reverberant
environments and, moreover, their low computational
complexity makes them well-suited to real-time
implementation with several sensors [7].In an SRPHAT algorithm, Consider the output from
microphone l, ml(t), in an M microphone system.
Then, the SRP at the spatial point x = [x, y, z] for a
time frame n of length T is defined as
=
+
(. )
(1)
=

DiBiase [18] showed that the SRP can be computed by

summing the GCCs for all possible pairs of the set of
microphones. The GCC for a microphone pair (k, l) is
computed as

= (2)
Where is the time lag, * denotes complex
conjugation, Ml() is the Fourier transform of the
microphone signal ml(t), and kl() = Wk()Wl () is
a combined weighting function in the frequency
domain. Taking into account the symmetries involved
in the computation of Eq.(2) and removing some fixed
energy terms [8], the part of Pn(x) that changes with x
is isolated as

=
(3)
= =+ ()
where kl(x) is the Inter-Microphone Time-Delay
Function (IMTDF). This function is very important,
since it represents the theoretical direct path delay for
the microphone pair (k, l) resulting from a point source
located at x. The IMTDF is mathematically expressed
as

www.ijaert.org

International Journal of Advanced Engineering Research and Technology (IJAERT) 66

Volume 4 Issue 4, April 2016, ISSN No.: 2348 8190

=
(4)

Where c is the speed of sound, and xk and xl are the

microphone locations.
Basically, the SRP-PHAT algorithm is implemented as
follows:
Define a spatial grid G with a given spatial resolution
r. The theoretical delays from each point of the grid to
each microphone pair are pre-computed using Eq. (4).
For each analysis frame, the GCC of each
microphone pair is computed as expressed in Eq. (2).
For each position of the grid x2 G, the contribution of
the different cross-correlations is accumulated, as in
Eq. (3).
Finally, the position with the maximum score is
selected.

and hence it transfers the more torque to the gear that

is connected to the shaft [9]-[10].

V. SPECIFICATION OF MOTOR
12V Dc Gear Motor
Motor Specification [9]:

Electrical Specification [9]:

IV. DC GEAR MOTOR

The DC motor works over a fair range of voltage. The
higher the input voltage more is the rpm (rotations per
minute) of the motor. For example, if the motor works
in the range of 6-12V, it will have the least RPM at 6v
and maximum at 12V. In terms of voltage, we can put
the equation as: RPM= K1 * V
Where, K1= induced voltage constant
V= voltage applied.
The working of the gears can be explained by the
principle of conservation of angular momentum [9] [10]. The gear having smaller radius will cover more
RPM than the one with larger radius. However, the
larger gear will give more torque to the smaller gear
than vice versa. The comparison of angular velocity
between input gear (the one that transfers energy) to
output gears gives the gear ratio. When multiple gears
connected together, conservation of energy is also
followed. The direction in which the other gear rotates
is always the opposite of the gear adjacent to it.
In any DC motor, RPM and torque are inversely
proportional. Hence the gear having more torque will
provide a lesser RPM and converse. In a geared DC
motor, the concept of pulse width modulation is
applied. The equations detailing the working and
torque transfer the gears are shown as:
Tin Win = Tout Wout
Where, Tin = input torque by the driver gear
Win = angular speed of the driver gear
Tout = output torque by the driven gear
Wout = angular speed of the driven gear.
In a geared DC motor the gear connecting the motor
and the gear head is quite small, hence it transfers
more speed to the larger teeth part of the gear head and
makes it rotate. The larger part of the gear further turns
the smaller duplex part. The small duplex part receives
the torque but not the speed from its predecessor which
it transfers to larger part of the other gear and so on.
The third gears duplex part has more teeth than others

Voltage : 12.0 VDC

Output Speed : 20010% RPM
No load output current : 50 mA
Rotation output : CW/CCW
Noise : No Gear Noise
Stall output : Slip Gear, Broken Gear not allowed
Output shaft of the axial clearance : 0.1-0.3 mm
Horizontal clearance requirement : 0.05 mm
No load speed : 5700 RPM
No load current : 30 mA
Rotation : CW

