TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol.12, No.1, Jan
uary 20
14
, pp. 275 ~
279
DOI: http://dx.doi.org/10.11591/telkomni
ka.v12i1.3967
275
Re
cei
v
ed Ma
y 23, 201
3; Revi
sed Aug 8,
2013; Accept
ed Aug 26, 2
013
Resear
ch of Embedded GIS Data Man
a
gemen
t
Strategies for Large Capacity
Shi Bei-lei*, Fang Kun, Wu Ke-ming, Ye Li-
w
ei
Naval U
n
iv
ersit
y
of Eng
i
ne
eri
n
g, W
huhan, C
h
ina, 13
50
717
7
976
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: meimei0
5
2
5
@
qq.com
A
b
st
r
a
ct
W
i
th the use of data for embed
de
d GIS system
conti
n
u
e
s to increas
e
and the req
u
i
r
ement o
f
app
licati
on for
embe
dde
d GIS system co
ntinu
e
s to i
m
pr
o
v
e, the qu
ad-t
r
ee in
dex
alg
o
rithms
and
bl
o
ck
cl
a
ssi
fi
ca
tio
n
da
ta
o
r
ga
ni
z
a
ti
on
mo
de
th
a
t
are
cu
rre
ntly us
ed to
han
dl
e l
a
rge
a
m
o
unts
of data r
e
flects
a
certain
li
mitati
o
n
. Co
mb
ini
ng t
he ch
aracterist
ics of
e
m
b
e
d
d
ed GIS data, t
he a
u
t
hors
put
forw
ard the
multi-
level
data in
de
xing a
nd dyn
a
m
ic d
a
ta lo
adi
ng, and re
ali
z
e
the data loa
d
i
ng w
hen req
u
ir
ed, and e
n
h
a
n
c
e
the real-ti
m
e re
spons
e spe
ed, solves
the l
i
m
it
ation o
n
lar
ge
volu
me data.
Ke
y
w
ords
: Emb
e
d
ded GIS, Multi-lev
e
l D
a
ta Indexi
ng, Dy
na
mic D
a
ta Lo
adi
ng
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
With the
dev
elopme
n
t of
mobile te
ch
n
o
logy
(m
obile
phon
es,
digit
a
l pe
rsonal
a
ssi
stant
)
,
more
and
mo
re p
eopl
e be
gin to feel
th
e ben
efits of
embed
ded
G
I
S technol
og
y, and be
gan
to
use in daily life of embedd
ed GIS. Similar to
the desktop GIS, sp
atial
data ma
nagem
ent is
also
suitabl
e for th
e embe
dde
d
GIS system. It is re
spo
n
si
b
l
e for ho
w to
orga
nize the
data in a file
or
databa
se, ho
w to mana
ge
the memory
spa
c
e
whe
n
a pro
g
ra
m st
arts, an
d ho
w to improve
the
retrieval
effici
ency i
n
la
rge
amount
s of
d
a
ta. But be
ca
use
of the
co
nstrai
nt of te
rminal e
quipm
ent
resou
r
ces of
the em
bedd
e
d
GIS, the
au
thor
also n
e
e
d
to
co
nsid
er the
app
rop
r
i
a
te st
rategi
es to
solve th
e
con
t
radictio
n
bet
wee
n
the
limi
t
ed sto
r
a
ge
a
nd a
la
rge
a
m
ount
of dat
a, and
satisfy
the
real
-time dem
and.
2. The Comp
arison of
Ch
arac
teris
t
ics
for Commo
n GIS Data a
nd Embedde
d GIS Data
2.1. Chara
c
teristics o
f
Common GIS Data
GIS database is the geog
raphi
cal featu
r
es
set in
a certain region,
mainly relate
d to the
manag
eme
n
t and o
r
ga
nization for the
grap
hics a
n
d
attribute
dat
a. For o
r
din
a
ry databa
se,
GIS
data ha
s the followin
g
ch
aracteri
stics:
(1) the GIS d
a
taba
se not o
n
ly has the at
tri
bute data of
geographi
cal
factor that is
simila
r
to co
mmon
d
a
ta ba
se,
but
also
ha
s la
rge n
u
mb
er o
f
spatial
d
a
ta
, and
they h
a
ve in
sep
a
ra
ble
conta
c
t.
