Inter
national
J
our
nal
of
Electrical
and
Computer
Engineering
(IJECE)
V
ol.
8,
No.
5,
October
2018,
pp.
3250
–
3258
ISSN:
2088-8708
3250
I
ns
t
it
u
t
e
o
f
A
d
v
a
nce
d
Eng
ine
e
r
i
ng
a
nd
S
cie
nce
w
w
w
.
i
a
e
s
j
o
u
r
n
a
l
.
c
o
m
W
iRoT
ip:
an
IoT
-based
W
ir
eles
s
Sensor
Netw
ork
f
or
W
ater
Pipeline
Monitoring
F
atma
Karray
1,2
,
Mariem
T
riki
2
,
Mohamed
W
assim
Jmal
1,2
,
Mohamed
Abid
1,2
,
and
Abdulfattah
M.
Obeid
3
1
Computer
and
Embedded
System
laboratory
,
National
Engineering
School
of
Sf
ax,
Uni
v
ersity
of
Sf
ax,
Sf
ax,
T
unisia
2
Digital
Research
Center
of
Sf
ax,
T
echnopole
of
Sf
ax,
Sf
ax,
T
unisia
3
National
Center
for
Electronics
and
Photonics
T
echnology
,
King
Abdulaziz
City
for
Science
and
T
echnology
(KA
CST),
Riyadh,
Saudi
Arabia
Article
Inf
o
Article
history:
Recei
v
ed
October
31,
2017
Re
vised
June
29,
2018
Accepted
July
14,
2018
K
eyw
ord:
W
ireless
Sensor
Netw
ork
Internet
of
Things
Node
Design
Leak
Detection
W
ater
Pipeline
Monitoring
ABSTRA
CT
One
of
the
k
e
y
components
of
the
Internet
of
Things
(IoT)
is
the
W
ire
less
Sensor
Netw
ork
(WSN).
WSN
is
an
ef
fecti
v
e
and
ef
ficient
technology
.
It
consists
of
senor
nodes;
smart
de
vic
es
that
allo
ws
data
collection
and
pre-processing
wirelessly
from
real
w
orld.
Ho
we
v
er
,
issues
related
to
po
wer
consumption
and
computational
per
-
formance
still
persist
in
classical
wireless
nodes
since
po
wer
is
not
al
w
ays
a
v
ailable
in
application
lik
e
pipeline
monitoring.
Moreo
v
er
,
the
y
could
not
be
usually
suitable
and
adequate
for
this
kind
of
application
due
to
memory
shortage
and
performance
constraints.
Designing
ne
w
IoT
WSN
system
that
matches
the
application
specific
re-
quirements
is
e
xtremely
important.
In
this
paper
,
we
present
W
iRoT
i
p,
a
WSN
node
prototype
for
w
ater
pipeli
ne
application.
An
e
xperimental
and
a
comparati
v
e
studies
ha
v
e
been
performed
for
the
dif
fe
rent
node’
s
components
to
achie
v
e
a
final
adequate
design.
Copyright
©
2018
Institute
of
Advanced
Engineering
and
Science
.
All
rights
r
eserved.
Corresponding
A
uthor:
F
atma
Karray
Computer
Science
department
Sf
ax,T
unisia
+21644295345
karray
.f
atma.enis@gmail.com
1.
INTR
ODUCTION
Internet
of
Things
(IoT)
allo
ws
us
to
transform
the
w
ay
of
our
perception
and
our
interaction
with
the
real
w
orld.
It
w
ould
mak
e
applications
g
ain
more
ef
ficienc
y
,
harness
intelligence
and
get
better
accurac
y
by
linking
the
ph
ysical
objects
to
the
information
netw
ork.
It
of
fers
also
a
promising
solution
of
v
arious
e
xisting
industrial
systems
such
as
w
ater
transportation
systems,
manuf
acturing
systems,
etc
[1].
W
ireless
Sensor
Netw
orks
(WSNs)
play
a
major
role
in
this
technology
as
intermediate
to
shape
the
ph
ysical
w
orld
to
human
perception.
No
w
adays,
WSN’
applications
are
getting
more
and
more
attention
from
the
in-
dustrial
and
the
academic
circles
[2]
[3]
[4].
One
of
the
most
crucial
application
of
WSN
is
w
ater
pipeline
monitoring
since
w
orries
about
potable
w
ater
ha
v
e
became
more
and
more
justified
[5].
In
that
line,
man
y
studies
ha
v
e
been
made
to
propose
solutions
for
leak
detection
and
location
in
w
ater
pipeline.
Most
of
them
are
focusing
on
the
softw
are
aspect
such
as
leak
detection
algorithms,
communication
protocols
etc.
Fe
w
others
are
tar
geting
the
hardw
are
parts
such
as
sensors,
WSN
platforms,
etc
[6].
Ho
we
v
er
,
the
most
common
concern
of
WSN
is
po
wer
consumption
since
it
determines
the
lifespan
of
the
whole
application.
The
battery-
p
o
we
red
sensors
are
responsible
for
g
athering
information
and
detecting
leaks
in
order
to
react
at
the
appropriate
time.
