TELKOM
NIKA
, Vol.9, No.1, April 2011,
pp. 9~18
ISSN: 1693-6
930
accredited by D
G
HE (DIKTI
), Decree No: 51/Dikti/Kep/2010
¢
9
Re
cei
v
ed Se
p 29
th
, 2010; Re
vised Feb
2
nd
, 2011; Accepte
d
April 2
nd
, 2011
Wireless Sensor Network for Landslide Monitoring
in Nusa Tenggara Timur
Herry
Z. Kott
a
1
, Kalv
ein Rantelob
o*
2
, Silv
ester Tena
3
, Gregorius Klau
4
1,2,
3
F
a
cult
y
of Scienc
e an
d En
gin
eeri
ng, Un
iv
ersitas Nus
a
C
end
an
a, Penfui
, Kupang, NT
T
4
Dinas Pertam
ban
ga
n dan E
n
ergi T
k
.
I N
T
T
,
Kupa
ng, NT
T
e-mail: gr
adz3
i
nki@
ya
ho
o.co
m
1
, rantelobo
@gmai
l
.com*
2
Abs
t
rak
T
anah
lo
ngsor
di
berb
a
g
a
i
d
aera
h
tel
ah
merup
a
kan
ba
ha
ya seri
us sec
a
ra substa
ntsia
l
bag
i j
i
w
a
ma
nus
ia d
an k
e
rug
i
an
materi
al. Untuk
men
ang
gul
an
gi d
a
n
men
gura
n
g
i
kerug
i
an ters
e
but ber
ba
gai
u
paya
untuk me
lak
u
k
an mo
nitori
ng tanah
lo
ngs
or dike
mban
gka
n
. Sala
h satu
ny
a yaitu
me
ma
n
f
aatkan tek
nol
og
i
jarin
g
a
n
s
enso
r
nirk
abe
l (JS
N
)
.
Hasi
l y
a
n
g
di
dap
atkan
d
a
ri
pen
eliti
a
n
yang
di
lak
u
ka
n d
i
desa
Ika
n
foti,
Kab. Ku
pa
ng,
Propi
nsi
NT
T
(pad
a k
oord
i
n
a
t 10
0
1
6
’
21.9
L
S
da
n
123
0
40
’59.8 BT
) s
e
b
a
gai
te
mpat
uj
i c
o
b
a
,
me
mberik
an
h
a
sil
ba
hw
a pe
nera
pan
JSN
dap
at di
ter
apk
an
den
ga
n b
a
i
k
. Peng
ukura
n
sens
or g
e
tar
a
n
(accel
e
ro
meter
)
pada
per
ang
kat Mica
z
di
gu
naka
n
untuk
me
nd
eteksi g
e
t
aran aki
bat a
dany
a per
ger
a
k
a
n
tanah.
Hasi
l d
a
ri p
e
n
e
liti
an
i
n
i
me
nu
njuk
an
ba
hw
a per
ub
aha
n n
ila
i acc
e
ller
a
tror
mula
i
dari
0,2
g
(gr
a
vity)
sampa
i
0,49
g
baik p
a
d
a
Acceller
o
metert
X ma
up
un Y
me
na
ndak
an t
ana
h mul
a
i b
e
r
gerak ta
pi b
e
l
u
m
secara si
gn
ifik
an. Nil
ai
0,5 g
ke atas a
dal
ah
n
ila
i ya
ng
me
nun
juka
n p
e
ru
bah
an p
e
rg
era
k
an tan
ah sec
a
r
a
signifik
an. M
u
l
a
i p
a
d
a
1
g
p
e
rger
akan
tan
ah su
da
h
mu
n
unj
ukan
aktivit
a
s yan
g
s
ang
at kuat d
a
n
su
da
h
me
ngk
uatirka
n
.
Dari
p
e
mant
aua
n
perg
e
ra
kan ta
na
h
de
nga
n JS
N i
n
i
di
hara
p
ka
n
akan
me
ngur
a
n
g
i
kerug
i
an
akib
a
t
benca
na g
eol
ogi tan
ah l
o
n
g
s
or khusus
nya
di NT
T
.
Dihar
apka
n
pe
nel
itia
n ini
me
mberik
a
n
lan
dasa
n
ba
gi
pen
erap
an JS
N di ber
bag
ai d
aera
h
di NT
T
dan Indo
nes
ia p
ada u
m
u
m
nya.
Ka
ta
k
unc
i
: acceler
o
meter, ja
ring
an sens
or nirka
bel (JSN),
Mica
z
,
p
e
mant
aua
n tana
h lo
n
g
sor
A
b
st
r
a
ct
Lan
dsli
des i
n
ma
ny reg
i
ons
constitute seri
ous ha
z
a
rds that cause su
b
s
tantial l
i
fe an
d financ
i
a
l
losses. T
o
ove
r
come an
d re
d
u
ce the
da
ma
ges, effort
s to mo
nitor l
a
n
d
s
lides
are d
e
ve
l
ope
d. One su
ch
techno
lo
gy utili
z
e
s a w
i
rel
e
ss
sensor netw
o
rk (W
SN
). Re
sults obtai
ned
from
studies
cond
ucted in t
h
e
Ikanfoti vi
lla
ge,
Kup
ang
Distric
t, Nusa T
e
ngg
ara T
i
mur
(NT
T
) Province
(S
10
o
16
’
2
1
.9
”
an
d E
123
o
40’
59
.8”
)
as p
ilot
pro
j
ect
,
give
res
u
lt th
at the
ap
plic
ati
on
of W
S
N c
a
n b
e
app
lie
d
p
r
operly.