(2)
the geo
graphi
c system
is a compl
e
x
giant
system
, whi
c
h
de
scri
bes the
re
sou
r
ce
an
d
environ
ment
with a large
amount of da
ta, even if
a
tiny area, the
application
s
are very b
r
oa
d,
su
ch a
s
geo
grap
hic
re
se
arch, enviro
n
m
ental prote
c
tion, land u
s
ag
e and pl
annin
g
, reso
urce
exploitation, ecol
ogi
cal en
vironme
n
t, munici
pal man
ageme
n
t, roa
d
con
s
tru
c
tio
n
and others.
The
cha
r
a
c
teri
stics mentio
ned
above de
cid
e
when
est
abli
s
h a GIS dat
aba
se, we
sh
ould follo
w a
nd
use
the g
ene
ral p
r
in
ciple
and m
e
thod
of the comm
on dat
aba
se,
on the
othe
r hand,
we
m
u
st
take
som
e
speci
a
l te
chni
que
s a
nd m
e
thods to
solv
e spatial
dat
a man
age
me
nt that do
es
no
t
exist in othe
r data. Becau
s
e of the
obv
ious
sp
atia
l chara
c
te
risti
c
s of GIS data
base, we
call
it
s
patial databas
e
[1].
2.2. Chara
c
teristics o
f
Embedded GI
S Data
With the dev
elopme
n
t of embed
ded te
chn
o
logy
an
d
GIS technol
ogy, the appli
c
ation of
embed
ded
G
I
S, which i
s
the pro
d
u
c
t
of the co
mb
i
nation of the
m
,expand
s
with irre
sistib
le
momentum of
rapid develo
p
ment.
Embe
dded GIS
data,
whi
c
h not
only ha
s the
cha
r
a
c
teri
stics of
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 2302
-4046
TELKOM
NIKA
Vol. 12, No
. 1, Janua
ry 2014: 275 – 2
7
9
276
GIS data of large
amou
nt of compl
e
x data, but al
so has th
e ch
ara
c
teri
stics
of itself that the
stora
ge
spa
c
e is sm
all an
d the cal
c
ulat
ion of retrieva
l is simpl
e
an
d quick. Thi
s
lead
s to 2 ma
jor
contradi
ction
s
mu
st be re
solved in
the d
e
velopme
n
t of embedde
d GIS data:
(1) T
he contradictio
n of large qua
nt
ities
of data and small storage
spa
c
e.
The complex
spa
c
e data
relative to the traditional
relational d
a
ta
base data, the data
quantity is l
a
rger, a
nd the
data di
strib
u
tion is
mo
re u
neven. But in
the emb
edd
ed sy
stem, t
he
stora
ge
sp
ace is very p
r
e
c
ious,
whi
c
h
requires mo
re
effective
stru
cture
an
d m
o
re effe
ctive way
to make GIS data sto
r
ed in
less
spa
c
e a
nd used in smaller me
mory.
(2) Th
e co
n
t
radictio
n bet
wee
n
embe
d
ded calcul
ation sp
eed a
nd com
p
lex
spatial
retrieval d
e
m
and.
GIS query the
databa
se
i
s
the main re
aso
n
re
late
d to spatial po
sition
an
d
not
directly
attribute dat
a whi
c
h le
d
to the
GIS databa
se
i
s
much
more
compli
cate
d
than traditi
onal
databa
se. Bu
t the shorte
r,
faster p
r
oce
s
s
re
qui
reme
nts
in emb
e
d
ded system, and embe
dd
ed
pro
c
e
s
sor it
self pro
c
e
s
sin
g
sp
eed, let
us m
u
st u
s
e
a me
cha
n
ism more flexible than
de
sktop
operating sy
stem to meet the req
u
ireme
n
ts of
embe
d
ded sy
stem. The contra
dictions ab
ove a
r
e
the cha
r
a
c
teri
stics of embe
dded GIS dat
a [2].