Therefore,
in
addition
to
ener
gy
preserv
ation,
enhancing
the
sensing
capabilities
by
ameliorating
the
output
signal
of
these
sensors
and
treating
their
information
is
quite
important
when
dealing
with
WSNs.
J
ournal
Homepage:
http://iaescor
e
.com/journals/inde
x.php/IJECE
I
ns
t
it
u
t
e
o
f
A
d
v
a
nce
d
Eng
ine
e
r
i
ng
a
nd
S
cie
nce
w
w
w
.
i
a
e
s
j
o
u
r
n
a
l
.
c
o
m
,
DOI:
10.11591/ijece.v8i5.pp3250-3258
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISSN:
2088-8708
3251
IoT
is
not
e
xplored
e
xtensi
v
ely
in
w
ater
pipeline
monitoring.
Fe
w
w
orks
e
xist
for
w
ater
management
in
general.
F
or
instance,
the
authors
in
[7]
sho
w
the
importance
of
IoT
for
w
ater
resource
engineering.
Some
literature
w
orks
are
reported
in
this
w
ork.
In
[8],
the
authors
describe
t
he
importance
of
IoT’
s
usage
for
w
ater
manage-
ment
companies
and
the
w
ay
of
IoT
inte
gration
for
this
management.
Mohod
[9]
in
the
same
w
ay
af
firms
the
feasibility
of
IoT
inte
gration
for
Dam
and
W
ater
Management.
The
authors
in
[10]
present
an
IoT
wireless
sensor
node
for
leak
detection
and
w
ater
quality
monitoring.
The
prototype
consists
of
a
microcontroller
,
a
PH
sensor
,
a
vibration
sensor
,
a
flo
w
sensor
and
a
le
v
el
sensor
.
Dhobale
et
al.
[11]
propose
an
IoT
WSN
system
for
w
ater
supply
management.
This
system
serv
es
to
automatically
measure
w
ater
le
v
el
in
dam
and
w
ater
flo
w
rate.
The
sensor
node
is
based
on
w
ater
flo
w
sensor
,
Arduino
board,
ultrasound
sensor
and
GSM
module.
The
authors
in
[12]
present
a
WSN
prototype
for
w
aterw
aste
monitoring
on
IoT
.
The
proposed
design
consists
of
a
W
ireless
sensor
node,
a
g
ate
w
ay
node,
a
SMS-g
ate
w
ay
and
IoT
Cloud
platform.
The
authors
describe
The
implementations
of
each
part.
The
sensor
node
is
composed
of
an
Arduino
Me
g
a
board,
a
wireless
commu-
nication
module,
a
sensor
interf
ace,
a
PH
sensor
,
a
conducti
vity
sensor
and
a
dissolv
ed
oxygen
sensor
.
There
is
no
po
wer
consumption
or
performance
e
v
aluation
in
the
proposal.
Nguyen
et
al.
[13]
describe
an
IoT
WSN
node
for
En
vironment
monitori
n
g.
The
node
consists
of
nRF51822
System
on
Chip
(SoC)
that
contains
an
ARM
corte
x
M0
microcontroller
and
an
ener
gy
harv
esting
module.
Y
ang
et
al.
[14]
propose
a
WSN
for
w
ater
consumption
monitoring
at
a
household
using
IoT
concept.
Ho
we
v
er
,
these
w
orks
are
considered
as
attempts
to
use
the
IoT
concept
for
w
ater
management
systems
and
further
impro
v
ements
are
needed
especially
in
terms
of
performance
and
ener
gy
optimization.
The
object
of
this
w
ork
is
to
propose
and
e
v
aluate
W
iRoT
ip,
an
IoT
ener
gy-ef
ficient
WSN
p
r
ototype
for
leak
detection
in
w
ater
pipelines.
Moreo
v
er
,
we
propose
dif
ferent
circuits
of
signal
conditioning
as
well
as
a
h
ybrid
leak
detection
algorithm
based
on
kalman
filter
used
in
data
processing.
This
paper
is
or
g
anized
as
follo
ws:
In
section
2,
we
present
the
W
iRoT
ip.
W
e
introduce
the
proposed
IoT
architecture
of
the
system,
the
softw
are
algorithm
and
the
node
design.
W
e
dra
w
the
e
xperimental
results
in
section
3
to
finally
finish
with
a
conclusion
and
further
future
w
ork
in
section
4.
2.
WIR
O
TIP
SYSTEM
DESIGN
In
this
section,
we
will
detail
our
proposed
solution.
The
architecture
of
IoT
system
will
be
pres
ented.
Softw
are
and
hardw
are
implementations
of
the
sensor
node
will
be
described.
2.1.
IoT
structur
e
of
the
pr
oposed
system
The
W
iRoT
ip
system
is
designed
for
pressurized
pipes.
The
proposed
IoT
architecture
consist
s
of
multi-layers
that
interact
and
cooperate
to
detect
and
to
locate
leaks
in
w
ater
pipelines.
