Indonesi
an
Journa
l
of El
ect
ri
cal Engineer
ing
an
d
Comp
ut
er
Scie
nce
Vo
l.
12
,
No.
3
,
Decem
ber
201
8
, p
p.
1030
~
1036
IS
S
N: 25
02
-
4752, DO
I: 10
.11
591/ijeecs
.v1
2
.i
3
.pp
1030
-
1036
1030
Journ
al h
om
e
page
:
http:
//
ia
es
core.c
om/j
ourn
als/i
ndex.
ph
p/ij
eecs
Flexibl
e Confi
guration
of Wir
eles
s Sens
or
Net
work
for
Monitori
ng of Ra
infall
-
Indu
ce
d L
andslid
e
Gian Qu
oc,
A
nh
1
, Duc
-
Tan
,
Tran
2
,
N
guye
n, Dinh
Ch
in
h
3
,
Ti
en Bui,
Di
eu
4
1
,2,3
Facul
t
y
of El
ec
tron
ic
s
and
T
e
le
comm
unic
atio
n,
VN
U
,
Hanoi
-
Univer
sit
y
of E
ngine
er
ing
an
d
Te
chno
log
y
,
144
Xuan
Thu
y
,
C
au
Gi
a
y
,
Hanoi, Vietna
m
1
Depa
rtment of
El
e
ct
roni
cs,
Na
m
Dinh
Univer
si
t
y
of Te
chnol
og
y
Educat
ion
,
Ph
u
Nghia, L
o
c
Ha
,
Nam
Dinh,
Vie
tna
m
4
Geogra
phic Inf
orm
at
ion
S
y
s
te
m
group, Depa
rtm
ent
of
Business a
nd
IT, Unive
rsi
t
y
Co
l
l
ege
o
f
Sou
the
ast
Norw
a
y
,
Gullbri
ngveg
en 36,
Bø
i Te
l
emar
k
N
-
3800,
Norw
a
y
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
M
ay
2
1,
2018
Re
vised Jul
1
2
,
2018
Accepte
d
O
ct
1
2
, 201
8
Esta
b
li
shm
ent
of
wire
le
ss
sensor
net
work
for
Mo
nit
oring
and
E
ar
l
y
W
arn
ing
S
y
stem
(MEW
S)
of
ra
infa
l
l
-
indu
ce
d
la
ndslid
e
is
s
ti
ll
a
challe
ng
ing
ta
sk
du
e
to
proble
m
s
of
en
erg
y
consum
pti
on.
Thi
s
rese
arch
a
im
is
to
pro
pose
a
n
ew
fle
xible
s
y
s
te
m
f
or
the
conf
igur
ation
of
MEW
S
for
rai
nf
al
l
-
induc
e
d
la
ndslid
es.
The
proposed
s
y
stem
is
an
integr
at
ion
a
st
ar
topolog
y
and
a
tree
to
polog
y
,
i
n
which,
th
e s
ta
r
t
opolog
y
is
aut
o
m
at
ic
a
lly
used
f
or the s
ensor no
des
aro
und
the
gat
ewa
y
node
,
where
as
the
tr
e
e
topo
log
y
is
a
dopte
d
for
the
othe
r
nodes.
Consequent
l
y
,
t
he
proposed
con
figura
t
ion
m
et
ho
d
is
fle
x
ible
and
c
apa
bl
e
to
save
the
en
erg
y
c
onsum
pti
on
in
th
e
MEW
S.
Expe
r
i
m
ent
resul
t
show
ed
tha
t
th
e
per
form
anc
e
of
the
MEW
S
with
the
propose
d
m
et
hod
W
ire
le
ss
Sensor
Network
is
be
tt
e
r
tha
n
thos
e
from
the
s
y
st
ems
onl
y
use
th
e
tr
ee
topo
log
y
or
th
e
star
topo
log
y
.
T
he
r
esult
o
f
th
is
s
tud
y
is
useful
for
designi
ng
W
SN
for
MEW
S
of
rai
nf
al
l
-
induced
l
andsli
des
.
Ke
yw
or
d
s
:
Lan
ds
li
de
s
W
i
reless
Senso
r
N
et
w
ork
Early
W
a
rn
i
ng
To
po
l
og
y
Viet
nam
Copyright
©
201
8
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Gian Q
uo
c
, An
h
,
Faculty
of Ele
c
tro
nics and
Tel
ecom
m
un
ic
at
i
on,
VNU
,
H
a
noi
-
Un
i
ver
sit
y o
f En
gin
eer
i
ng a
nd Tec
hnol
og
y,
144 X
uan T
huy, Ca
u Giay
, H
ano
i,
V
ie
tnam
,
Em
a
il
:
gianq
uo
canh@
gm
ail.co
m
1.
INTROD
U
CTION
Mon
it
ori
ng
a
nd
Ea
rly
W
ar
ning
Syst
em
(MEWS
)
f
or
rain
fal
l
-
induce
d
la
nds
li
de
is
an
e
ff
ic
i
ent
to
ol
f
or
la
nd
sli
de m
anag
em
ent
and
it
s
risk re
duct
io
n.
Ba
sic
al
ly
, a
M
E
W
S
for l
an
dsl
ide
consi
sts
of
sever
al
c
om
po
nen
t
s
su
c
h
as
(1)
se
ns
or
net
wor
ks
for
m
on
it
ori
ng
pa
ram
et
ers
of
the
la
ndsli
de;
(
2)
te
le
com
m
un
ic
at
io
n
syst
em
;
(3
)
m
on
it
or
ing
a
nd
analy
sis
sta
ti
on
;
(
4)
decisi
on
m
aking
;
a
nd
(
5)
respo
ns
e
procedu
res.
Howe
ver,
desi
gn
a
nd
i
m
ple
m
entat
io
n
of
la
ndsli
de
ME
W
S
a
re
not
easy
ta
sk
s
bec
ause
they
a
re
de
pende
nt
on
va
rio
us
fact
or
s
.
On
e
of
these
facto
rs
i
s
the
te
le
com
m
un
ic
at
ion
sy
stem
that
us
es
to
sen
d
tim
ely
the
m
on
it
or
i
ng
par
am
et
ers
of
t
he
la
nd
sli
de
to
t
he
stat
ion
for fu
rt
her analy
zi
ng
a
nd m
aking
dec
isi
on
.
An
ef
fecti
ve
te
le
com
m
un
ic
at
i
on
syst
em
cou
ld
be
de
rive
d
with
the
use
of
a
r
el
at
ive
ne
w
te
c
hnology
of
W
i
reless
Se
ns
or
Netw
ork (
W
SN
)
that
h
a
s
prov
e
n via
ble,
re
li
abili
ty
, f
le
xibi
li
t
y and easy
f
or exte
ndin
g
in
so
m
e
recent
works
[1
-
4]
.