W
e
d
e
t
ect and
me
as
ure
vibrati
ons ca
us
ed by
la
ndsl
i
de
s by
vibr
ation
sensor (
a
ccel
e
rometer) o
n
Mi
ca
z
dev
ices. T
he res
u
lts of th
is
study indicat
e
that changes in
accele
ro
meter
valu
es ran
g
i
ng
from 0.
2 g (gr
a
vity) to 0.49 g
of either th
e X
o
r
Y of accelero
meter ind
i
cate th
at soil be
gins t
o
mov
e
but not
signific
antly. Valu
e abov
e 0.5
g is a value th
at
indic
a
tin
g
a sig
n
ifica
n
t chan
ge
of groun
d moti
on. T
he val
ue
of 1 g and
abo
ve of grou
nd
motion i
ndic
a
tes
a
very strong acti
vity and sho
u
l
d
be alar
med. It
is expecte
d tha
t
this research
provi
des the fo
und
atio
n for the
app
licati
on of
W
S
N in vari
ou
s areas i
n
NT
T
Province
an
d
Indon
esi
a
in
g
ener
al, for est
ablis
hi
ng thor
o
ugh
and re
lia
bl
e ea
rly w
a
rning sys
tem (EW
S
).
Ke
y
w
ords
: ac
celer
o
meter, la
ndsli
de
mo
nitor
i
ng, Mica
z
,
w
i
reless se
nsor n
e
tw
ork (W
SN)
1. Introduc
tion
Within fe
w y
ears, the
ea
rth's p
opul
atio
n
ha
s in
crea
sed
very
sha
r
ply, ca
uses
human
need mo
re la
nd to live: (a) people in po
or co
untri
e
s
are be
ginnin
g
to explore mountain
s
to live
and farm, (b
) peopl
e of rich
cou
n
trie
s are m
o
vin
g
into area
s of high to get pea
ce a
n
d
comfo
r
table
p
l
ace to
live. As a
re
sult, settlements
occupy stee
p a
r
e
a
s,
which p
r
e
v
iously resulted
in movement
of land every year in all
corne
r
s
of th
e wo
rld. Lo
sses
suffere
d from moveme
n
t
o
f
land
slide
s
for one year is e
qual to the da
mage cau
s
ed
by the earthq
uake for 20 years.
The lan
d
sli
d
e
is a de
structi
v
e geologi
cal
pro
c
e
ss. Su
ch pro
c
e
s
s tha
t
at each o
ccurren
ce
is alm
o
st
always results i
n
losse
s
both i
n
mate
rial a
n
d
immate
rial.
It always ha
p
pen
s eve
r
y year
and ma
ke
s lo
sses of lives
and prope
rtie
s. So re
search is need
ed to desi
gn a system that ca
n
help to preve
n
t catast
rophi
c geol
ogi
cally haza
r
d
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
¢
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 9, No. 1, April 2011 : 9 – 18
10
In this pape
r, a land
slide d
e
tection
syste
m
was
desi
g
n
i
ng usi
ng wi
re
less se
nsor n
e
twork
(WS
N) ve
ry
steep ramp
s u
p
toward
s th
e
foot of
the
sl
ope
excee
d
in
g the b
a
lan
c
e
of po
we
r tha
t
hold its
slop
es.). La
nd
sli
de in this a
c
tivity re
fers is define
d
a
s
the mass m
o
vement of land
(incl
udin
g
rocks), layer
of
sedim
enta
r
y depo
sit t
hat
has
not con
s
olidated
or la
yer of soil on
the
slop
es that a
r
e very ste
e
p
ramp
s up to
wards th
e foot of the slo
pe ex
ce
edin
g
the balan
ce
of
power that ho
ld its slop
es.
Applicatio
n of
WS
N te
chn
o
l
ogy for lan
d
slide
mitigatio
n of g
eolo
g
ical ha
za
rd
s
ha
s g
r
o
w
n
rapidly. The
developm
ent
is ba
sed
on the ch
ar
acte
r
of WSN
which are e
a
sily
develop
ed an
d
relatively e
c
o
nomical. Th
e
use
of
conve
n
tional
se
n
s
o
r
technol
ogy
often en
co
unt
ers difficultie
s in
data
colle
ctio
n that
came
from th
e ch
ara
c
teri
st
ics
of land
slide
s
,
ca
usi
ng
so
me p
r
oble
m
s in
resea
r
ch to
predi
ct
and
d
e
tect la
nd
slid
es. T
he
pre
s
ence of
WS
N p
r
ovide
s
si
mplicitie
s in
t
he
measurement
of field d
a
ta,
and m
a
kes th
e land
slid
e p
r
edictio
n sy
ste
m
be
com
e
s
more
effectiv
e
.