To solve the
s
e 2 cont
radi
ctions, 2 a
s
pe
cts can b
e
co
n
s
ide
r
ed:
(1)
Usin
g the appropri
a
te data com
p
re
ssi
on technolo
g
y. The stora
ge space of
embed
ded d
e
vice
s is far l
e
ss than the
desktop
co
m
puter
stora
g
e
spa
c
e. Altho
ugh the data
can
be sto
r
ed i
n
block o
r
can
be do
wnlo
a
ded dyna
mi
cally throug
h the network,
but the effect
ive
method for ef
ficient com
p
ression of GIS
data is
the m
o
st effective and e
c
on
omi
c
al metho
d
.
(2) Th
e rea
s
o
nable spatial index and re
al
-tim
e re
spo
n
se. The low proce
s
sing spe
ed and
limited memo
ry spa
c
e of embedd
ed pro
c
e
s
sor, ma
ke
s it impossibl
e to load large amou
nts of
data into
me
mory for op
e
r
ation
at the
sam
e
time,
To a
c
hi
eve
real
-time
re
spo
n
se for d
a
ta
operational
reque
st, the a
ppro
p
ri
ate m
e
thod of
data
index a
nd
d
a
ta organi
zati
on i
s
ne
ede
d
to
adopt to divid
e
the data fo
r re
du
cing th
e amou
nt
of read d
a
ta at o
n
ce, a
nd the
n
the amo
unt
of
cal
c
ulatio
n to CPU is
red
u
ced and the el
ectri
c
po
we
r is save
d.
3. Analy
s
is o
f
the
Data M
a
nageme
nt
Strateg
y
3.1. Data M
a
nagemen
t
Strategy
for Embedded GI
S Data
For
embe
dde
d GIS, due to
the limitation
of CP
U spee
d, battery p
o
w
er, i
n
tern
al
memory
cap
a
city, external mem
o
ry
cap
a
city and
scree
n
di
spla
y quantity, we can not com
p
letely copy the
data ma
nag
e
m
ent ba
se
d
on PC i
n
to e
m
bedd
ed
GIS.
Generally, the metho
d
whi
c
h i
s
com
b
ined
regio
n
qua
d-t
r
ee ind
e
x storag
e stru
ctu
r
e with
the spatial data o
r
gani
zatio
n
o
f
blockin
g
an
d
layering
is a
dopted
for the ma
p d
a
ta
. After dividi
ng the
ma
p
into blo
c
k, the g
eog
rap
h
i
c
al
feature
s
in the block are
classified on the basi
s
of the importa
nce, and
the cl
assificatio
n
s
are
orga
nized
by
compl
e
me
nta
r
y no
n-red
u
n
dant m
e
mo
ry
st
ru
cture, in
this
way, it i
s
realized th
at t
h
e
data rea
d
ing
in the block according to the need of
classificatio
n
. The geo
gra
p
h
ical featu
r
e
s
in
the same
cla
ssifi
cation a
r
e store
d
into
layers, and
are di
splaye
d by layers
control. For the
embed
ded GI
S platform ba
sed o
n
Win
d
o
ws mobile
, data
storage use CEDB
or the combin
ation
of EDB with file form [3].
3.2. Data M
a
nagemen
t
O
p
timization
for Embedde
d GIS Platfor
m
3.2.1. Shorta
ge of Embed
d
ed GIS Platform for
Data Manag
e
me
nt
The
meth
od whi
c
h
i
s
com
b
ined re
gion quad
-tre
e
ind
e
x storage
structure
with th
e sp
atial
data organi
za
tion of blo
cki
ng an
d layeri
ng solved th
e
cont
radi
ction
betwe
en la
rg
e ca
pa
city da
ta
and
limite
d
storag
e spa
c
e
in
a ce
rtain e
x
tent,
but wit
h
its
expan
di
ng rang
e of
appli
c
ation
a
n
d
increa
sing d
e
m
and of bu
si
ness , it also
reflect
s
certai
n limitations [4].
1. Data Inde
x and Organi
zation:
(1)
The bl
ock an
d cl
assif
i
cation m
e
th
od an
d
layers control
ma
ke
s the d
a
ta
sent to
memory
a
s
li
ttle as po
ssible, but
data
in the
laye
r
are
still
all lo
aded
into
me
mory. Althou
gh
embed
ded
G
I
S applicatio
n relate
d to
data modifi
cation op
erati
ons, it ra
rely
involved in
the
operation fo
r all featu
r
e
s
in the laye
r.