WSN
is
a
lso
su
it
able
for
ME
W
S
est
abl
ished
i
n
ha
rsh
env
i
ronm
ents
because
sen
sor
nodes
us
e
it
s
inte
rn
a
l
batte
ry.
T
he
m
a
in
disa
dv
a
ntage
of
WSN
is
t
hat
it
use
s
lo
w
band
width
an
d
sho
rt
ra
nge
com
m
un
ic
at
ion
.
M
or
e
ov
e
r,
WSN
exists
li
m
it
a
ti
on
s
of
processin
g
ca
pabi
li
t
y
a
nd
sto
ra
ge
of
i
nd
i
vidual
sens
or
nodes
[
5]
.
Howev
e
r,
if
WSN
is
us
e
d
f
or
a
long
-
te
rm
MEW
S
of
la
ndsli
de
,
dissi
pation
of
e
nergy
in
t
he
sens
or
nodes
,
m
ic
ro
con
t
ro
ll
er
unit
s,
an
d
tra
ns
cei
ve
rs
of
WSN
is
a
c
riti
cal
issu
e.
T
hus
optim
i
zat
ion
of
t
he
e
nerg
y
consum
ption
f
or
WSN
is
sti
ll
n
ee
ded b
et
te
r
so
luti
ons.
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Fle
xi
ble Co
nfigu
r
atio
n of Wi
rel
ess S
e
nsor
Ne
tw
or
k for
M
on
it
ori
ng
…
(
G
ian
Q
uo
c
, A
n
H
)
1031
Lit
eratur
e
rev
i
ew
sho
ws
that
var
i
ou
s
s
ol
ution
s
f
or
reduci
ng
e
ne
rg
y
c
onsu
m
ption
an
d
exten
ding
the
WSN
li
feti
m
e
hav
e
bee
n
pro
po
s
ed
i.e
.
data
com
pr
essio
n
[
6,
7]
,
lo
w
e
nergy
arc
hitec
ture
sens
or
no
des
[8,
9]
,
routin
g
p
r
oto
c
ols
[
10]
,
la
ye
r
op
ti
m
iz
ation
a
nd
sam
pling
r
a
te
changi
ng
[
11]
.
I
n
ge
ner
al
,
these
prov
i
de
bette
r
dynam
ic
trade
off
s
olu
ti
ons
a
m
on
g
perform
ance;
c
omm
un
ic
at
e
range,
a
nd
netw
ork'
s
li
f
et
i
m
e.
Nev
e
rth
el
ess,
op
ti
m
iz
ation
of
the
e
nergy
co
ns
um
ption
i
n
s
ens
or
node
s
of
WSN
is
sti
ll
crit
ic
al
becau
se
the
dis
pe
ns
ed
e
nergy
is
pro
portio
nal
to
t
he
s
quare
of
the
distance
from
the
se
nsor
no
des
t
o
the
m
on
it
or
in
g
a
nd
a
naly
sis
sta
ti
on.
F
or
exam
ple,
a
batte
ry
6600m
A
only
us
es f
or 7
.
6
days for
t
heir
sens
or
nod
e
s
in
act
ive
m
od
e
[
11]
,
an
d
t
his
is
cl
ear
no
t
suffici
ent
f
or lo
ng tim
e M
E
W
S
.
This
pap
e
r
a
ddresses
t
he
af
or
e
m
entione
d
cri
ti
cal
issue
by
pro
po
si
ng
a
ne
w
fle
xib
le
m
eth
od
for
th
e
config
ur
at
io
n
of
ME
WS
f
or
rainf
al
l
-
i
nduce
d
la
ndsli
des
ai
m
ing
to
e
xten
d
the
li
fe
ti
m
e
of
th
e
ME
WS
s
ys
tem
.
The
pro
po
se
d
config
ur
at
io
n
i
s
an
i
nteg
rati
on
of
t
he
sta
r
t
opol
og
y
a
nd
the
tree
to
po
l
og
y
that
ena
ble
fle
xib
le
scenari
os
.
Acc
ordin
gly,
w
he
n
a
la
nd
sli
de
is
m
on
it
or
ed
a
nd
analy
zed
it
s
in
sta
bili
ty
by
m
e
an
of
Fact
or
of
Safety
(F
oS)
,
t
he
tree
to
polo
gy
is
use
d
if
FoS
la
r
ge
r
t
han
1,
cal
le
d
no
rm
al
conditi
on
.
F
or
t
his
c
ase,
s
ens
or
no
des
of
WSN
near
t
he
gateway
s
witc
h
to
r
ou
te
r
m
od
e
to
save
t
he
energy
co
nsum
ption.
On
c
ontr
ast
,
the
sta
r
to
polo
gy
will
be
a
uto
m
atical
ly
us
ed,
a
nd
in
this
case,
se
ns
or
no
des
fa
r
f
ro
m
th
e
gate
wa
y
in
the
sta
r
to
polo
gy
will
co
nsum
e
m
uch
energy,
bu
t
the
net
wor
k
is
reli
able.
B
ecause,
e
ve
ry
sens
or
node
will
send
a
nd
rece
ive
data
directl
y
to
the
gateway, t
her
e
fore,
t
he op
e
rat
ion
of a se
nsor
node d
oes n
ot
dep
e
nd
on o
t
he
r nodes.
2.
BACKG
ROU
ND OF
WI
RELE
SS
SEN
SO
R NET
WO
R
K AN
D
ITS
TOPOLO
GY
This
sect
i
on
de
scribes
bri
efly
on
W
i
reless
Se
ns
or
Net
work
(
WSN),
T
ree
T
opology
a
nd
Star
to
polo
gy
us
es i
n
ME
WS f
or
rain
fall
-
in
du
ce
d
la
ndsli
de
.
2.1
.
Wir
el
ess
Sens
or Netw
or
k
A
WSN
c
ould
be
de
fine
d
as
a
set
of
se
nse
d
de
vices
(also
cal
le
d
se
nsor
nodes
)
that
is
sp
at
ia
ll
y
distrib
uted
in
t
he
rain
fall
-
in
duced
la
nd
sli
de
bei
ng
c
onside
red,
i
n
w
hich,
sens
or
s
in
th
es
e
no
des
co
uld
sense
par
am
et
ers
of
t
he
la
nd
sli
de
(i.
e.
pore
-
water
in
sl
op
e
s
oil,
ti
lt
an
d
vibrat
ion)
a
nd
wireless
ly
com
m
un
ic
ate
th
e
ob
ta
ine
d
par
a
m
et
er
value
s
vi
a
wireless
li
nks.
T
he
data
va
lues
a
re
the
n
trans
ferred
to
a
sin
k
to
co
nn
ect
to
th
e
internet
or o
t
he
r netw
orks
t
hro
ugh
a
g
at
e
way
[12]
.