Figure
1
sho
ws
the
components
of
dif
ferent
layers.
The
First
lay
er
is
WSN
layer
in
which
the
data
is
collected
and
pre-processed
locally
(not
in
the
serv
er).
In
f
act,
a
pre-processing
in
node
could
sa
v
e
ener
gy
dissipated
in
frequent
data
transmission
or
sending
useless
information.
Hence,
a
h
ybrid
leak
detection
m
ethod
based
on
Kalman
filter
(HLDKF)
is
implemented
to
detect
leaks
in
w
ater
pipes.
The
sensor
node
in
our
case
allo
ws
dif
ferent
tasks
lik
e
data
filtering,
data
processing,
data
compressing,
data
fusion,
dat
a
aggre
g
ation,
etc.
After
softw
are
implementation,
a
hardw
are
e
xperimental
study
is
performed
to
select
and
design
the
dif
ferent
components
of
the
node
in
subsection
2.3..
The
second
lay
er
is
the
netw
orking,
the
service
and
the
storage
layer
.
In
this
layer
,
the
communication
between
nodes,
g
ate
w
ays
and
the
base
station
is
performed.
The
sensors
are
fix
ed
in
sleep
mode
and
get
data
e
v
ery
8
hours.
The
HLDF
is
run
to
filter
data
and
to
test
leak
occurrence.
When
a
leak
occurs,
the
data
is
collected
with
a
high
sampling
rate.
The
compressed
data
and
leak
information
are
firstly
forw
arded
from
nodes
to
cluster
heads,
in
which
the
leak
position
is
calculated,
and
then
to
the
cloud.
The
final
lay
er
is
the
application
layer
in
which
the
user
could
interact
with
the
sensors’
information.
In
this
step,
v
arious
analyses
are
performed
and
visualized
online.
An
interacti
v
e
user
interf
ace
is
de
v
eloped
to
access
the
pipelines
information.
In
the
application,
users
are
allo
wed
to
access
to
leaks
information,
statistics,
graphs,
pipeline
state
and
netw
ork
information.
2.2.
W
iRoT
ip
Leak
detection
and
data
filtering
module
Kalman
filter
(KF)
[15]
is
an
ef
fi
cient
predicti
v
e
and
estimator
.
The
usage
of
such
algorithm
j
ointly
with
WSN
has
not
been
e
xplored
yet
for
w
ater
pipeline
application
to
the
best
of
our
kno
wledge.
Ho
we
v
er
,
some
papers
ha
v
e
used
KF
for
lea
k
detection.
F
or
e
xample,
Benkherouf
in
1988
proposes
an
Extended
KF
Evaluation Warning : The document was created with Spire.PDF for Python.
3252
ISSN:
2088-8708
Figure
1.
W
iRoT
ip
IoT
architecture
(EKF)
in
the
conte
xt
of
pipeline
monitoring
[16].
The
authors
in
[17]
suggest
a
linear
KF
for
leak
detection
based
on
pressure
and
flo
w
measurements.
The
y
s
up
pos
e
that
in
a
gi
v
en
time
step,
the
measurement
state
is
similar
with
the
one
of
pre
vious
week.
Jung
et
al
[18]
propose
a
leak
detection
method
based
on
a
statistical
process
control
and
a
KF
to
detect
b
ursts
in
pipelines.
Ho
we
v
er
,
these
w
orks
do
not
e
xplore
the
KF
in
the
conte
xt
of
WSN-WPM
application.
W
e
implement
a
HLDKF
to
perform
data
filtering
and
leak
detection
in
w
ater
pipes.
Figure
??
sho
ws
The
flo
wchart
of
the
algorithm.
KF
is
a
recursi
v
e
data
processing
algorithm
for
dynamic
systems.
It
emplo
ys
a
set
of
mathematical
equations
to
produce
an
optimal
estimation
of
the
system
[19].
This
algorithm
with
be
used
in
the
ne
xt
section
for
e
v
aluating
the
performance
of
the
processing
module.
2.3.
W
iRoT
ip
Node
design
T
o
design
an
ener
gy-ef
ficient
node
and
to
perform
a
good
choice
of
the
node
components,
a
t
heoretical
and
e
xperimental
study
about
the
node
components
is
necessary
.
A
typical
sensor
node
consists
of
four
modules
[20]:
•
The
pr
ocessing
unit:
F
or
the
processing
unit,
we
ha
v
e
used
an
Arduino
Uno
[21].
This
platfor
m
is
open
hard
and
f
acilitated
hardw
are
understanding.
It
allo
ws
also
easy
inte
gration
of
sensors
and
communica-
tion
de
vices
(transcei
v
ers).
•
The
communication
unit:
This
unit
is
composed
of
a
transcei
v
er
that
transmits
and
recei
v
es
data
wire-
lessly
.
A
nRF24l01
is
used
for
this
unit
as
sho
wn
in
Figure
2.
The
nRF24l01
[22]
is
a
2.4
GHz
Radio
T
ranscei
v
er
.
