Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
C
om
put
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
9
, No
.
6
,
Decem
ber
201
9
, p
p.
5295
~
5303
IS
S
N:
20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v
9
i
6
.
pp5295
-
53
03
5295
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
Senso
r dat
a ident
ification
based r
eagent
ca
bin
et
mana
gement s
ystem
Chan
gs
u
Kim
, Hon
gyou
l
Kim,
an
d
H
oeky
ung Jun
g
Depa
rtment
o
f
C
om
pute
r
Engi
n
e
eri
ng,
Pai
Chai
Univer
sit
y
,
S
ou
t
h Korea
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
A
pr
21
, 201
9
Re
vised
Ju
l
6
,
201
9
Accepte
d
J
ul
17
, 2
01
9
Rec
en
tly
,
a
r
ea
g
ent
c
abi
n
et
is
us
ed
in
a
l
abor
at
or
y
or
a
l
abor
a
tor
y
is
req
uir
ed
to
hav
e
a
s
y
s
te
m
ca
p
able
of
ide
n
t
if
y
ing
a
d
ange
ro
us
situa
tion
thro
ugh
sensor
dat
a
as
var
iou
s
sensors
are
uti
lized
.
Th
e
exi
sting
s
y
ste
m
ide
nti
fi
es
the
d
ange
rous
si
tua
ti
on
throug
h
var
ious
sensor
d
at
a
,
bu
t
th
ere
is
a
p
roble
m
tha
t
th
e
serve
r
p
erf
orm
s
al
l
the
o
per
ations
and
th
e
oper
a
ti
on
of
th
e
device
is
per
form
ed
m
an
ual
l
y
.
In
ord
er
to
solv
e
th
is
pr
oble
m
,
thi
s
pap
er
proposes
a
s
y
stem
th
at
c
a
n
ide
nti
f
y
the
d
a
nger
ous
situa
t
io
n
and
aut
o
m
atic
al
l
y
op
era
t
e
the
equi
pm
en
t
t
hrough
th
e
inter
nal
envi
ronm
en
t
da
ta
of
th
e
re
ag
ent
c
abi
ne
t.
Ide
nti
f
ic
a
ti
on
of
the
ha
za
rdous si
t
uat
ion is done
th
rough
the
m
ast
er
node
used
in
the
re
age
nt
c
abi
ne
t,
no
t
th
e
serve
r.
The
se
rv
er
c
an
con
ti
nuo
usl
y
upd
at
e
the
sens
or
data
t
hrough
the
m
ast
er
node
and
m
onit
or
th
e
re
al
-
t
i
m
e
stat
us
of
the
re
age
n
t
c
abinet
through
the
appl
i
ca
t
ion.
I
n
t
his
wa
y
,
i
t
is
ex
pec
t
ed
th
at
the
r
isk
situ
at
io
n
will
be
prom
ptly
addr
essed
b
y
ide
n
ti
f
y
ing
th
e
dang
ero
u
s
situa
ti
on
in
the r
ea
gen
t ca
bin
et an
d
oper
a
ti
ng
the
devi
c
e.
Ke
yw
or
d
s
:
Em
bed
ded Sys
tem
IoT
Re
agen
t
Rem
ote Con
tr
ol
Sensor
Copyright
©
201
9
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
:
Ho
e
kyun
g
J
un
g
,
Dep
a
rtem
ent o
f
C
om
pu
te
r
E
ng
i
neer
i
ng,
Pai
C
hai
Un
i
ve
rsity
,
1
55
-
40
Ba
e
j
ae
-
ro
,
Se
ogu
,
Dae
Jeo
n,
S
outh
Ko
rea
.
Em
a
il
:
hk
jung
@p
c
u.
ac
.kr
1.
INTROD
U
CTION
Re
centl
y,
rea
ge
nt
ca
bin
et
m
anag
em
ent
syst
e
m
s
have
bee
n
st
ud
ie
d
t
o
preve
nt
or
c
op
e
with
sa
fety
acci
den
ts
occ
urrin
g
in
la
bor
at
or
ie
s[
1
-
3
]
.
T
he
rea
gen
t
ca
bin
et
m
anag
e
m
ent
syst
e
m
m
easur
es
t
he
internal
env
i
ronm
ent
of
t
he
reag
e
nt
c
abinet
t
hro
ugh
ga
s
se
nsors
a
nd
e
nv
i
ronm
ental
sens
or
s
,
or
m
easur
es
to
xic
ga
ses
that
m
ay
occu
r
insi
de
t
he
rea
gen
t
ca
bin
et
[4
-
7]
.
Also,
t
he
s
ens
or
data
is
use
d
to
chec
k
wh
et
her
a
da
nge
r
ous
sit
uation
occur
s
in
the
reag
e
nt
cabi
net.