WSN
has
bee
n
su
ccess
f
ully
app
li
ed
f
or
vari
ou
s
real
-
w
orl
d
pro
blem
s
su
ch
as
f
ood
a
nd
agr
ic
ult
ure
m
on
it
or
ing
[
13]
,
flas
h
-
flo
od
al
erti
ng
[14]
,
so
il
m
oistur
e
m
on
it
or
i
ng
[
15]
,
ci
vil
struct
ur
es
m
on
it
ori
ng
[
16]
,
unde
rgrou
nd
st
ru
ct
ur
e
m
on
it
ori
ng
[
17
]
,
a
nd
unde
rgrou
nd
c
oal
m
ine
m
on
it
or
i
ng
[18]
.
H
oweve
r,
few
res
earche
s
on
ap
plica
ti
on
of
WS
N
for
la
nd
sli
de
stu
dy
hav
e
bee
n
c
arr
ie
d
out
i.e.
la
nd
sli
de
det
ect
ion
[
19]
,
la
nd
sli
de
pr
e
dicti
on
[20]
,
a
nd
la
nd
sli
de
early
wa
rn
i
ng
s
yst
e
m
[2
1,
22]
.
T
her
e
fore,
re
s
earch
on
WSN
for
la
ndsli
de
M
E
WS
sh
oul
d be stil
l carried
out.
2.2.
T
opolo
gy
of Wire
le
ss
Senso
r
Ne
two
rk
Desig
natio
n
of
an
ef
fecti
ve
la
nd
sli
de
ME
WS
requires
deter
m
ining
an
ap
propriat
e
topolo
gy
fo
r
WSN.
This
is
a
crit
ic
al
i
m
po
rtance
to
ensure
both
t
he
reli
abili
ty
and
the
en
er
gy
co
ns
er
vatio
n
of
WSN
[
23
]
.
Lit
eratur
e
rev
ie
w
s
hows
t
hat
var
i
ou
s
ty
pe
s
of
to
po
l
og
y
for
WSN
ha
ve
bee
n
us
e
d
i.e.
un
dergro
und
topolo
gy
a
nd
hy
br
id
topolo
gy
[
23
]
,
li
near
to
po
l
og
y
[24]
,
m
esh
network
t
opol
og
y
and
pee
r
-
to
pee
r
-
to
polo
gy
[
12]
,
sta
r
topolo
gy
[25]
,
and
tree
to
polo
gy
[
16]
.
Dif
fere
nt
to
polo
gies
wer
e
propose
d
beca
us
e
ob
st
r
uctio
ns
i
n
t
he
env
i
ronm
ent
of
real
-
world
prob
le
m
s
m
ay
lim
it
or
pr
e
ve
nt
c
ommun
ic
at
io
n
bet
w
een
so
m
e
node
s.
In
this
sect
i
on,
t
he
sta
r
to
polo
gy
and the t
ree to
po
l
og
y
a
s s
hown in
Fig
ure
1
are s
hortly
d
isc
us
se
d.
E
R
E
E
R
C
E
E
C
E
n
d
D
e
v
i
c
e
S
i
n
k
n
o
d
e
W
i
r
e
l
e
s
s
C
o
n
n
e
c
t
i
o
n
R
R
o
u
t
e
r
a
)
S
t
a
r
T
o
p
o
l
o
g
y
b
)
T
r
e
e
T
o
p
o
l
o
g
y
E
E
E
E
E
C
E
Figure
1. Net
w
ork
t
opology:
(a
)
Star
t
opol
ogy and (
b) Tree
topolo
gy
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
12
, N
o.
3
,
Dece
m
ber
2
01
8
:
1030
–
1036
1032
Fo
r
the
sta
r
to
polo
gy,
t
he
c
oor
din
at
or
(sink
node
)
in
Fi
g
u
re
1
(
a
)
act
s
a
s
the
netw
ork
co
ntr
oller
a
nd
th
e
oth
e
r
de
vices
are
cal
le
d
“en
d
-
de
vices”.
T
he
end
-
dev
ic
es
do
not
com
m
un
ic
at
e
directl
y
with
each
ot
he
r
but
op
e
rate
in
dep
e
nd
e
ntly
.
Mor
e
im
po
rtantl
y,
they
are
not
af
fected
by
oth
e
r
e
nd
-
de
vices
w
he
n
these
de
vices
do
no
t
op
e
rate.
Wh
e
n
the
sta
r
to
po
l
og
y
is
act
ive,
t
he
en
d
-
de
vices
com
m
un
ic
at
e
directl
y
with
the
co
ordi
nato
r
(sink
node
).
In
a
ddit
ion
,
a
n
end
-
dev
ic
e
m
a
y
switc
h
to
the
idle
m
od
e
or
th
e
sle
ep
m
od
e
in
or
der
to
re
duce
the
po
w
e
r
consum
ption
a
nd
e
xten
d
the
durati
on
of
the
operati
on
ti
m
e
o
f
the
se
ns
or
no
de.
Be
ca
us
e
th
e
end
-
de
vices
do
not
com
m
un
ic
at
e
t
hro
ugh
t
he
routers
to
t
he
c
oor
din
at
or,
the
m
or
e
distanc
e
bet
ween
the
en
d
-
de
vice
a
nd
t
he
coor
din
at
or,
th
e m
or
e
ene
rg
y
sp
e
nd
i
ng
is.
T
he
sta
r t
opology
sh
ould
be u
se
d f
or
la
ndsli
de
m
on
it
or
ing
when t
he
la
nd
sli
de
is act
ive stat
us
.
Re
gardin
g
to
t
he
tree
to
polo
gy
in
Fig
ure
1
(
b
)
,
the
netw
ork
consi
sts
of
one
coor
din
at
or,
se
ver
al
r
oute
rs,
and
en
d
-
de
vic
es.
T
he
c
oor
din
at
or
set
s
up
t
he
netw
ork,
se
le
ct
s
the
ope
ra
ti
ng
c
ha
nn
el
,
gi
ve
ad
dresses
f
or
the
routers
a
nd
the
en
d
de
vices.
The
routers
c
om
m
un
ic
at
e
dir
ect
ly
with
the
coor
din
at
or.
In
the
tree
to
polo
gy,
th
e
routers
t
ran
s
fe
r
the
data
f
ro
m
the
en
d
-
de
vice
s
to
the
co
ordi
nato
r,
t
her
e
for
e,
the
routers
a
re
al
ways
act
iv
e
that
consum
e
energ
y.