The
choice
of
this
transcei
v
er
is
due
to
its
lo
w
po
wer
,
its
lo
w
cost
and
its
compatibility
with
Arduino
board,
yet,
for
final
prototype
or
product,
nRF24L01+
or
the
SoCs
nRF24LE1
or
nRF24LU1+
are
more
suitable
[22].
F
or
the
g
ate
w
ay
,
we
ha
v
e
used
the
Ethernet
communication.
Ethernet
(also
kno
wn
as
the
IEEE
802.3
standard)
[23]
is
a
standard
for
data
transmission
for
local
area
netw
ork.
W
e
ha
v
e
used
for
our
prototypes
Arduino
Ethernet
Shield
V1
[24]
to
connect
the
Arduino
to
the
internet
(g
ate
w
ay)
as
sho
wn
in
Figure
2.
The
acquisition
of
W
ifi
shield
w
as
not
possible.
That
is
wh
y
,
we
ha
v
e
used
Ethernet
shield.
•
The
sensing
unit:
This
unit
is
in
char
ge
of
g
athering
data
from
ph
ysical
en
vironment.
This
part
is
v
ery
crucial
as
the
accurac
y
of
an
y
sensor
will
af
fect
the
all
system
(to
design
sturdy
system).
Press
ure
and
IJECE
V
ol.
8,
No.
5,
October
2018:
3250
–
3258
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISSN:
2088-8708
3253
Figure
2.
Ethernet
T
ranscei
v
er
connected
to
Arduino
Uno
board
Flo
w
sensors
are
used
as
inputs
of
the
HLDKF
as
detailed
belo
w:
1.
YF-S201
Hall
Effect
W
ater
Flo
w
Sensor:
This
sens
or
is
used
to
measure
the
flo
w
v
ariations
in
w
ater
pipes
[25].
It
is
a
destructi
v
e
sensor
aligned
with
the
w
ater
pipe.
The
accurac
y
of
this
sensor
is
about
+/-
10
%
and
the
flo
w
rate
v
aries
from
1
to
30
liters
per
minute.
This
sensor
is
widely
emplo
yed
due
to
its
puls
e-based
mechanism
that
allo
ws
lo
w
po
wer
consumption.
Figure
3
sho
ws
the
w
ay
in
which
this
sensor
is
attached
to
the
pipe
and
to
the
sensor
node.
Figure
3.
YF-S201
Hall
Ef
fect
W
ater
Flo
w
Sensor
attached
to
our
demonstrator
and
to
the
sensor
node
2.
F
or
ce
Sensiti
v
e
Resistor
(FSR)
sensor:
FSR
is
an
analog
sensor
used
to
measure
pressure
in
w
ater
pipes.
It
is
a
polymer
thick
film
de
vice
characterized
by
its
easy-to-use
and
lo
w
cost
[26].
Figure
4
illustrates
the
dif
ferent
parts
of
the
sensor
.
When
a
force/pressure
is
e
x
ercised
to
the
sensor
,
the
resistor
element
is
deformed
and
the
air
is
pushed
from
the
spacer
.
The
accurac
y
of
sensor
is
+
=
10
%.
Figure
4.
FSR
sensor
In
our
case,
the
sensor
is
attached
to
the
outside
of
the
pipeline
and
fix
ed
with
a
join.
The
pipeline
pressure
produces
a
contact
force
between
the
pipe
and
the
join.
When
a
leak
occurs,
i
t
causes
pressure
v
ariations
which
af
fect
the
contact
force.
•
The
po
wer
management
unit
manages
and
pro
vides
ener
gy
to
the
all
sensor
node
components.
In
this
paper
,
we
ha
v
e
not
w
ork
ed
on
this
unit.
Ho
we
v
er
,
some
ener
gy
optimization
techni
ques
are
used
to
sa
v
e
the
all
po
wer
of
the
node.
This
study
is
v
ery
important
to
master
the
hardw
are
part
and
select
the
best
components
of
the
IoT
node.
Evaluation Warning : The document was created with Spire.PDF for Python.
3254
ISSN:
2088-8708
3.
RESUL
TS
AND
AN
AL
YSIS
The
W
iRoT
ip
prototype
is
tested
using
a
demonstrator
installed
in
our
research
center
[27].
Figure
5
sho
ws
an
almost
rectangular
section
composed
of
25
m
polyeth
ylene
pipes.
These
pipes
ha
v
e
32
mm
as
an
Figure
5.
W
iRoT
ip
T
estbed
e
xternal
diameter
.
The
y
support
up
to
12
bar
of
pressure.
The
choice
of
this
sort
of
pipes
is
thanks
to
their
lo
w
cost,
their
resistance
and
insensiti
vity
to
chemical
and
electrical
corrosion.
Furthermore,
the
y
are
used
in
the
real
distrib
ution
systems
of
our
country
.
More
general,
the
use
of
plastic
pipes
has
increasingly
widespread
all
o
v
er
the
w
orld.
The
setup
consists
also
of
tw
o
v
alv
es
in
inlet
and
outlet
points
in
order
to
v
ary
the
users
demands
by
v
arying
the
pressure.