The main re
aso
n
behin
d
the use of wi
rele
ss
sen
s
o
r
networks is
the flexibility
in data den
si
ty,
easy l
ong
-di
s
tan
c
e
co
m
m
unication
s
to variou
s p
o
ints, im
prov
ing ove
r
all
a
c
cura
cy, a
n
d
the
capability of real time control.
WSN i
s
a
group of smart sensors,
where
each poi
n
t of sensor has
the ca
pability
for sen
s
ing,
pro
c
e
ssi
ng a
nd commu
ni
cating, but if they are
devel
oped i
n
term
s of
numbe
rs or conne
cted to e
a
ch oth
e
r into
networks,
they will be abl
e to function
as the monito
r of
a physi
cal
sta
t
e of the eart
h
colle
ctively. WSN i
s
a ne
w gen
eration
of sen
s
o
r
sy
stems, althou
gh
still limited in data pro
c
e
s
si
ng ca
pability and ba
nd
widt
h for com
m
un
ication [1].
Variou
s
studi
es related to
grou
nd moti
on wi
th diffe
rent focu
s ha
d alre
ady be
en don
e.
Among othe
rs [2] - [7], Jamaludi
n et al [2] analyzed the temp
eratu
r
e of th
e soil an
d sl
ope
con
d
ition
s
, [3] and [4] fo
cu
sed
on th
e d
e
t
ection
of a p
r
ototype mod
e
l
, while
[5] - [
7
] tried
a
WSN
test bed for real-time me
a
s
ureme
n
t.
To the best of our kn
owl
edge, re
se
arch
on WS
N and its proto
t
ype in the
area o
f
montmorill
oni
te (expan
sive
clays) ha
s not been
wel
l
studying ye
t. This re
sea
r
ch
attempts to
contri
bute
to
the impl
emen
tation of
WS
N, not
onl
y t
he p
r
ototype
model,
but al
so to
comp
are it
with dire
ct m
easure
m
ent
s at the test site resear
ch.
This research is a co
ntin
uation of pre
v
ious
resea
r
ch [8]
of the geographi
c informa
t
ion system
s (GIS) appli
c
a
t
ion for settle
m
ent of the
hill
zone
on Nusa
Tenggara Timur (N
TT)
Province. The expe
cted
contribution of
this research will
be useful in integratin
g the
GIS system in
an early wa
rning
system
(EWS) [9] - [1
1].
2. Rese
arch
Metho
d
2.1 Acceler
ometer
Data Analy
s
is
Accel
e
romete
r re
ad
s a
bi
axial accel
e
rati
on chan
ge
s by p
r
ovidi
ng a la
rg
e
voltage
differen
c
e V
out
, so that to know h
o
w mu
ch the angle
chang
es that o
c
cur by u
s
ing
formula bel
o
w
S
V
V
offset
out
−
=
−
1
sin
α
(1)
whe
r
e:
α
: Tilt angle
out
V
: Voltage output
accelerometer
offset
V
: Offset voltage p
o
wer suppl
y
(0.4
V)
S
: Sensitivity
acceleromete
r (167
m
V
/ G, 17
%)
Accel
e
romete
r wo
rks by sendin
g
high
-l
evel
analo
g
signal
s in terms of stre
ss
(voltage
)
prop
ortio
nal t
o
the a
c
cele
ration me
asurements. S
e
n
s
ing
me
cha
n
i
s
m of th
e a
c
cele
rom
e
ter i
s
descri
bed a
s
V
out
=
(Sc
a
le_fac
tor * Acc
e
leration) +
Off
s
et_voltage
(2
)
whe
r
e:
Scale_factor
: sensor
sensitivit
y
(
V
olt/g)
Acceleration
: acceleration value of each a
x
is (g)
Offset_voltage
: voltage (zero-g
)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
¢
Wirel
e
ss Sen
s
or
Network for Lan
dsli
de
Monitori
ng in
Nu
sa Ten
gga
ra … (He
r
ry Z
.
Kotta)
11
The g
ene
rate
d value
of accele
rom
e
ter,
X and Y, sho
w
the
ch
ang
es in
ho
rizon
t
al and
vertical an
gle
.
In other wo
rds, the value
of the accele
rometer n
ode
X means
cha
nging
corne
r
to
the left and
right a
c
cordi
ng to the X
-
axis, whil
e t
he value
of
Y mean
s tha
t
node
ch
an
ges
accele
rom
e
te
r with ve
rtica
l
slop
e a
s
Y-axis.
Fig
u
re
1 sh
ows a
visual rep
r
e
s
entation of t
he
movement of node
s in the
event of chan
ges in a
c
cele
romete
r X an
d Y.