Howe
ver,
th
e
distance
fro
m
the
end
-
dev
i
ces
to
the
r
ou
te
rs
is
s
horter
th
a
n
th
os
e
from
the
en
d
-
dev
ic
es
to
the
coor
din
at
or
of
the
sta
r
to
polo
gy
(
Fig
u
re
1),
t
her
e
fore,
the
WSN
syst
em
usi
n
g
the
tree
to
po
l
og
y
consum
es
le
ss
energy
tha
n
t
ha
t
of
the
sta
r
to
polo
gy.
T
hus,
t
he
tree
to
polo
gy
sho
uld
be
us
e
d
wh
e
n
t
he
la
ndsli
de
is q
uite st
able s
ta
tus.
3.
E
X
PERI
MEN
TS A
ND R
E
S
ULT
3.1
.
Desig
nat
ion
th
e
Wir
el
ess Sens
or Ne
t
w
ork
S
ystem
f
or Ra
in
f
all
-
induced L
an
d
sli
de
In
t
his
resea
rc
h,
the
W
irel
ess
Sensor
Netw
ork
(
WSN)
syst
e
m
was
desi
gned
with
6
se
nsor
nodes
a
s
sh
ow
n
in
Fig
ure
2
place
d
in
po
sit
io
ns
on
a
slop
e
s
urface
e
sta
blished
i
n
t
he
sens
or
la
b
of
V
NU
Un
i
versi
ty
of
En
gin
eeri
ng
a
nd
Tech
nolo
gy
(
Viet
nam
).
T
he
sink
node
wa
s
placed
near
the
slo
pe
at
a
sa
fe
distance.
A
we
at
he
r
sta
ti
on
,
w
hich
connects
direct
ly
to
the
sin
k
node,
prov
i
des
m
on
it
or
in
g
da
ta
of
rai
nf
al
l
a
nd
it
s
inte
ns
it
y
an
d
durati
on.
T
he
s
ens
or
nodes
ne
ar
the
si
nk
node
can
s
witc
h
be
tween
t
he
tw
o
functi
ons,
t
he
rout
er
a
nd
the
end
-
dev
ic
e.
Be
side
s
the
batte
ry
powe
r,
t
he
se
nsor
nodes
opera
ti
ng
as
the
r
ou
t
ers
are
al
s
o
power
e
d
by
so
la
r
pan
el
s
.
The
se
nsor
no
de
s that are
f
a
r f
ro
m
sink
node
are set
up as t
he
end
-
de
vices.
S
i
n
k
n
o
d
e
a
n
d
G
a
t
e
w
a
y
R
o
u
t
e
r
S
N
C
r
i
t
i
c
a
l
s
l
i
p
s
u
r
f
a
c
e
W
e
b
S
e
r
v
e
r
C
o
m
p
u
t
e
r
S
t
a
t
i
o
n
R
o
u
t
e
r
S
N
S
N
S
N
S
N
:
S
e
n
s
o
r
n
o
d
e
S
l
o
p
e
s
u
r
f
a
c
e
R
a
i
n
g
a
u
g
e
Figure
2. The
netw
ork of
se
nsor
no
des use
d i
n
this
researc
h
It
is
note
d
that
these
e
nd
-
dev
i
ces
can
c
omm
un
ic
at
e
to
the
coor
din
at
or
(th
e
sta
r
to
polo
gy)
or
t
hroug
h
the rou
te
r
s (
the
tree topolo
gy)
. D
at
a is trans
f
err
e
d
to the si
nk no
de
f
or
furt
her
processi
ng.
A
g
at
e
way co
nn
ect
s
the
W
S
N
t
o t
he
internet
to
ex
pand t
he
abili
ty
of
c
omm
un
icati
on
.
A c
om
pu
te
r st
at
io
n use
s
the
fiel
d s
urv
ey
data
and real
-
ti
m
e d
at
a acqu
i
red f
r
om
sen
sor
node
s for analy
zi
ng the
stat
us
of t
he
sl
op
e
.
3.2
.
Desig
nat
ion
of t
he
Sen
so
r
Nodes
for
Ra
in
fa
ll
-
in
du
ced La
ndsli
de
Bl
ock
diag
ra
m
of
a
sens
or
no
de
is
s
ho
wn
in
Fi
gure
3a,
inclu
ding
sens
ors,
Was
pm
ot
bo
a
rd,
rech
a
rg
ea
ble
ba
tt
ery,
XBee
-
Pr
o
ZigBee
R
F
m
od
ule.
Se
ns
ors
s
houl
d
be
sel
ect
ed
ba
sed
on
the
de
gr
ee
of
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Fle
xi
ble Co
nfigu
r
atio
n of Wi
rel
ess S
e
nsor
Ne
tw
or
k for
M
on
it
ori
ng
…
(
G
ian
Q
uo
c
, A
n
H
)
1033
accuracy,
the
l
ifet
i
m
e
of
op
e
r
at
ion
,
a
nd
pow
er
c
on
s
um
ption
[23]
.
T
he
re
f
or
e
,
in
this
res
earch
,
th
ree
se
ns
ors,
a
so
il
m
oistur
e
s
ens
or
,
a
te
m
per
at
ur
e
s
ens
or,
and
a
n
acce
le
r
at
ion
se
ns
or
w
ere
us
e
d.
Acc
ordin
gly,
W
as
pm
ote
bo
a
r
d,
wh
ic
h
is
powe
red
by
a
batte
ry
an
d
a
s
olar
pan
el
,
is
use
d
f
or
the
se
nsor
nodes
to
reduce
e
nergy
co
nsum
i
ng
Be
cause
t
he
op
erati
on
ti
m
e
of
the
Wasp
m
ote
boa
rd
us
in
g
ba
tt
ery
only
can
be
1
t
o
5
ye
ar
s
dep
e
ndin
g
on
rate
of
c
omm
un
ic
at
ion
us
e
d,
the
refor
e
,
i
n
this
stu
dy,
a
Wasp
m
ote
PRO
v1.2
is
em
plo
ye
d
a
nd
desi
gn
e
d
with
an
A
Tm
ega1
281
m
ic
ro
co
ntr
oller
(
fr
e
quenc
y
14
M
Hz
,
SR
AM
8KB,
Flash
12
8K
B
),
w
he
reas
XBee
is
us
e
d
f
or
wirelessl
y
c
omm
un
ic
at
ion
.
W
e
use
a
rec
harg
eable
batte
ry
(
6600m
Ah
a
nd
3.7V)
f
or
the
s
ens
or
no
des.
F
or
th
ose
act
s as the r
ou
t
ers,
t
he
e
xtra
s
olar pa
nel is e
m
plo
ye
d.
Figure
3.
(a
)
B
lock diag
ram
o
f
a se
nsor
no
de
; (b) a
phot
o of a real se
nsor
nod
e
Re
gardin
g
the
ZigBee
net
wor
k
,
we
sel
ect
the
ZigBee
f
or
la
ndsli
de
m
on
it
or
i
ng
i
n
this
st
ud
y
du
e
to
lo
w
powe
r
co
nsum
ption.