A
1000
m
3
reserv
oir
is
used
as
a
w
ater
source.
T
o
control
the
inlet
and
outlet
w
ater
,
we
emplo
y
tw
o
flo
w
meters.
As
the
pipes
are
made
at
the
same
le
v
el,
the
w
ater
is
mo
ving
along
the
pipes
by
an
electrical
pump
with
1
hp
motor
pro
viding
up
to
4
bar
when
the
output
v
alv
e
is
closed
and
up
to
2.5
bar
in
open
circuit.
The
supports
are
designed
to
ha
v
e
v
ariable
heights
that
we
will
e
xplore
in
the
future
to
see
the
ef
fect
of
this
v
ariation
on
the
pressure
and
to
test
our
algorithm
in
v
aried
conditions.
Finally
,
The
leaks
are
induced
using
tw
o
g
arden
taps.
This
demonstrator
is
used
to
test
the
proposed
sensor
node.
The
prototype
is
made
up
of
the
Arduino
board,
the
nRF24l01
transcei
v
er
,
the
flo
w
sensor
and
the
relay
.
T
able
1
il
lustrates
also
the
po
wer
consumption
distrib
ution
of
the
node
prototype.
In
this
table,
each
of
t
he
node
components
is
shut
do
wn
to
see
it
ef
fect
in
the
whole
po
wer
of
the
node.
T
able
1.
Po
wer
distrib
ution
in
the
Arduino-nRF24l01
prototype
Arduino
Relay
Flo
w
sensor
algorithm
nRF24l01
Current
(mA)
Po
wer
(mW)
on
on
on
HLDKF
Tx
(11
mA)
85.47
396.3
on
on
on
HLDKF
Rx
(18
mA)
92.47
419.4
on
on
Of
f
HLDKF
Idle
(2
mA)
34
381.5
on
of
f
of
f
HLDKF
of
f
32
160
sleep
(
PWR
DO
WN)
of
f
of
f
HLDKF
of
f
26
130
T
able
2
illustrates
the
po
wer
consumption
of
the
sensors
used
for
W
iRoT
ip.
These
sensors
are
lo
w
po
wer
and
the
y
k
ept
shut
do
wn
as
longer
as
possible
to
sa
v
e
the
po
wer
of
the
node.
T
able
3
represents
the
po
wer
distrib
ution
analysis
of
the
g
ate
w
ay
prototype
which
is
based
on
Ethernet
shield
V1.
W
e
note
that
the
Ethernet
module
has
high
po
wer
consumption.
Ho
we
v
er
,
it
is
used
due
to
material
con-
straints.
Figure
6
summarize
the
po
wer
profiles
of
W
iRoT
ip
node
and
Gate
w
ay
.
W
e
remember
that
thi
s
proposal
is
a
prototype
for
more
ener
gy
sa
ving
a
PCB
board
needs
to
be
designed.
Moreo
v
er
,
due
to
the
lack
of
information
of
po
wer
consumption
in
the
other
approaches,
we
implement
our
algorithm
in
tw
o
other
sensor
nodes:
Arduino
Due
and
MKR1000
to
compare
our
w
ork
with
others
as
repre-
IJECE
V
ol.
8,
No.
5,
October
2018:
3250
–
3258
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISSN:
2088-8708
3255
T
able
2.
Experimental
Po
wer
Consumption
of
each
Sensor
Sensor
V
oltage
(V)
Curr
ent
(mA)
P
o
wer
(mW)
YF-S201
sensor
5
2.47
12.35
FSR
5
2.3
11.5
T
able
3.
Po
wer
distrib
ution
in
the
Arduino-Ethernet
prototype
Arduino
Relay
Ethernet
mode
Current
(mA)
Po
wer
(mW)
on
on
Tx
(167
mA)
239
1195
on
on
Rx
(160
mA)
232
1160
on
on
Idle
(150
mA)
222
1110
Figure
6.
W
iRoT
ip
Po
wer
Consumption
Summary
sented
in
T
able
4.
As
we
can
see,
in
the
T
able
4,
our
proposal
still
ha
v
e
the
lo
west
po
wer
consumption
while
impro
v
ements
still
needed
in
this
respect.
T
able
4.
Comparison
of
W
iRoT
ip
with
other
Approaches
T
ime
(
s)
Po
wer
(mW)
W
iRoT
ip
80
187.5
Arduino
Due
30
557
MKR1000
40
600.5
4.
CONCLUSION
In
this
paper
,
we
de
v
elop
an
IoT
WSN
node
prototype
for
w
ater
pipeline
monitoring
appl
ication.
V
arious
tests
and
implementations
ha
v
e
been
performed
for
dif
ferent
units
to
design
an
ener
gy
a
w
are
reliable
system.
The
sensing
unit
w
as
a
crucial
unit.
In
f
act,
calibrations
and
amplifier
ha
v
e
been
added
to
adjust
the
signal
coming
from
polyeth
yle
ne
pipes.
Communic
ation
and
po
wer
management
techniques
ha
v
e
bee
n
also
e
v
aluated.