For l
a
yers th
at had
no
ch
ange
or mod
i
fied very
sm
all
amount, there is no ne
ed to load all dat
a into memory at one time.
(2) Th
e incon
v
enien
ce of data exchan
ge
to
large GIS softwa
r
e. Because the data
index
and o
r
ga
nization metho
d
s a
r
e u
s
e
d
in different
ways, b
e
fore exch
ange
d
,
data must
be
conve
r
sed int
o
the structu
r
e of
co
rrespo
nding pl
atform at first,
this method hi
nd
ered th
e sharing
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Re
sea
r
ch of Em
bedded G
I
S Data Mana
gem
ent Strategie
s
for La
rg
e Cap
a
cit
y
(S
hi Bei-lei)
277
and exchan
gi
ng of data be
tween pl
atforms to som
e
e
x
tent, also ca
use
d
the co
n
v
ersio
n
proce
s
s
too cumb
erso
me.
2.
Data Storage:
Usin
g
CEDB as
the databa
se, we impro
v
ed the efficien
cy of
informatio
n q
uery which i
s
irre
spe
c
tive to spat
ial dat
a. But CEDB as the data
b
a
se in p
r
a
c
tical
appli
c
ation al
so refle
c
ted
some sh
ort
c
o
m
ings:
(1) Th
e
CEDB dat
aba
se
doe
s
not
suppo
rt th
e
PC te
rminal
o
peratin
g. Usually the
attribute data
is written into the file, when the pr
o
g
ra
m first run
s
, it read
s the attribute data from
the file to
form databa
se. So when the
prog
ram i
s
first runni
ng, it
is usu
a
lly su
spend
ed for the
transi
ent p
r
o
c
ess of buildi
n
g datab
ase [
5
]. When
th
e
attribute dat
a is too la
rg
e
(more than
5M)
or the datab
a
s
e table h
a
s
too many
fields (g
re
ater th
an 12), it ma
y lead to failure of co
ntribu
te
databa
se a
n
d
affect the stable ope
ration
of the prog
ra
m.
(2) T
he fussy data que
ry an
d modify operation lead to l
o
w efficie
n
cy.
CEDB
datab
ase
supp
ort
only inqui
ry o
f
a si
ngle
field, the
sup
port for SQL
is
also
very
simple sub
s
et
.
3.2.2. Optimization o
f
Embedded GI
S Platform for Data Man
a
gemen
t
The limitation
s
of the e
m
b
edde
d GIS pl
atform
reflect
ed in terms
o
f
data mana
g
e
ment,
esp
e
ci
ally the operatio
n of the large am
o
unt of
data in
the city pipe netwo
rk a
nd other ne
ede
d
is
particula
rly obvious. Using
the multi-level data
indexi
ng, dynamic
data loadin
g
and SQL Mo
bile
databa
se ca
n solve these probl
ems v
e
ry well [6
].
Optimizin
g
in
the data managem
ent of the
embed
ded GI
S platform is
mainly from the followi
ng a
s
pe
cts:
1
.
Usin
g Me
thod of the Multi-le
vel Data Indexi
n
g
and Dynam
ic Data Lo
a
d
ing to
Optim
i
ze Dat
a
Indexi
ng an
d Orga
nizatio
n
Mode:
(1) The
multi-level data i
n
d
e
xing: The
whole m
ap i
s
d
i
vided an
d
stored
a
c
cordi
n
g to the
admini
s
trativ
e regio
n
as
the unit, and
the index
of map is est
ablished to
swit
ch bet
we
en
admini
s
trativ
e regio
n
s. F
o
r the bigge
r admini
s
trati
v
e region, the multi-leve
l index can be
establi
s
h
ed to stora
ge an
d switch. For the data of
single map, the data is org
anized by layers.
Thro
ugh the
maximum and minim
u
m display scale ma
rk is added in
each layer
head
information, the platform realiz
es the controlling of
classifi
cation between
la
yers ,and data
will
be loa
ded
int
o
the m
e
mo
ry when
di
spl
a
yed in
th
e current scale. Thro
ugh
t
he method of
m
u
lti-
level data
ind
e
xing, data
o
r
gani
zatio
n
i
s
maintain
ed
consi
s
tent
with
GIS softwa
r
e on
the
PC f
o
r
the excha
ngi
ng and
sha
r
in
g of data with
PC.