T
his
is
an
RF
protoc
ol
,
in
w
hich
,
tw
o
lo
wer
la
ye
r
s
MAC
(Me
dia
Access
C
on
t
r
ol)
a
nd
Ph
ysi
cal
la
ye
r
def
i
ned
by
IEE
E
80
2.15.4
we
r
e
us
e
d.
ZigBee
is
ta
rg
et
e
d
for
us
in
g
i
n
WSN
with
lo
w
powe
r,
l
ow
band
width,
a
nd
s
hort
com
m
un
ic
at
io
n
ra
nge.
T
he
ZigBee
net
work
incl
udes
a
c
oor
din
a
tor,
r
oute
rs
an
d
e
nd
dev
ic
es
.
T
he
c
oor
din
at
or
c
ho
os
es
P
A
N
I
D
and
ope
rati
ng
channel,
ge
nerat
es
ad
dr
es
ses
to
routers
a
nd
e
nd
dev
ic
es
.
In
t
his
w
ork
,
the
c
oord
i
nato
r
act
s
as
the
si
nk
node,
w
her
e
the
r
ou
te
rs
j
oi
n
the
netw
ork
com
po
ne
nts,
transm
it
and
r
ecei
ve
t
he
m
on
it
or
i
ng
data
t
o
t
he
c
oor
din
a
tor
an
d
tran
sf
er
data
from
oth
e
r
de
vices
to
th
e
coor
din
at
or.
T
he
e
nd
-
dev
ic
es
are
se
nsor
no
de
s that m
on
it
or the
par
am
et
ers
an
d
tra
nsfer t
o t
he
c
oor
din
at
or.
3.3
.
W
orkin
g Pri
ncipl
e of
t
he WS
N Sys
te
m
The
sens
or
node
s
se
ns
e
t
he sl
ope
par
am
et
ers
and
tra
ns
fe
r t
he
obta
ine
d
data
to
t
he
si
nk
node
, t
he
n,
the
gateway
uploa
ds
t
his
data
to
the
web
databa
se
via
the
inter
net.
T
he
data
is
analy
ze
d
at
t
he
sta
ti
on
us
i
ng
t
he
GeoStu
dio
s
of
t
war
e
to
ass
ess
the
sta
tu
s
of
t
he
slo
pe,
sta
ble
or
un
sta
ble,
in
te
rm
of
Fact
or
of
Sa
fety
(
FoS)
as
sh
ow
n
in
Fig
ure
4
.
It
i
s
note
d
t
hat
base
d
on
t
he
F
oS
val
ue
s,
t
he
syst
em
cou
l
d
be
s
witc
hed
betwee
n
the
t
re
e
topolo
gy and t
he
sta
r
to
po
l
ogy.
In
this
stu
dy, w
e d
ivi
de
t
he
op
e
rati
on o
f
th
e
syst
em
into
t
wo
sce
nar
i
os
: 1)
no
rm
al
op
er
at
ion
m
od
e
;
2)
warnin
g
operat
ion
m
od
e.
The
first
sce
na
rio
is
a
safe
c
onditi
on
wh
e
n
the
w
eat
her
is
good,
po
re
water
pr
e
ssur
e
is
low
an
d
the
slop
e
vibrat
es
belo
w
a
thr
esh
old
.
The
sec
on
d
scen
ario
is
a
n
unsa
fe
co
ndit
ion
wh
e
n
the
wea
the
r
is ba
d,
pore
wa
te
r
press
ur
e
is
high a
nd the
slop
e
v
i
br
at
es
ov
er a
t
hr
es
hold.
In
th
e
norm
al
op
e
rati
on
m
od
e,
FoS
is
m
uc
h
great
er
tha
n
on
e
,
the
netw
ork
us
es
the
t
re
e
topolo
gy.
Wh
e
reas
i
n
the
warnin
g
opera
ti
on
m
od
e,
sl
ope
sta
bili
ty
is
low
or
vi
br
at
es
ov
e
r
a
t
hr
es
hold,
t
he
pri
ori
ty
of
t
he
syst
e
m
is
data
colle
ct
ion
;
th
eref
or
e
,
the
hig
h
sam
ple
rate
and
t
ran
s
fe
r
r
at
e
are
us
e
d
t
o
m
on
it
or
the
slop
e'
s
conditi
on.
I
f
F
oS
inc
reases,
t
he
slo
pe
com
es
back
t
he
safe
sta
te
,
the
netw
ork
s
witc
h
f
rom
the
sta
r
topo
log
y
to
th
e
tree
to
polo
gy.
Acc
ordin
gl
y,
the
se
ns
or
node
s
t
urn
i
nto
the
idle
m
od
e
to
sa
ve
ene
rg
y
.
T
he
pow
er
f
or
t
he
sens
or
s
is tu
rned
off,
a
nd the
Was
pm
ote b
oa
rd an
d
t
he
RF
m
od
ule also s
witc
h
to
the
idl
e m
od
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
12
, N
o.
3
,
Dece
m
ber
2
01
8
:
1030
–
1036
1034
S
t
a
r
t
S
l
o
p
e
s
t
a
b
i
l
i
t
y
a
n
a
l
y
s
i
s
G
e
t
d
a
t
a
f
r
o
m
s
e
n
s
o
r
n
o
d
e
s
a
n
d
r
a
i
n
g
a
u
g
e
S
t
a
r
T
o
p
o
l
o
g
y
C
o
n
f
i
g
u
r
a
t
i
o
n
E
n
d
F
o
S
g
r
e
a
t
e
r
t
h
a
n
o
n
e
T
r
e
e
T
o
p
o
l
o
g
y
C
o
n
f
i
g
u
r
a
t
i
o
n
T
r
u
e
F
a
l
s
e
S
e
t
S
a
m
p
l
e
R
a
t
e
F
S
2
S
e
t
S
a
m
p
l
e
R
a
t
e
F
S
1
V
i
b
r
a
t
i
o
n
>
t
h
r
e
s
h
o
l
d
T
r
u
e
F
a
l
s
e
Figure
4.
Wo
r
ki
ng
pri
nciple
of the
prop
os
e
d sy
stem
3.4
.
Results
and
Discussio
n
The
res
ult
of
t
he
est
i
m
at
ion
of
the
powe
r
c
on
s
um
ption
is
sh
ow
n
in
Ta
ble
1.
It
co
uld
be
see
n
t
hat
t
he
powe
r
a
se
ns
or
node
co
nsum
e
d
i
n
t
he
act
ive
m
od
e
is
10
0.255m
W
.
F
or
the
batte
ry
wit
h
th
e
capaci
ty
of
6600m
A
h
and
volt
age
of
3.7V,
the
m
axim
u
m
po
we
r
s
upply
f
or
a
se
nsor
node
is
24
420m
W
h.