VI. POWER SUPPLY

VII. CONCLUSION
The goal of designing a platform for camera which can
turn its direction to face wherever a nearby hand
clapping or other impulses comes from. The servo
motor generates pulse signal back to the circuit to
avoid the interference of other signals. Since our band
pass filter is from the range of 150 hertz to 800 hertz,
so the sound with greater frequency or lower
frequency will not affect the circuit. This could help to
eliminate the sound but on the other hand, if clapping
frequency is not in the range will also lead to miss of
detection. In this project, we use DC geared motor
instead of the servo motor, because servo motor is
noisy as compared to the DC geared motor. And dc
geared motor has a greater advantage with the power
supply, as in servo motor the power is supplied
through the clock pulses but in the dc geared motor,
power is directly supplied in voltage. The another
disadvantage of the servo motor is that it typically
have a movement range of 180 degree but dc geared
motor has a movement range of 360 degrees.

www.ijaert.org

International Journal of Advanced Engineering Research and Technology (IJAERT) 67

Volume 4 Issue 4, April 2016, ISSN No.: 2348 8190

and more than ten conference papers. He is working as a

project coordinator from last ten years.

REFERENCES
[1] Han Tang Bihan Lao Chuan You Cornell
university/Department of Electrical and computer
engineering/sound
Localizing
camera/people.ece.cornell.edu
[2] G. C. Carter, Coherence and time delay
estimation, Proc. IEEE, vol. 75, pp. 236255,
February 1991
[3] G. C. Knapp. C. H., Carter, The generalized
correlation method for estimation of time delay.

IEEE Transactions on Acoustics, Speech

and Signal Processing, vol.ASSP-24, no.
4, pp. 320327, August 1976.
[4] T.G.Ulrich Klee and J. McDonough, Kalman
filters for time delay of arrival-based source
localization, EURASIP Journal on Applied Signal
Processing, pp. 115, 2006.
[5] J.Benesty, Adaptive eigen value decomposition
algorithm for passive acoustic source localization,
Bell Labs Tech. Memo,1998.
[6] B.C. Dalton, Audio-based localisation for
ubiquitous sensor networks, Masters thesis,
Massachusetts Institute of Technology, September
2005.
[7] Y.T. Chan and K. C. Ho, An efficient closed-form
localization solution from time difference of arrival
measurements, IEEE, 1994.
[8] M.M. P. Svaizer and M. Omologo, Acoustic
source localization in a three- dimensional space
using crosspower spectrum phase.Proceedings of
IEEE International Conference on Acoustics,
Speech, and Signal Processing (ICASSP97), 1997.
[9] www.wikipedia.com
[10] www.eeweb.com

AUTHORS BIOGRAPHY:

Sumna Khan
(khansumna786@gmail.com)
Student of Electronics and Communication Engineering,
Moradabad Institute of Technology, Moradabad, Uttar
Pradesh (244001), and Area of interest includes Robotics
and Software applications (Android).

Sanyam Saxena
(sanyamsaxena310@gmail.com)
Student of Electronics and Communication Engineering,
Moradabad Institute of Technology, Moradabad, Uttar
Pradesh (244001), and Area of interest includes Robotics
and Embedded system.

Shivani Rajput
(shivanirajput66@gmail.com)
Student of Electronics and Communication Engineering,
Moradabad Institute of Technology, Moradabad, Uttar
Pradesh (244001), and Area of interest includes Pspice and
Embedded System.

Simrat Singh Gill

(simrat.gill@gmail.com)
Assistant professor in the department of electronics and
Communication Engineering at M.I.T Moradabad . He has
completed his masters degree in the field of communication
system and bachelors degree in electronics and
communication engineering at M.I.T Moradabad. His main
research interest lies in the area of wireless communication,
wireless sense network, MIMO system, Robotics and
Embedded system. In recent year, he focused on using
modern wireless techniques for developing project based on
embedded system. He has published five papers in journal

Shubham
Kumar
(shubchauhan13@gmail.com)
Student of Electronics and Communication Engineering,
Moradabad Institute of Technology, Moradabad, Uttar
Pradesh (244001), and Area of interest includes Hardware
Development and Robotics.

www.ijaert.org