All
this
w
ork
has
permit
us
to
design
and
e
v
aluate
W
iRoT
ip
node.
As
future
w
ork,
a
PCB
board
will
be
de
v
eloped
to
get
our
o
wn
ultra
lo
w
po
wer
product
with
sensor
board
e
xtension.
T
ests
and
e
xperiments
will
be
performed
not
only
in
the
demonstrator
b
ut
also
in
real
field.
Evaluation Warning : The document was created with Spire.PDF for Python.
3256
ISSN:
2088-8708
A
CKNO
WLEDGMENT
The
authors
w
ould
lik
e
to
thank
the
King
Abdulaziz
City
for
Science
and
T
echnology
(KA
CST)
which
supports
this
w
ork
under
a
research
grant
(project
no.
35/1012).
REFERENCES
[1]
A.
Rghioui
and
A.
Oumnad,
“Internet
of
things:
Surv
e
ys
for
measuring
human
acti
vities
from
e
v
ery-
where,
”
International
J
ournal
of
Electrical
and
Computer
Engineering
(IJECE)
,
v
ol.
7,
no.
5,
p.
2474,
2017.
[2]
C.-Y
.
Chong
and
S.
P
.
K
umar
,
“Sensor
netw
orks:
e
v
olution,
opportunities,
and
challenges,
”
Pr
oceedings
of
the
IEEE
,
v
ol.
91,
no.
8,
pp.
1247–1256,
2003.
[3]
A.
A.
Jaber
and
R.
Bick
er
,
“Design
of
a
wireless
sensor
node
for
vibration
monitoring
of
industrial
ma-
chinery
,
”
International
J
ournal
of
Electrical
and
Computer
Engineering
,
v
ol.
6,
no.
2,
p.
639,
2016.
[4]
H.
Sayuti,
R.
Rashid,
N.
A.
Latif
f,
M.
A.
Rahim,
A.
Abdul,
M.
Sarijari
,
and
N.
Ghazali,
“Smart
home
and
ambient
assisted
li
ving
based
on
the
internet
of
things,
”
International
J
ournal
of
Electrical
and
Computer
Engineering
,
v
ol.
7,
no.
3,
p.
1480,
2017.
[5]
M.
M.
Mek
onnen
and
A.
Y
.
Hoekstra,
“F
our
billion
people
f
acing
se
v
ere
w
ater
scarcity
,
”
Science
ad-
vances
,
v
ol.
2,
no.
2,
p.
e1500323,
2016.
[6]
F
.
Karray
,
A.
Garcia-Ortiz,
M.
W
.
Jmal,
A.
M.
Obeid,
and
M.
Abid,
“Earnpipe:
A
testbed
for
smart
w
ater
pipeline
monitoring
using
wireless
sensor
netw
ork,
”
Pr
ocedia
Computer
Science
,
v
ol.
96,
pp.
285–294,
2016.
[7]
S.
Deshmukh
and
P
.
Barapatr
,
“Internet
of
things
based
system
for
w
ater
resource
engineering,
”
in
Em-
anations
in
Modern
T
ec
hnolo
gy
and
Engineering
(ICEMTE),
2017
International
Confer
ence
on
,
v
ol.
5.
IEEE,
2017,
pp.
240–242.
[8]
T
.
Robles,
R.
Alcarria,
D.
M.
de
Andr
´
es,
M.
Na
v
arro,
R.
Calero,
S.
Iglesias,
and
M.
L
´
opez,
“
An
iot
based
reference
architecture
for
smart
w
ater
management
processes,
”
J
oWU
A
,
v
ol.
6,
no.
1,
pp.
4–23,
2015.
[9]
N.
Mohod,
“Usability
of
internet
of
things
[iot]
for
dam
safety
and
w
ater
management,
”
International
J
ournal
of
Resear
c
h
in
Advent
T
ec
hnolo
gy
,
v
ol.
5,
no.
1,
pp.
29–30,
2017.
[10]
S.
Ezhilv
anji
and
S.
Malark
odi,
“
An
ef
ficient
w
ater
distrib
ution
system
for
india
using
iot,
”
International
Inno
vative
Resear
c
h
J
ournal
of
Engineering
and
T
ec
hnolo
gy
,
2017.
[11]
G.
S.
P
.
G.
P
.
J.
H.
K.
A.
Dhobale,
Kiran
M.
and
U.
A.
Mande,
“Iot
based
smart
w
ater
supply
management
system,
”
Advanced
Resear
c
h
in
Computer
and
Communication
Engineering
,
v
ol.
6,
no.
3,
2017.
[12]
Y
.
Zakaria
and
K.
Michael,
“
An
inte
grated
cloud-based
wireless
sensor
netw
ork
for
monitoring
industrial
w
aste
w
ater
dischar
ged
into
w
ater
sources,
”
W
ir
eless
Sensor
Network
,
v
ol.
9,
no.
08,
p.
290,
2017.
[13]
C.
M.
Nguyen,
J.
Mays,
D.
Plesa,
S.
Rao,
M.