Figure 1. Visualization of Nod
e
Angle
Cha
nge
s [13]
Figure 2. Measu
r
em
ent an
d monitori
ng
of real-time
data land
slid
es a
s
a real
-ti
m
e system
(refer to [7])
2.2 WSN
Co
nfigura
t
ion o
n
data collec
t
ion
In this research, the u
s
in
g
ad-h
o
c type
o
f
netwo
rk to
p
o
logy simil
a
r
to [12]. This i
s
a type
of network to
pology in
whi
c
h
a central '
r
oot' no
de
(t
he
top level
of the hie
r
a
r
chy) is
con
n
e
c
ted
to
one
or mo
re
other no
des t
hat a
r
e
one
l
e
vel lo
wer
in
the hie
r
a
r
chy (i.e., the
second l
e
vel) wit
h
a
point-to
-
poi
nt link betwe
en
each of the se
con
d
le
vel node
s and th
e top level ce
ntral 'ro
o
t' no
de.
While
ea
ch o
f
the se
con
d
l
e
vel nod
es th
at are
con
n
e
c
ted to the to
p level centra
l 'root' n
ode
will
also
have o
n
e
or m
o
re oth
e
r no
de
s that
are
one level
lowe
r in the
hiera
r
chy (i.e.
,
the third lev
e
l)
con
n
e
c
ted to
it. Gateway t
hat provide
s
l
i
nk
betw
een
sen
s
o
r
s an
d
central p
r
o
c
e
s
sor act
s
as root,
and
nod
es re
fer to
every
sensor in th
e
system.
A
c
tu
ally,
every
n
o
de can
send
data
di
re
ctly to
gateway, but in certai
n ca
se whe
n
link b
e
twee
n
one n
ode to gateway is bro
k
en,
it can se
nd d
a
ta
via other nod
e (ad
-
ho
c).
The land
slid
es monito
rin
g
in real-t
im
e system, that be used
as the sche
me
in this
resea
r
ch, is shown in Figu
re 2.
3. Results a
nd Analy
s
is
3.1 Measur
e
ment and
Da
ta An
aly
s
is
Measurement
usin
g a p
r
o
t
otype is inte
nded to fa
cili
tate analysi
s
of data on
variou
s
con
d
ition
s
. Simulation of th
e movement
of land to
see
the avalanch
e
effect is do
ne manu
ally by
stre
ssing the
grou
nd. Outp
ut of MoteView software
[1
3], [14] which
is used in thi
s
work is
sho
w
n
in Figu
re
3. It uses fo
rmul
a
(1
) that di
re
ctly c
onverte
d
into the M
o
te
View a
nalysi
s
software. T
h
e
results of dire
ct analysi
s
in
grap
hical form for each co
ndition are sh
own in Fig
u
re
4.
The analy
s
is
of each d
a
ta usin
g a protot
ype is as follo
ws:
a.
Con
d
ition I: the normal co
ndition (b
efore the
forced
movement of
landsli
de) i
s
depicte
d
in
Figure 4. It a
ppea
rs that
all nod
es
(sen
sors), 5,
6, 8,
9 an
d 1
0
rem
a
ined
un
ch
an
ged i
n
b
o
th
X and Y acce
lerato
r data.
b.
Condition II:
the condition that is affected
by ground movement (l
andslides 1)
depi
cted in
Figure 5. Thi
s
figu
re
sho
w
s th
at no
de
5 from
time
13:02 to
13:0
3
su
bje
c
t to
cha
nge, the
cha
nge p
o
siti
on of x accel
e
rom
e
ter, ap
proximately -0.3 to -0.7 g while in y accele
rom
e
ter
cha
nge
d sign
ificantly
from
positio
n
0
to +1 g.
Th
ere
a
r
e vi
sible
but
not si
gnificant
ch
ang
es in
node 6 a
nd 9.
c.
Condition III: the condition t
hat is affected by ground move
ment (l
andslides
2) depi
cted
in
Figure 6. No
de 10 ha
s ch
ange fro
m
time 13:04 unt
il 13:05. Posi
tion of x accelero
meter i
s
Evaluation Warning : The document was created with Spire.PDF for Python.
¢
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 9, No. 1, April 2011 : 9 – 18
12
cha
ngin
g
fro
m
+0.38 to 0
g, while y accele
romete
r is ch
angin
g
positio
n from
0 to about
-0.1 g.
Figure 3. The
topology of WSN o
n
the data retri
e
val
d.
Con
d
ition IV: the con
d
itio
n that is affe
ct
ed by soil
movement (l
and
slide
s
3
)
depi
cted in
Figure 7. At
node 8 from time 13:05 to
13:06 of
sig
n
ificant chan
ge, in x acce
lerom
e
ter it
cha
nge
d po
si
tion from 0 to
-3.34 g
whil
e
in y accele
ro
meter it chan
ged p
o
sition f
r
om 0 to
-
0.306 g. No
d
e
6 experie
nced no si
gnificant cha
nge.
e.