T
her
e
f
or
e
,
the
w
orki
ng
tim
e
of the se
nsor
nod
e
m
ay
ex
te
nd
rou
gh
ly
to
24
3.5 hou
rs.
In
fact,
the
ef
fi
ci
ency
of
th
e
ba
tt
ery
is
m
uch
sm
al
le
r.
Fo
r
e
xa
m
ple,
if
the
eff
ic
ie
ncy
is
75
%
,
the
w
orkin
g
tim
e
decr
eases
to
182.6
ho
urs.
T
his
w
orki
ng
ti
m
e
is
not
enou
gh
for
th
e
lo
ng
ti
m
e
m
on
it
ori
ng
syst
em
of
la
nd
sli
de
.
In
or
der
to
exte
nd
the
li
feti
m
e
of
t
he
netw
ork,
th
e
sens
or
nodes
sh
oul
d
on
ly
be
act
ive
in
li
m
it
e
d
ti
m
e
and tu
rn
e
d
t
o
t
he
idle m
od
e
.
Table
1
.
T
he
p
ow
e
r
c
onsu
m
ed by a se
nsor
nod
e
Mod
u
le
Po
wer
(
m
W)
Idle
m
o
d
e
Activ
e
m
o
d
e
W
asp
m
o
te
0
.18
4
9
.5
XBee
-
PRO
Z
ig
Be
e
1
0
m
W
(
+1
0
d
B
m
)
f
o
r
Internatio
n
al variant
Acceler
o
m
ete
r
1
.15
5
So
il
m
o
istu
re
sen
s
o
r
6
.6
Te
m
p
e
rature
sen
so
r
(
LM
3
5
)
33
In
t
he
norm
al
op
e
rati
on
m
ode,
the
e
nd
-
dev
i
ces
only
act
ive
3
sec
onds
afte
r
sta
yi
ng
10
m
i
nu
te
s
i
n
the
idle m
od
e.
T
he
po
wer t
he
se
nsor
node
us
es
is ro
ughly
0.6
56m
Ah
. T
he
refor
e
, t
he
li
fetim
e of
the
se
nsor
node
is
1163
days
c
onsiderin
g
t
he
e
f
fici
ency
is
75%.
I
n
wa
rn
i
ng
operati
on
m
od
e,
t
he
se
nsor
node
act
i
ve
3
seco
nds
after
sta
y
in
g
1
m
inu
te
in
the
idle
m
od
e.
T
he
li
fetim
e
of
the sen
s
or
nod
e
is 161
days
c
onsideri
ng
the
e
ff
ic
ie
ncy
is 7
5%.
The
a
bove
resul
t
is
cl
early
better
tha
n
in
t
he
previ
ou
s
w
orks
i
n
[
12]
in
te
rm
of
the
powe
r
co
ns
um
ptio
n.
In
[
12]
,
the
network
is
switc
he
d
from
the
tre
e
to
po
l
og
y
to
t
he
sta
r
t
opolog
y
w
hen
the
rai
nf
al
l
reache
s
t
o
t
he
determ
ined
th
r
esh
old
.
It
happ
ens
e
ve
n
th
ough
the
F
oS
is
sti
ll
high
(i.e.
FoS
>>1).
I
n
our
w
ork,
t
he
WSN
s
yst
e
m
is
switc
he
d
f
r
om
the
tree
t
opology
to
t
he
st
ar
t
opology
on
ly
wh
e
n
F
oS
r
eaches
to
1.
T
hu
s
,
it
can
sa
ve
m
or
e
powe
r.
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Fle
xi
ble Co
nfigu
r
atio
n of Wi
rel
ess S
e
nsor
Ne
tw
or
k for
M
on
it
ori
ng
…
(
G
ian
Q
uo
c
, A
n
H
)
1035
On
e
of
t
he
m
os
t
crit
ic
al
iss
ues
of
the
W
SN
is
the
tra
nsm
issi
on
reli
a
bili
ty
.
In
t
his
researc
h,
the
transm
issi
on
r
e
li
abili
ty
o
f
the p
r
opos
e
d WS
N
syst
e
m
is asse
ssed usin
g
Pac
ket D
el
ive
ry Rat
io (
P
DR)
m
e
asur
e
.
This
is
the
ra
ti
o
betwee
n
the
num
ber
of
m
essages
that
the
se
ns
or
nodes
tra
nsm
it
te
d
and
the
nu
m
ber
of
m
es
sage
s
the
sin
k n
ode
r
ecei
ved
.
Th
e
re
su
lt
is
show
n T
ables
2
an
d
3 a
nd
Fig
ur
e
5.
It
cou
l
d
be se
e
n t
hat
the
total
num
ber
of
pac
kets
the
sens
or
node
s
s
ent
is
73
0,
wh
e
reas
the
t
otal
num
ber
of
pa
ck
et
s
the
sin
k
node
recei
ve
d
is
730
f
or
bo
t
h
the
star t
opol
og
y a
nd the
tree to
polo
gy, t
her
ef
ore, t
he
P
DR param
et
er is 100
%
.
Figure
5.
Sce
na
rios f
or
outd
oor
experim
ent; (a) st
ar top
olog
y;
(
b) tree t
opol
og
y
Table
2
.
Packe
t Deli
ver
y R
at
i
o (PD
R
)
for
t
he
tree to
polo
gy
No
d
e
Total o
f
tr
an
s
m
itte
d
pack
ets
Total o
f
r
eceived
p
ackets
Percent of
r
eceive
d
pack
ets
1
140
140
100%
2
103
103
100%
3
127
127
100%
4
109
109
100%
5
133
133
100%
6
118
118
100%
Table
3.
Packe
t Deli
ver
y R
at
i
o (PD
R
) f
or t
he
star to
polo
gy
No
d
e
Total o
f
tr
an
s
m
itte
d
pack
ets
Total o
f
r
eceived
p
ackets
Percent of
r
eceive
d
pack
ets
1
140
140
100%
2
103
103
100%
3
127
127
100%
4
109
109
100%
5
133
133
100%
6
118
118
100%
4.
C
ONCL
US
IO
N
This
resea
rch
ha
s
propose
d
a
ne
w
f
le
xi
ble
sys
tem
fo
r
t
he
c
on
fig
ur
at
io
n
of
M
E
W
S
f
or
rain
fa
ll
-
induced
la
nd
sli
des
. T
he
prop
os
e
d
syst
e
m
is a c
om
bin
at
ion
of the
sta
r
to
polo
gy a
nd
the tree
to
po
l
ogy t
o u
se i
n dif
fer
e
nt
scenari
os
of
m
on
it
ori
ng
t
hat
i
s
not
only
to
s
ave
e
nergy
a
nd
but
al
so
to
im
pr
ov
e
operati
on
al
reli
abili
ty
of
t
he
ME
W
S syste
m
. Th
e sta
r
to
po
l
og
y i
s
us
ed w
he
n
the m
ov
em
e
nt r
at
e of the la
nd
sli
de
is hig
h (FoS ≤
1), whe
reas
the tree t
opology i
s em
plo
ye
d w
hen the
slo
pe
is stable
(FoS
> 1).