Nguyen,
and
J.-C.
Chiao,
“W
ireless
sensor
nodes
for
en
vironmental
monitoring
in
internet
of
things,
”
in
Micr
owave
Symposium
(IMS),
2015
IEEE
MTT
-S
International
.
IEEE,
2015,
pp.
1–4.
[14]
S.-H.
Y
ang,
X.
Chen,
X.
Chen,
L.
Y
ang,
B.
Chao,
and
J.
Cao,
“
A
case
study
of
internet
of
things:
A
wireless
household
w
ater
consumption
monitoring
system,
”
in
Internet
of
Things
(WF-IoT),
2015
IEEE
2nd
W
orld
F
orum
on
.
IEEE,
2015,
pp.
681–686.
[15]
H.
Alra
w
ashdeh
and
J.
A.
Sumadu,
“The
kalman
filter
performance
for
dynamic
change
in
system
parameters,
”
International
J
ournal
of
Electrical
and
Computer
Engineering
(IJECE)
,
v
ol.
3,
no.
6,
pp.
713–723,
2013.
[Online].
A
v
ailable:
http://iaesjournal.com/online/inde
x.php/IJECE/article/vie
w/3857
[16]
A.
Benkherouf
and
A.
Allidina,
“Leak
detection
and
location
in
g
as
pipelines,
”
in
IEE
Pr
oceedings
D
(Contr
ol
Theory
and
Applications)
,
v
ol.
135.
IET
,
1988,
pp.
142–148.
[17]
G.
Y
e
and
R.
A.
Fenner
,
“Kalman
filtering
of
h
ydraulic
measurements
for
b
urst
detection
in
w
ater
distri-
b
ution
systems,
”
J
ournal
of
pipeline
systems
engineering
and
pr
actice
,
v
ol.
2,
no.
1,
pp.
14–22,
2010.
[18]
D.
Jung
and
K.
Lanse
y
,
“W
ater
distrib
ution
system
b
urst
detection
using
a
nonli
near
kalman
filter
,
”
J
ournal
of
W
ater
Resour
ces
Planning
and
Mana
g
ement
,
v
ol.
141,
no.
5,
p.
04014070,
2014.
[19]
R.
E.
Kalman,
“
A
ne
w
approach
to
linear
filtering
and
prediction
problems
,
”
J
ournal
of
basic
Engineering
,
v
ol.
82,
no.
1,
pp.
35–45,
1960.
[20]
F
.
Karray
,
W
.
M.
Jmal,
M.
Abid,
D.
Houssaini,
A.
M.
Obeid,
S.
M.
Qasim,
and
M.
S.
BenSaleh,
“
Archi-
tecture
of
wireless
sensor
nodes
for
w
ater
monitoring
applications:
From
microcontroller
-based
system
to
soc
solut
ions,
”
in
En
vir
onmental
Instrumentation
and
Measur
ements
(IMEK
O),
2014
5th
IMEK
O
TC19
Symposium
on
,
2014,
pp.
20–24.
[21]
Arduino.
Arduino/genuino
uno
description.
[Online].
A
v
ailable:
IJECE
V
ol.
8,
No.
5,
October
2018:
3250
–
3258
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISSN:
2088-8708
3257
https://www
.arduino.cc/en/Main/ArduinoBoardUno
[22]
nRF24L01
Single
Chip
2.4GHz
T
r
ansceiver
Pr
oduct
Specification
,
Nordic
semiconductor
,
Jul.
2007.
[23]
Z.
Lin
and
S.
Pearson,
“
An
inside
look
at
industrial
ethernet
communication
protocols,
”
T
e
xas
Instruments,
White
P
aper
,
2013.
[24]
Arduino.
Arduino
ethernet
shiel
d
v1.
[Online].
A
v
ailable:
https://www
.arduino.cc/en/Main/ArduinoEthernetShieldV1
[25]
Data-sheet,Pr
odcut
Intr
oduction
,
YIF
A
the
plastics
Ltd.
[26]
F
.
S.
R.
A.
O.
of
the
T
echnology
,
“Data-sheet,
”
Interlink
Electr
onics
.
[27]
Digital
research
centre
of
sf
ax.
[Online].
A
v
ailable:
http://www
.crns.rnrt.tn/
Evaluation Warning : The document was created with Spire.PDF for Python.
3258
ISSN:
2088-8708
BIOGRAPHIES
OF
A
UTHORS
F
atma
Karray
is
a
Phd
student
at
the
Computer
and
Embedded
Systems
(CES)
laboratory
and
at
the
Digital
Research
Center
of
Sf
ax,
T
unisia
(2014).
She
has
recei
v
ed
the
Engineering
de
gree
in
embedded
systems
from
the
National
School
of
Engineer
of
Sf
ax,
T
unisia
in
2013.
Her
current
research
interests
are
in
the
fields
of
embedded
systems,
W
irel
ess
Sensor
Netw
orks,
Signal
processing,
IoT
applications,
Smart
Infrastructures
monitoring,
System
on
Chip
design
and
Lo
w
po
wer
design.