Figure 1. Multi-lever in
dex of map
(2)
Dynami
c
data loadi
ng
based on l
a
yer: the laye
r
p
r
ocessin
g
is u
s
ed fo
r si
ngle
map, at
the same ti
me, the method of dyna
mic data lo
a
d
is used in
the map loading. Durin
g
map
loadin
g
, only the map laye
r index table
is rea
d
and
stored in mem
o
ry perm
ane
ntly. Only when
queri
ed o
r
modified, the d
a
ta is re
ad int
o
memo
ry for corre
s
po
ndin
g
operation b
y
seeki
ng ind
e
x
index of district
dis
t
ric
t
1
d
i
st
rct
2-1
d
i
st
ri
ct
2-2
d
i
st
ri
ct
2-3
se
c
o
nda
r
y
inde
x of dis
t
ric
t
2
dis
t
ric
t
1
dis
t
ric
t
1
La
ye 1
La
ye 2
La
ye 3
points, lines,
a
r
ea
s
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 2302
-4046
TELKOM
NIKA
Vol. 12, No
. 1, Janua
ry 2014: 275 – 2
7
9
278
numbe
r of
co
rre
sp
ondi
ng l
a
yer an
d feat
ure from t
he
map laye
r ind
e
x table, and
data is
rele
ased
from the me
mory after th
e operation. it
save more m
e
mory spa
c
e
in this way.
2. Data Modif
i
cation:
Due to the data dynamic lo
adin
g
method an
d layer proce
ssi
ng for
singl
e m
ap, t
he d
a
ta
(in
c
l
uding
the
sp
atial data
an
d attrib
ute d
a
t
a) mo
dificati
on i
s
m
ana
g
ed by
dynamic in
de
x data table establi
s
hin
g
.
(1) T
he mo
dification of attribute d
a
ta a
nd
its structu
r
e: By a unique nu
mbe
r
of the
feature
(ID), the feature a
ttri
bute data
index value
s
is read fr
o
m
the layer
index table
s
in
memory, a
n
d
then the
attri
bute
stru
cture of the fe
at
u
r
e i
s
read
form file a
s
well
as
data dyn
a
m
ic
index table is established.
When modifi
cation is
fini
shed, the inde
x table and its conte
n
ts a
r
e
relea
s
e
d
, and
the modified data is saved
in the databa
se.
(2)
The
modi
fication fo
r th
e sp
atial dat
a is
simila
r
with attrib
ute
data, but d
u
e
to the
spatial data i
s
use
d
more
frequently than the attr
ibute data. In
orde
r to ensure the re
al-t
ime
respon
se, th
e MapGIS-E
MS platform
is optimi
z
ed. Thro
ugh
the optimizat
io
n of
operation, not
only the abilit
y and efficien
cy for Ma
pGI
S
-EMS platform data m
a
n
ageme
n
t wa
s increa
sed, b
u
t
also the
pro
c
e
ss of d
a
ta
exchan
ge
wa
s sim
p
lif
ie
d. Before op
timization, th
e data bet
ween
embed
ded
e
quipme
n
ts o
r
embe
dded
equipm
ent with PC a
r
e e
x
chan
ged th
rough
switchi
ng
platform in th
e form of file. After optimizat
ion, althoug
h data betwe
en the emb
e
dded d
e
vice
and
PC a
r
e ex
cha
nged
still in t
he form of file
, but
emb
edd
ed devi
c
e
s
can exchan
ge
data di
re
ctly. By
Usi
ng the m
u
lti-level data i
ndexing, dyn
a
mic d
a
ta lo
ading a
nd
store the
attrib
ute data thro
ugh
the SQLMob
ile for map
data, the
efficien
cy
co
mpari
s
o
n
be
fore optimization and a
fter
optimization i
s
showed in t
able 1,
1:5 m
illion ma
p of
Wuhan
City (after conversi
on is 2.2M, 13
layers, 87
47 feature
s
) is u
s
ed:
Table 1. Co
m
parative Effici
ency Before and After Opt
i
mization
Test methods
Before optimization
After optimization