Ex
per
im
ent
resu
lt
sho
wed
th
at
the
pe
rfo
rm
ance
of
the
M
E
W
S
with
t
he
pro
posed
m
eth
od
WSN
is
bette
r
tha
n
th
ose
from
the
con
ven
ti
onal
syst
e
m
s
us
ing
only
the
tree
t
opolog
y
or
the
sta
r
to
po
l
og
y.
The
re
su
lt
of
this stu
dy is
use
fu
l
for desi
gni
ng w
i
reless se
nsor
n
et
wor
ks
f
or ME
WS of
r
ai
nf
al
l
-
in
duced
landsl
ides.
ACKN
OWLE
DGE
MENT
The
a
naly
sis
a
nd
w
rite
-
up
w
ere
ca
rr
ie
d
out
as
a
pa
rt
of
t
he
first
a
utho
r'
s
PhD
stu
dies
a
t
Faculty
of
Ele
ct
ro
nics
and Telec
omm
un
ic
at
ion
,
VNU,
Hanoi
-
Un
i
versi
ty
o
f
E
nginee
rin
g
a
nd Tech
nolo
gy
.
R
o
u
t
e
r
a
)
S
t
a
r
T
o
p
o
l
o
g
y
b
)
T
r
e
e
T
o
p
o
l
o
g
y
E
E
E
E
E
C
E
S
N
.
1
S
N
.
2
S
N
.
3
S
N
.
5
S
N
.
4
S
N
.
6
C
o
o
r
d
i
n
a
t
o
r
(
S
i
n
k
n
o
d
e
)
5
0
m
9
5
m
9
5
m
1
0
5
m
1
0
5
m
5
0
m
E
E
E
E
C
S
N
.
1
S
N
.
2
S
N
.
3
S
N
.
5
S
N
.
4
S
N
.
6
C
o
o
r
d
i
n
a
t
o
r
(
S
i
n
k
n
o
d
e
)
5
0
m
5
0
m
5
0
m
6
0
m
6
0
m
5
0
m
R
R
R
o
u
t
e
r
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
12
, N
o.
3
,
Dece
m
ber
2
01
8
:
1030
–
1036
1036
REFERE
NCE
S
[1]
Ramesh
MV
.
Real
-
ti
me
wireless
sensor
net
work
f
or
landsli
de
dete
ct
ion
.
Sensor
Technol
ogi
es
and
Applic
a
ti
on
s,
20
09
SENS
ORC
OMM'
09
Th
ird
In
te
r
nat
ion
al
Conf
erence
on
;
2009
:
IE
EE
.
[2]
Fos
al
au
C
,
Zet
C,
Pe
tri
sor
D
.
I
mpleme
ntation
of
a
landsl
ide
monitoring
syst
em
as
a
wir
el
e
ss
sensor
net
wo
rk
.
Ubiquit
ous Com
puti
ng,
Elec
tronics &
Mobi
le Co
m
m
unic
at
ion
Co
nfe
ren
c
e
(UEM
CON
),
IEEE
An
nual
;
2016:
I
EEE.
[3]
Chang
DTT
,
Ts
ai
Y
-
S,
Yang
K
-
C.
Stud
y
of
R
eal
-
Ti
m
e
Slope
St
abi
lit
y
Monitor
i
ng
S
y
stem
Us
ing
W
ire
le
ss
Sensor
Network
(W
SN
)
.
Indone
sian
Jou
rnal
of El
e
ct
ri
ca
l
Eng
ine
ering
an
d
Computer
Sc
ience
.
2013;11(3)
:
1478
-
1488.
[4]
Gian
QA
,
N
gu
y
en
DC,
Tr
an
DN
,
Tra
n
D
-
T
.
Monitori
ng
of
La
ndslid
es
in
Mountai
nous
Re
gions
base
d
on
FEM
Modell
ing
and
Rai
n
Gaug
e
Me
asure
m
ent
s.
Int
e
rnational
Journ
al
of
El
e
ct
rica
l
and
Computer
E
ngine
ering
(
IJECE)
.
2016;6(5):
2106
-
2113.
[5]
Rawa
t
P,
Singh
KD
,
Chaouc
hi
H,
B
onnin
JM
.
W
ire
le
ss
sensor
net
works
:
a
su
rve
y
on
re
ce
nt
deve
lopments
a
nd
pote
ntial
s
y
n
erg
i
es.
The
Journal of
supercomputi
ng
.
2014;68(1)
:1
-
48.
[6]
Ngu
y
en
D
-
C,
D
uc
-
Ta
n
T
,
Tra
n
D
-
N.
Applicatio
n
of
compress
ed
sensing
in
ef
f
ec
t
iv
e
power
consumption
of
WSN
f
or
la
ndslide sce
nar
io
.
Mul
ti
m
edi
a
a
nd
Broadcasti
ng
(AP
Media
Cast),
2015
As
ia
Pa
ci
fi
c
Confer
ence
on
;
2015:
IEEE.
[7]
You
L,
Han
Y
,
Li
S,
Su
X.
Soure
and
Tra
nsm
ission
Control
for
W
ire
le
ss
Visual
Sensor
Networks
with
Com
pre
ssi
v
e
Sensing
and
E
ner
g
y
Har
vesting.
Indon
esian
Journal
of
El
e
ct
rica
l
Enginee
ring
and
Computer
Sc
ie
n
c
e
.
2013;11(5):
2468
-
2474.
[8]
Raghuna
th
an
V,
Schurge
rs
C
,
P
ark
S,
Srivasta
v
a
MB.
Ene
rg
y
-
a
ware
wir
el
ess
m
ic
rosensor
ne
tworks.
IEEE
S
ign
al
proce
ss
ing
magaz
ine
.
2002
;19(2)
:40
-
50.
[9]
Ze
t
C
,
Fos
al
au
C,
Petri
şor
D,
Hogaş
I.
Study
of
the
pow
er
co
nsum
ti
on
of
a
l
andslide
sensor
node
.
Ubiqui
to
us
Com
puti
ng,
E
lec
troni
cs
&
Mobil
e
Com
m
unic
at
io
n
Confer
en
ce (U
EMCON
),
IEEE
Annual; 2016:
I
EE
E
.
[10]
Panta
z
is
NA
,
Ni
koli
daki
s
SA
,
Ve
rga
dos
DD
.