F
atma
Karray
serv
es
as
a
re
vi
e
wer
in
international
journals
and
c
onferences
and
as
or
g
anizer
in
national
e
v
ents
and
w
orkshops.
Mariam
T
riki
Electrical
engineer
from
the
national
school
of
engineering
of
Sf
ax
,
T
unisia
(2016).
She
specialized
in
electronics
and
ne
w
technologies.
She
w
ork
ed
as
an
RD
engineer
at
the
Computer
and
Embedded
Systems
Laboratory
(CES
labs)
and
at
the
Digital
Research
Center
of
Sf
ax.
She
is
passionate
about
IoTs,
sensors,
analog
systems
and
electronic
de
vices.
Mohamed
W
assim
Jmal
is
an
Associate
Professor
at
the
Higher
Institute
of
Applied
Sciences
and
T
ecnology
of
Gafsa,
T
unisia
since
2012.
Hi
s
research
acti
vity
is
conducted
within
CES
Laboratory
.
He
has
recei
v
ed
the
Engineering
de
gree
in
Electrical
Engineering,
from
the
National
Engineering
School
of
Sf
ax
in
2005
and
the
Master
de
gree
in
Automat
ic
and
Industrial
Informatics,
from
the
same
Engineering
School,
in
2007.
He
got
the
PhD
in
Electrical
Engineering
in
2013.
His
current
research
interests
are
in
the
field
of
W
ireless
Sensor
Netw
orks
(WSN)
and
the
Embedded
Systems.
The
y
are
focused
on
the
implementation
of
wireless
s
ensor
netw
orks
applications
in
Reconfigurable
System.
He
has
se
v
eral
publications
in
man
y
conferences
and
Journals.
Mohamed
W
assim
JMAL
serv
ed
in
national
and
international
conference
or
g
anization:
IDT
,
ICM,
TWESD,
SensorNets.
Mohamed
Abid
is
Head
of
”Computer
Em
bedded
System”
laboratory
CES-ENIS,
T
unisia.
He
is
w
orking
no
w
as
a
Professor
at
the
Engineering
National
School
of
Sf
ax
(ENIS),
Uni
v
ersity
of
Sf
ax,
T
unisia.
He
recei
v
ed
the
PhD
de
gree
from
the
National
Institute
of
Applied
Sciences,
T
oulouse
(France)
in
1989
and
the
”thse
d’tat”
de
gree
from
the
National
School
of
Engineering
of
T
unis
(T
unisia)
in
2000
in
the
area
of
Computer
Engineering
and
Microelectronics.
His
current
research
interests
include
hardw
are/softw
are
co-design,
System
on
Chip,
Reconfigurable
System,
and
Em-
bedded
System,
biometric,
etc.
He
has
also
in
v
estig
ated
the
design
and
implementation
issues
of
FPGA
embedded
systems.
Dr
.
Abid
is
joint
coordinator
or
an
acti
v
e
member
of
se
v
eral
Interna-
tional
Research
and
Inno
v
ation
projects.
He
w
as
Supervisor
or
Co-supervisor
of
more
than
50
PhD
doctors.
He
is
author
or
co-author
of
more
than
150
publications
in
Journals
and
of
more
than
300
papers
in
international
conferences.
He
is
also
author
or
c
o-
author
of
man
y
guest’
s
papers,
Joint
author
of
man
y
book’
s
chapters.
Dr
.
Abid
has
serv
ed
a
lso
as
Guest
professor
at
se
v
eral
international
uni
v
ersities
and
as
a
Consultant
to
research
and
de
v
elopment
in
T
elnet
Incorporation.
Abdulfattah
M.
Obeid
is
currently
general
manager
at
Saudi
T
echnology
De
v
elopment
and
In
v
est-
ment
Compan
y
(T
A
QNIA)and
associated
profess
or
at
National
Center
for
Electronics
and
Photonics
T
echnology
,
King
Abdulaziz
City
for
Science
and
T
echnology
(KA
CST),
Riyadh,
Saudi
Arabia.
He
has
recei
v
ed
the
Bachelor
of
Science
(B
Sc)
Electrical,
Electronics
and
Communications
Engineer
-
ing
in
King
Saud
Uni
v
ersity
in
1994,
the
Master
de
gree
of
Electrical
Engineering
in
Michig
an
State
Uni
v
ersity
,
East
Lansi
ng,
MI,
USA
in
1999
and
the
PhD
de
gree
in
Electrical
Engineering
and
In-
formation
T
echnology
,
TU-Darmstadt,
Darmstadt,
German
y
in
2006.
His
current
research
interests
are
in
the
fields
of
embedded
systems,
W
ireless
Sensor
Netw
orks,
System
on
Chip
design
and
Lo
w
po
wer
design.
Obeid
serv
es
as
a
re
vie
wer
in
international
journals
and
conferences
and
as
or
g
anizer
in
international
conferences
and
w
orkshops.
IJECE
V
ol.
8,
No.
5,
October
2018:
3250
–
3258
Evaluation Warning : The document was created with Spire.PDF for Python.