Con
d
ition V: the co
ndition
that is affect
ed by groun
d
movement (l
and
slide
s
4
)
depi
cted in
Figure 8. In this co
ndition
, node 9 and
10 from
time 13:08 until
13:09 of not significant
cha
nge.
f.
Con
d
ition
s
VI: the con
d
itio
n that is affe
ct
ed by gro
u
n
d
moveme
nt (land
slid
es
5) depi
cted in
Figure 9. It seem
s that
node
s 9 a
n
d
10 ch
ang
e from time 13
:09 until 13:
10 but not
signifi
cantly. At node 9: b
o
t
h x and y a
c
cele
rom
e
ter i
s
chan
ging
p
o
sition f
r
om
-0.2 to -0.1
g.
At node 10: b
o
th x and y acceleromete
r are chan
ging
positio
n from
0.22 to 0.1 g.
g.
Condition VII: the condition t
hat is affected by ground move
ment (l
andslides
6) depi
cted
in
figure
10. It a
ppea
rs that o
n
ly node
9 from time
1
3
:1
1 until 1
3
:12
of signifi
cant
cha
nge
s. X
accele
rom
e
te
r un
cha
nged
while y accel
e
rom
e
ter
cha
nged p
o
sitio
n
from -0.2 to
-0.1 g. At
node 1
0
: both
x and y accel
e
rom
e
ter a
r
e
unchan
ged.
h.
Condition VIII: the condition that is
affected
by ground movement
(landslides
7)
illustrated in
Figure 11. It appe
ars that both nod
es 9
and 10 fr
om
time 13:13 un
til 13:14 of ch
ange, the x
accele
rom
e
te
r chan
ged
alt
houg
h not
sig
n
ificant, a
nd
y accele
rom
e
ter at n
ode
9
cha
nge
s
of
positio
ns
+0.
1
to +0.34 g; at node 10
ch
ange
s from
y accele
rom
e
te
r positio
n -0.3
to -0.56 g.
i.
Con
d
ition IX: the co
ndition
that is affecte
d
by
gro
und
movement (l
a
ndsli
de
s 8) ill
ustrate
d
in
Figure 1
2
. It app
ears th
a
t
only no
de
9 from
time 13:16 until
1
3
:17of cha
n
g
e
,
ch
ang
e
of
positio
n x accele
rom
e
ter
0.7 to -0.1
g, while th
e cha
nge of p
o
sitio
n
y accele
ro
meter
-0.6 to
+0.48 g.
j.
Con
d
ition X:
the conditio
n
that is
affecte
d
by
g
r
o
und
movement
(la
ndsli
de
s 9
)
ill
ustrate
d
i
n
Figure 1
3
. T
he
singl
e
re
maining
no
de
10 f
r
om
time
13:18
to
13:
19
cha
nge
d i
n
fluctu
ation,
stable x accel
e
rom
e
ter cha
nged raised from -0.1 to +0.84 g and th
en fell to a -0.64 g, while
y accel
e
ro
me
ter on chang
e
of position around
+0.2 to +0.52 g.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
¢
Wirel
e
ss Sen
s
or
Network for Lan
dsli
de
Monitori
ng in
Nu
sa Ten
gga
ra … (He
r
ry Z
.
Kotta)
13
Figure 4. Co
ndition I: normal
Figure 5. Condition II: landslide 1
Figure 6. Condition III: landslide 2
Figure 7. Co
ndition IV: landslid
e 3
Evaluation Warning : The document was created with Spire.PDF for Python.
¢
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 9, No. 1, April 2011 : 9 – 18
14
Figure 8. Co
ndition V: landslid
e 4
Figure 9. Co
ndition VI: landslid
e 5
Figure10. Condition VII: la
ndsli
de 6
Figure 11. Condition VIII: l
andslide 7
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
¢
Wirel
e
ss Sen
s
or
Network for Lan
dsli
de
Monitori
ng in
Nu
sa Ten
gga
ra … (He
r
ry Z
.
Kotta)
15
Figure 12. Condition IX: landsli
de 8
Figure 13. Condition X: landsli
de 9
From tho
s
e
data, it appears that chan
ges in
a
c
cele
romete
r valu
es ra
ngin
g
from 0.2g
(gravity) to 0.49g in both X and Y indicate
s soil
ha
s begu
n to move but not significa
ntly. The
value of
0.5
g o
r
mo
re
is
a value
that i
ndicati
ng
a
si
gnifica
nt cha
nge
of g
r
ou
n
d
motio
n
. Sta
r
ting
on 1
g
of gro
und m
o
tion i
ndicates a ve
ry strong
si
g
nal a
c
tivity and
sho
u
ld b
e
alarmed. T
h
ese
results
will be demonst
rat
ed and tested on a
sam
p
le of research sites
in the Ikanfoti village
,
Kupang
Dist
ri
ct, Nusa Ten
ggara Timu
r Province (10
o
16’ 21.9” LS
and 12
3
o
40’ 59,8”
BT).
