Our Bachelor's final year project for active localization of a sound source using FPGA, microphones and stepper motor. Simulation was also accomplished on Simulink.
Introduction to Machine Learning Unit-4 Notes for II-II Mechanical Engineering
FPGA Based Acoustic Source Localization Project
1. FPGA
Based
Acous.c
Source
Localiza.on
Prabin
Maharjan
Prabin
Man
Baidya
Presha
Joshi
Shristi
Nhuchhe
Pradhan
-‐
Final
Year
Project
Presenta.on
March 18, 2010
Supervisors: Prof. Dr. Dinesh Kumar Sharma
Asst. Prof Jitendra Kumar Manandhar
Internal Examiner: Dr. Nanda Bikram Adhikari
External Examiner: Er. Shiva Raj Baral
Tribhuvan University
Institute of Engineering, Pulchowk Campus, Nepal
2. ! Task
of
locating
the
position
of
sound
source
using
sound
=ield
measurements
! Best
example:
auditory
system
of
human
ear
3. ! Integrating
acoustic
source
localization
with
a
video
camera
so
that
camera
automatically
follows
a
lecturer
or
active
speaker
in
conference
room
! Security
surveillance-‐
camera
automatically
captures
video
whenever
burglar
makes
sound
! Robotics,
Humanoids
4. Capture
sound
signal
from
source
using
microphones
with
ampli=ication
Detect
sound
signal
and
determine
the
position
of
the
sound
source
Control
the
movement
of
stepper
motor
to
the
direction
of
sound
source
24. ! Localizes
the
sound
source
only
in
left
and
right
direction
! Affected
by
background
noise
! Stepper
motor
doesn‘t
react
quickly
to
the
sound
signals
25. ! High
resource
utilization
of
FPGA
board
! Cascading
different
modules
! Different
clock
for
ADC,
stepper
motor
controller
26. ! Using
more
than
2
microphones
for
exact
localization
! Localization
of
acoustic
source
in
3-‐D
environment
! Multiple
source
detection
! Reducing
the
effect
of
background
noise
and
reverberation
27. ! FPGA
prototyping
by
VHDL
examples,
Pong
P.
Chu
! Industrial
Circuits
ApplicaFon
Note
Stepper
Motor
Basics
! Circuit
Design
Using
VHDL,
Pedroni
! Spartan
3E
user
guide
! “A
comparison
of
the
ILD
and
TDOA
sound
source
LocalizaFon
algorithms
in
a
train
environment”
! “Reliability
measurement
of
Fme
difference
of
arrival
esFmaFons
for
mulFple
sound
source
localizaFon”