The SQLmobile is adopted to
replace the origin
al CEDB as the
attribute datab
ase
A single database cannot be
greater than
5M, the
database table fields cannot
more than 1
2
A single data table cannot
be greate
r
than
5
12M, the
database cannot
be greate
r
than 4G,
unlimited number
of fields
The SQLmobile database and th
e
same display
me
nthod are
adopted to first r
un the pro
g
ram t
o
dra
w
1:1 m
ap in
real-time
w
hen
the all lay
e
rs on
35~38 seconds
18~20 seconds
SQLmobile data
base is adopte
to quer
y in the m
ap
w
i
th the
condition that High
w
a
y
grade is
arterial road
and
name is luoshilu
1~2 seconds, and not good
at surpporting t
h
e complex
contition quer
y
i
n
g
Less than 1 second, and
good at surpp
o
rti
ng the
complex contition quer
y
i
n
g
SQLmobile data
base is adopte
to add the ne
w field of manager
and to insert def
ault value in the
ne
w
field in the communal
facilit
ies lay
e
r(1
145 points)
About 2 seconds
Less than 1 second
4. Summar
y
Due to
the
ch
ara
c
teri
stics
of embe
dde
d
GIS data
itself, resultin
g in
the pa
rticul
ari
t
y of its
data man
age
ment, and th
ere a
r
e
some
limitati
ons o
n
the data m
anag
ement o
f
embedd
ed
GIS
platform u
s
e
d
at pre
s
e
n
t, this pap
er
pre
s
ent
s the
use of d
a
ta
cla
ssifi
catio
n
indexing,
data
dynamic l
o
a
d
ing a
s
well
as the SQ
L
m
obile d
a
tab
a
se to o
p
tim
i
ze the e
m
b
edde
d GIS d
a
ta
manag
eme
n
t module, the data access efficiency a
nd sup
p
o
r
t on large volu
me data can
be
signifi
cantly improve
d
.
Referen
ces
[1]
T
ang Minan,
W
ang
Xi
aomi
n
g, Yuan S
h
u
a
n
g
. Site Se
l
e
ctio
n of Mech
anic
a
l Parki
ng S
y
st
em Base
d o
n
GIS w
i
th
AFRARBMI.
Indones
ian Jo
urna
l of Electrical E
ngi
neer
ing
. 2
013;
11(7): 39
35-
39
44.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Re
sea
r
ch of Em
bedded G
I
S Data Mana
gem
ent Strategie
s
for La
rg
e Cap
a
cit
y
(S
hi Bei-lei)
279
[2]
F
an W
en-
you
,
Ren Nian-
hai, Li
u Qin. Research o
n
Embed
ded
Mobil
e
Data
b
a
se
an
d Its
Ke
y
T
e
chno
lo
g
y
.
Micro com
puter information
. Beiji
ng. 20
08; 6: 49-51
[3]
Xi
e Z
h
o
ng, F
e
ng Mi
ng, Ma C
han
g-ji
e. Inde
x St
rategies for
Embed
de
d-GIS Spatia
l Dat
a
Mana
geme
n
t.
Journ
a
l of Chi
n
a Univ
ersity of Geoscie
n
ces
. W
uhan. 20
06;
9: 653-6
5
8
[4]
Ye Li-
w
e
i
,
Xi
e Z
hon
g. Re
search
and I
m
pleme
n
tatio
n
on Dat
a
Mana
geme
n
t Optimizati
on o
f
Embed
de
d GIS Platform.
Micro computer i
n
formatio
n
. Beiji
ng. 200
9; 25:1
38-1
39
[5]
Ying
Li
u, Yant
ao Z
h
u, Yuro
n
g
Li,
Cha
o
N
i
.
T
he Embedd
e
d
Informati
on
Acquis
i
tion
S
ystem of F
o
rest
Resource.
Indo
nesi
an Jo
urna
l of Electrical En
gin
eeri
n
g
. 20
1
2
; 10(7): 18
43-
184
8
[6]
Yifeng
W
u
, Ho
ngch
ao W
a
ng.
Appl
icati
on
of GP
RS an
d G
I
S in Bo
iler
R
e
mote Mo
nitor
i
ng S
y
stem.
T
E
LKOMNIKA Indon
esi
an Jou
r
nal of Electric
al Eng
i
ne
eri
n
g
.
2012; 1
0
(8): 2
159-
216
8
Evaluation Warning : The document was created with Spire.PDF for Python.