3.2 Experiments a
nd An
aly
s
is on Test bed Lo
cation
The te
st bed
or te
st site
is lo
cate
d a
t
Ikanfoti village, Kup
ang
Dist
rict, Ea
st Nu
sa
Tengg
ara Province o
n
10
o
16’ 21.9” L
S
and 123
o
4
0' 59.8” BT
(as Pilot Proje
c
t). The p
r
im
ary
con
s
id
eratio
n
of the
site
lo
cation
that p
r
epared
by Bo
bona
ro
Co
mp
lex Formation
is th
at the
sit
e
con
s
i
s
ts of cl
ay mineral o
r
montmorillo
n
i
te
(expan
sive clays) and rock. Such sit
e
s are regi
on
ally
wide
sp
rea
d
a
nd ca
n rep
r
e
s
ent the gene
ral con
d
ition o
f
the area in
NTT.
Analys
is
of eac
h
data meas
urement is
as
follows
:
a.
Con
d
ition I: the no
rmal
co
ndition (befo
r
e the enf
o
r
ce
d moveme
nt of land
slide
)
i
s
de
picte
d
in
Figure 14. It
appe
ars that all the node
s (se
n
sors
), 5, 6, 8, 9 and 10 remai
ned u
n
ch
ang
ed i
n
both X and Y accele
rom
e
te
r.
b.
Condition II:
the condition that is affected by
ground movement
(land
slides 1) depicted in
Figure 15. It
can
be
se
en
that only no
d
e
6
subj
ect to
cha
nge
from
time 10:37
to 10:38.
The
cha
nge
po
sition of x a
c
cel
e
rom
e
ter
app
roximately -0.
8
to -0.6
g while the y a
c
cele
rom
e
ter
have -0.25 to
about 0 g.
c.
Condition III: the condition that is affected
by soil
movement (landslides 2) depicted i
n
Figure 16. Th
is figure
sh
o
w
s the
cha
n
g
e
of t
he node
s 8, 9 and 1
0
from time 10
:45 to 10:46.
Nod
e
1
0
, whi
c
h
ch
ang
ed
significa
ntly, in x a
c
celerom
e
ter
ch
ang
ed
from
po
sitio
n
-0.7 to
0
g
and of y acce
lerom
e
ter fro
m
0.25 to ab
out 0.94
g. Node 9 is
only experie
nci
ng
small
cha
nge
from po
sition
-0.1 to 0.2
g
at both x and
y ac
cele
rom
e
ter. Node
8
only slig
htly chang
ed, i.e.
the x accel
e
rometer from positio
n to po
sition -1.1
2 to -0.9 g.
Evaluation Warning : The document was created with Spire.PDF for Python.
¢
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 9, No. 1, April 2011 : 9 – 18
16
d.
Con
d
ition IV: the co
nditio
n
that is aff
e
cted
by soil
movement
(land
slide
s
3
)
depi
cted in
Figure 1
7
. It appe
ars that
only n
ode
s
5 an
d 8 f
r
om
time 10:4
9
t
o
10:5
0
a
r
e
subj
ecte
d to
cha
nge. At
n
ode
5, the
x a
c
celeromete
r
cha
nge
d p
o
si
tion to 0
-0.5
g while i
n
y
a
c
celeromete
r
cha
nge
d fro
m
po
sition
0.
2 to
0.7 g.
Node
8 a
r
e
ch
angin
g
p
o
siti
on, for x a
c
celero
meter u
n
til
the positio
n -1.58 to -0.9 g
;
while the y acceleromete
r cha
nge
d from
0.6 to about 0.8 g
.
e.
Con
d
ition V:
the co
ndition
that is affe
cted
by g
r
ou
nd
movement
(l
and
slide
s
4
)
depi
cted in
Figure 1
8
. It
can
be
seen
that nod
es 5,
8 a
nd
9
from time
10
:5
1 to
1
0
:
52
ar
e s
u
b
j
ec
te
d
to
cha
nge. At node 5; the x accele
rom
e
te
r ch
ang
ed
p
o
sition from
0 -0.5 g to 0
.
2 g, while y
accele
rom
e
te
r to 0.7 g. For node 8; x accele
rom
e
ter
appe
ars to ch
ange fro
m
po
sition -1.
57g
to -0.9 g whil
e in y accele
romete
r ch
an
ge from 0.6 g to 0.8 g. T
he sign
al no
de 8 end
s at
10:51. Fo
r
n
ode
9; the x
accel
e
ro
met
e
r
cha
nge
to
po
sition 1
g
0.1 (signifi
ca
ntly) whil
e y
accele
rom
e
te
r from -0.1 g
to 0.6 g. Signal node
9 also see
m
s to h
a
ve sub
s
id
ed
arou
nd the
time of 10:5
1
. Nod
e
s
6
and 1
0
ju
st still prov
ide
a co
nsta
nt sign
al (no
cha
nge
). Th
e
terminatio
n o
r
lo
ss of
sign
al no
de
s 5
a
nd 9
cau
s
ed
by sig
n
ifica
n
t gro
und
moti
on, so n
ode
s
can n
o
long
er emit a signal
to the gateway/network.