Ener
g
y
-
e
fficie
nt
rou
ting
pr
oto
col
s
in
wire
l
ess
sensor
n
etw
orks:
A
surve
y
.
IEE
E
Comm
unications
sur
ve
ys
&
tut
orials
.
2013;
15(2):551
-
591.
[11]
Ngu
y
en
CD,
Tr
a
n
TD,
Tra
n
ND
,
Hu
y
nh
TH,
Ngu
y
en
DT
.
Flexi
b
le
and
eff
icien
t
wir
el
ess
sensor
ne
t
works
for
d
et
e
ct
i
ng
rai
nfa
ll
-
indu
ce
d
la
ndslide
s
.
In
t
er
nati
onal Journal
of
Distribu
te
d
S
ensor Ne
tworks
.
2015;2015.
[12]
Burat
ti
C
,
Conti
A,
Dard
ari
D,
Verdone
R.
An
over
vie
w
on
wi
rel
ess
sensor
n
e
tworks
technolo
g
y
and
evol
ut
io
n.
Sensors
.
2009;9(
9):6869
-
6896.
[13]
Yu
X,
W
u
P,
Ha
n
W
,
Zh
ang
Z.
A
surve
y
on
wir
el
e
ss
sensor
net
wor
k
infra
stru
ct
ur
e
f
or
agr
i
cul
tu
re.
C
omputer
Standar
ds
&
I
nte
rface
s
.
20
13;35(1):
59
-
64.
[14]
Casti
ll
o
-
Eff
er
M
,
Quin
tela
DH
,
Moreno W
,
Jord
an R
,
W
esthoff
W
.
Wire
le
ss
sen
sor
net
works
for
fl
ash
-
fl
ood
al
erting
.
Devic
es,
Cir
cui
t
s
and
S
y
s
te
m
s,
2
004
Proce
ed
ings
of
the
Fifth
IEEE
Inte
rn
at
ion
al
Ca
rac
as
Conf
ere
n
c
e
on;
2004:
IEEE
.
[15]
Kerke
z
B,
Glase
r
SD
,
Ba
le
s
RC
,
Mea
dows
MW
.
Design
and
per
f
orm
anc
e
of
a
wir
el
ess
sensor
n
etw
ork
for
ca
t
chment‐
sca
le snow
and
s
oil
m
oisture m
eas
ure
m
ent
s.
Wat
e
r R
esource
s
Re
s
earc
h
.
2012
;48(
9).
[16]
Chint
alapudi
K,
Fu
T,
Paek
J,
Ko
tha
ri
N,
R
angwa
la
S,
Caf
fr
e
y
J,
Govindan
R,
Joh
nson
E,
Masri
S.
Monitori
ng
c
ivil
struct
ure
s wi
th a
wire
l
ess sensor
net
work.
IEEE
I
nte
rnet
Computi
ng
.
2006;10(2)
:2
6
-
34.
[17]
Li
M,
L
iu
Y.
Un
derground
structure
monitoring
wit
h
wireless
sensor
net
works
.
Proce
ed
ings
of
th
e
6th
in
te
rn
at
ion
al
conf
ere
n
ce on
In
form
at
ion
pro
ce
s
sing i
n
sensor
ne
tworks;
2007:
A
CM.
[18]
Li
M,
Li
u
Y.
U
nder
ground
coal
m
ine
m
onit
orin
g
with
wir
el
ess
sensor
net
work
s.
ACM
Tr
ansacti
ons
on
Senso
r
Net
works (
TO
SN
)
.
2009;5(2):
10.
[19]
Ramesh MV
.
De
sign,
dev
el
opm
e
nt,
and
d
epl
o
y
m
ent
of
a
wire
le
ss
sensor ne
twork
f
or
detec
t
ion
of
l
a
ndslide
s.
Ad
Hoc
Net
works
.
2014
;
13:2
-
18.
[20]
Sheth
A,
T
ej
as
wi
K,
Mehta
P
,
Pare
kh
C,
B
an
sal
R,
Mer
cha
nt
S,
Singh
T,
D
e
sai
UB,
Th
ekka
t
h
CA,
To
y
ama
K.
Sensli
de:
a
sens
or
net
work
bas
e
d
landsli
d
e
pred
ic
ti
on
system
.
Pr
oce
ed
ings
of
the
3rd
in
te
rn
at
ion
al
conf
er
ence
o
n
Embedde
d
ne
tw
orke
d
sensor s
y
s
te
m
s; 200
5:
AC
M.
[21]
Arnhardt
C,
As
c
h
K,
Azz
am
R,
Bil
l
R
,
Ferna
nd
e
z
-
Stee
g
er
T
,
Ho
m
fel
d
S,
Kallash
A,
Nieme
y
er
F,
Rit
te
r
H,
To
loc
z
y
k
i
M.
Sensor
b
ase
d
L
andsli
de
E
ar
l
y
W
arn
ing
S
y
st
e
m
-
SLEWS
.
Deve
lopment
of
a
g
e
oservic
e
infr
astr
uct
ure
as
basis
f
or
ea
rl
y
w
arn
ing
s
ystems
for
la
ndsli
des
b
y
integra
ti
o
n
of
re
al
-
t
ime
se
nsors
.
Geote
chn
ologi
en
sc
ie
n
ce
r
eport
.
2007
;10:
7
5
-
88.
[22]
Intri
er
i
E,
Gigl
i
G,
Mugnai
F,
Fa
nti
R
,
Casagli
N
.
Design
and
implementation
of
a
l
andsli
d
e
ea
r
l
y
warni
ng
s
y
s
te
m
.
Engi
ne
ering
Ge
ology
.
2012;147
-
148:124
-
136.
[23]
Ak
y
il
d
iz
IF,
St
unte
be
ck
EP.
W
ire
le
ss
under
g
round
sensor
n
et
works
:
Rese
ar
ch
ch
al
l
enge
s.
Ad
Hoc
Net
wor
ks
.
2006;4(6):
669
-
6
86.
[24]
Hu
X,
W
ang
B
,
J
i
H.
A
wire
le
s
s
sensor
ne
twork‐ba
sed
stru
ct
ur
al
health
m
onit
o
ring
s
y
s
te
m
for
highwa
y
br
idge
s
.
Computer‐Ai
ded
Civil
and
Infras
truct
ure
Engi
ne
e
ring
.
2013;28(3)
:193
-
209.
[25]
Srbinovska
M,
Gavrovski
C,
Di
m
ce
v
V,
Krkol
e
va
A,
Boro
za
n
V.
Env
ironmental
par
amete
rs
m
onit
oring
in
pre
ci
si
on
agr
ic
u
lt
ure
using
wire
l
ess sensor
net
works
.
Journal
of
Cle
an
er
Pr
oduct
ion
.
2015;8
8:297
-
307.
Evaluation Warning : The document was created with Spire.PDF for Python.