Figure 14. Conditioni I: no
rmally
Figure 15. Condition II: landslide 1
Figure 16. Condition III: landslide 2
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
¢
Wirel
e
ss Sen
s
or
Network for Lan
dsli
de
Monitori
ng in
Nu
sa Ten
gga
ra … (He
r
ry Z
.
Kotta)
17
Figure 17. Co
ndition IV: landslid
e 3
Figure 18. Co
ndition V: landslid
e 4
The re
se
arch
has be
en d
one u
s
ing eit
her
a p
r
ototype or sampl
e
of measu
r
ements
dire
ctly on
th
e site,
giving
the result tha
t
t
he a
ppli
c
ation of
WS
N
can b
e
applie
d p
r
op
erly. T
he
value of x
an
d y a
c
celero
meter
(Mi
c
a
z
device)
ar
e use
d
to dete
c
t
vibratio
ns cau
s
e
d
by ground
movement i
n
land
slid
e a
r
ea. The
resu
lts of thi
s
works indi
cate
that chan
ge
s a
c
cele
rom
e
ter
values ra
ngin
g
from
0.2
g t
o
0.49
g i
n
b
o
th x an
d y a
c
celeromete
r
indicates soil
begin
to m
o
ve
but not signif
i
cantly. Above 0.5 g
is a value that indicatin
g
a si
gnifica
nt cha
nge of land
sl
ide
.
Starting on
1
g and
above
of land
slide h
a
ve a very
st
rong
signal
activity and sho
u
ld be
alarm
e
d.
This conditio
n
will
be
interpreted
by Ea
rly Wa
rning
S
y
stem (SP
D
)
as
a d
ang
er
con
d
ition
and
to
give sign
al to the comm
unit
y
surroun
ding
the location
of landsli
de
s area [8].
4. Conclusio
n
The pa
per h
a
s p
r
e
s
ented
the research
that
demon
strates the im
plementatio
n of WS
N
for monito
rin
g
land
slide in
NTT province. It can
be concl
ude
d that the cha
nge
s
in the value o
f
X
and Y a
ccel
e
romete
r re
sul
t
from gro
und
motion from
a positio
n of
0.2 g to 0.49
g indicate th
e
occurre
n
ce o
f
slow g
r
ou
n
d
motion. Value of
0.5 g o
n
X and Y acceleromete
r sho
w
sig
n
ificant
cha
nge
s in g
r
oun
d move
ment and val
ue of 1 g and above of groun
d motion
could lea
d
to
disa
ster. Th
e
system ca
n
be integrate
d
with
an ea
rly warning system as pa
rt of continu
ed
resea
r
ch to provide info
rmation for i
mmediate d
a
nger info
rma
t
ion to the publi
c
abo
ut the
locatio
n
of landsli
de. Furth
e
r re
sea
r
ch is expe
ct
ed to be realized i
n
a system that is con
n
e
c
ted
into the Internet network
by
utilizing the
GIS system.
Evaluation Warning : The document was created with Spire.PDF for Python.
¢
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 9, No. 1, April 2011 : 9 – 18
18
Ackn
o
w
l
e
dg
ments
This
re
sea
r
ch
wa
s fund
ed
by Dire
cto
r
at
e Gen
e
ral
of High
er Ed
uca
t
ion (DGHE
)
,
Ministry
of National E
ducation Re
p
ublic of Indon
esia
(Hib
ah Kompeten
si Ba
tch II 2010). Than
ks a
r
e al
so
addresse
d to Dr. Wirawan
,
DEA and Prof. Dr. Ts
uyo
s
hi Usa
g
a
w
a
as pee
rs, th
e PREDICT and
Lab. Multime
d
ia Tele
com
m
unication
s of ITS,
Lab.
UNDANA Mi
ning Engi
n
e
e
r
ing, Energy and
Mining Depa
rtment of NTT Province, and
the
various parties involved in this works.
Referen
ces
[1]
W
i
ra
w
an,
Rac
h
man S, Pr
ato
m
o I, Mita N.
Desig
n
of
Low
Cost W
i
rel
e
ss
Sensor
Netw
o
r
ks - Base
d
Enviro
n
m
enta
l
Monitori
ng
Sys
t
em for
Dev
e
l
o
pin
g
C
ountry
.
Proc. Int. Conf
erenc
e APC
C
14th As
ia-
Pacific. T
o
kyo, Japa
n. 200
8; 1-5.
[2]
Jamal
udi
n MZ
, Arip
in
NM, Isa
AM, Moh
a
me
d HW
L.
W
i
r
e
le
ss Soi
l
T
e
mp
er
ature
an
d Sl
op
e Incl
inati
o
n
Sensors
for S
l
op
e Mo
nitor
i
n
g
Syste
m
.
Pr
ocee
din
g
s of
I
n
ternati
o
n
a
l C
onfere
n
ce
on
Energ
y
a
n
d
Enviro
nment (ICEE). Selan
g
o
r
, Mala
y
s
ia. 20
06.
[3]
Mane
esha V R
a
mesh, San
g
e
e
th Kumar, P Venkat Ra
nga
n.
W
i
reless Se
nsor Netw
ork for Lan
dsli
de
Detectio
n
. Pro
c
eed
ings
of th
e 20
09 Int
e
rn
ation
a
l C
onfer
ence
on W
i
r
e
l
e
ss Net
w
o
r
ks
(ICW
N). Las
Vegas N
e
va
da
, USA. 2009; 2: 89-95.
[4]
W
ang Ri
xu. Ni
e Le
i.
T
he Des
i
gn
and A
p
p
lic
ation
of T
i
me Predict
i
on a
nd
Spatia
l Pred
icti
on i
n
F
u
shu
n
W
e
st Open Pi
t Lands
lid
e M
onitor
i
ng
an
d
Predicti
on Sys
t
em.
Proc. Int. Confer
enc
e
Secon
d
W
R
I
Globa
l Con
g
re
ss on Intelli
ge
n
t
Sy
stems.
Xi
a
m
en, Chi
na. 20
10; 3-7.
[5]
Ryan
Ditya
Permadi.
Si
mulasi Si
ste
m
Deteksi T
anah L
ong
so
r deng
an Jaringan Sen
s
or
Nirkab
el. Fin
a
l Proje
c
t. Surabaya: Ele
c
trical En
ginee
ring ITS. 2009
.
[6]
Cesar
e
Ali
ppi,
Romol
o
C
a
mpl
ani, Cristi
an G
a
lp
erti, Manu
el
Roveri.
Effecti
v
e des
ign
of
W
S
Ns: fro
m
the la
b to the r
eal w
o
rl
d
. Proc
eeding
of IEEE Inter
national
Conference on S
ensing T
e
chnology
(ICST
)
3
rd
.
T
a
ipe, T
a
iw
a
n
. 20
08; 1-9
.
[7]
Jung
w
o
o
Le
e
M S, Real-tim
e
Monitor
i
ng
of
Lan
ds
li
de
usin
g W
i
rel
e
ss Se
nsor N
e
t
w
ork.
PhD T
hesis.
Ohio: T
he Ohio State Universit
y
; 20
09.
[8]
Herr
y
Z
Kotta, Silvester T
ena, Gregorius Kl
a
u
, K Rantelo
b
o
.
Applicati
on of
Geograp
hic
a
l
Informatio
n
System (GIS) for Mapp
ing
L
ands
lid
e Susc
eptib
ility:
A Ca
se Study of Ti
mor Te
ng
ah S
e
lata
n, NT
T
Provinc
e
. Proceed
ings of N
a
ti
ona
l Semin
a
r o
n
A
ppl
ied T
e
ch
nol
og
y, Scie
nc
e, and Arts (1
st
AP
T
E
CS).
Surab
a
y
a. 20
0
9
; 13-17.
[9]
Chu
anh
ua
Z
h
u
,
Xu
ep
ing
W
a
n
g
. La
ndsl
i
d
e
S
u
scepti
b
il
ity M
app
ing:
A
C
o
mparis
on
of Info
rmati
o
n
an
d
W
e
ights-of-Evi
denc
e Met
hod
s in
T
h
ree
Go
rges Ar
ea.
Int
e
rnati
ona
l C
o
n
f
erence
o
n
En
vironme
n
t
a
l
Scienc
e an
d Information Ap
pli
c
ation T
e
c
hno
l
o
g
y
. W
uhan, C
h
in
a. 200
9; 34
2-34
6.
[10]
K
y
o
ji Sass
a.
Editors
. Lan
dsli
de: Risk Ana
l
ysis
and S
u
sta
i
na
ble D
i
saster
Manag
eme
n
t. Proceed
in
g
of F
i
ret Genera
l
Assembl
y
of the Int. Consor
ti
um on La
nds
lid
e. Spring
er-Ber
lin. NY. 200
5.
[11]
WEN Hai-ji
a, L
I
Xin, ZHANG
Jia-la
n.
An
Eva
l
ua
ti
o
n
-
Ma
nag
em
en
t In
fo
rma
ti
o
n
System
of H
i
g
h
Slo
p
e
Geo-risk for Mounta
i
n
ous Cit
y Based on GIS.
Internation
a
l
Confer
ence o
n
Information
Scienc
e an
d
Engi
neer
in
g (ICISE). Nanji
n
g
,
China. 20
09;
197
6-19
78.
[12]
Karl Hol
ger, Willi
g Andr
eas. Protocols a
nd
Architectures f
o
r Wireless Se
nsor Net
w
orks
.
John W
ille
y
and So
ns. 200
5.
[13]
Crossbo
w
T
e
chnology
.
X Me
sh User
’
s
Ma
n
ual – R
e
visi
on
C.
Crossbo
w
T
e
chn
o
lo
g
y
, Inc. 2007.
[14]
Crossbo
w
T
e
chnology
.
Mote
View
Users Ma
nua
l
. Crossbo
w
T
e
ch
no
log
y
,
Inc. 2007.
Evaluation Warning : The document was created with Spire.PDF for Python.