I
n
add
it
io
n,
t
he
a
pp
li
cat
io
n
or
e
m
bed
ded
boar
d
en
ables
real
-
tim
e
m
on
it
or
ing an
d co
ntr
ol of the
sta
te
o
f
t
he rea
gen
t ca
bine
t[
8
-
10
].
The
e
xisti
ng
re
agen
t
ca
bin
et
s
afety
m
anag
e
m
ent
syst
e
m
is
us
e
d
to
m
easu
re
the
i
n
te
r
nal
env
i
ronm
ent
of
th
e
rea
gen
t
thr
ough
en
vir
onm
ent
sensor,
gas
se
ns
or,
a
nd
ot
her
se
nsor
s
and
t
ran
sm
it
it
to
the
serv
e
r
an
d
confirm
s
the
oc
currence
of
a
da
nger
ous
sit
uation.
Wh
e
n
a
da
nger
ous
sit
uation
is
ide
nt
ifie
d,
t
he
m
a
nage
r
m
anu
al
ly
co
nt
ro
ls
the
devi
ce
thr
ough
t
he
tra
ns
m
it
ted
no
ti
ficat
io
n
m
essage
[11
-
13]
.
H
oweve
r
,
since
the
m
anag
er
m
anu
al
ly
m
an
ages
t
he
dang
erous
sit
uation,
a
sa
fety
acci
den
t
ca
n
occ
ur
beca
us
e
t
he
m
anage
r
do
e
s
not
ta
ke
prom
pt
act
ion
wh
e
n
the
m
anag
er
is
a
bs
e
nt.
I
n
ad
diti
on,
si
nce
t
he
ser
v
er
handles
al
l
the
w
ork,
m
anag
in
g
a
la
rg
e
num
ber
of
rea
gen
t
cabi
ne
ts
m
a
y
no
t
be
able
to
i
dent
ify
dange
rous
sit
uations
or
sen
d
no
ti
ficat
io
n
m
e
ssages q
uickly[
14
-
16
].
In
orde
r
t
o
s
ol
ve
t
his
pr
ob
le
m
,
we
pro
pose
a
syst
em
that
identifie
s
a
dangero
us
sit
uatio
n
in
a
rea
gen
t
cabinet
a
nd
a
uto
m
at
ic
ally
s
el
ect
s
a
dev
ic
e
that
can
c
ope
with
it
and
then
se
nds
a
n
ope
rati
on
co
m
m
and
autom
at
ic
ally.
To
ide
ntify
ha
zardo
us
sit
uations
in
the
rea
gen
t
ca
bin
et
,
a
sens
or
node
colle
ct
s
the
in
te
rn
a
l
env
i
ronm
ental
data
of
th
e
rea
gen
t
ca
bin
et
a
nd
a
m
ast
er
node
perform
s
dan
ge
r
ou
s
sit
uation
ide
ntific
at
ion
a
nd
dev
ic
e c
on
t
ro
l
com
m
and
ope
r
at
ion
s th
rou
gh
the co
ll
ect
ed
s
ens
or
data. The
m
ast
er n
ode iden
ti
fies a
dangero
us
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
6
,
Dece
m
ber
201
9
:
5295
-
5303
5296
sit
uation
th
r
ough
t
he
se
nsor
data
tran
sm
i
tt
e
d
eve
ry
sec
ond
f
ro
m
the
se
ns
or
n
ode.
Al
so
,
t
he
m
ast
er
node
sel
ect
s
the
de
vi
ce
that
ca
n
c
op
e
with
t
he
i
den
ti
fie
d
da
ngerous
sit
uatio
n
an
d
sen
d
the
op
e
rati
on
c
omm
and
autom
at
ic
ally.
Ther
e
fore,
if
a
dangero
us
sit
uation
occurs,
it
will
be
treat
ed
prom
ptly
and
safe
re
agent
m
anag
em
ent wi
ll
b
e ac
hie
ved
.
2.
SY
STE
M DESIGN
This
syst
em
c
ollec
ts
the
inte
rn
al
en
vir
onm
ental
data
of
t
he
rea
gen
t
ca
bi
net
a
nd
ca
n
a
uto
m
at
ic
all
y
op
e
rate
t
he
de
vice
in
case
of
a
da
ng
e
rous
s
it
uation.
T
he
pro
posed
syst
e
m
us
es
a
se
nsor
no
de
t
hat
c
ollec
ts
sens
or
data
a
nd
a
m
ast
er
no
de
that
ide
ntifie
s
a
da
nger
ous
sit
uation
th
rou
gh
c
ollec
te
d
se
ns
or
data
a
nd
sen
ds
op
e
rati
on
c
omm
and
s
to
the
de
vice.
T
he
se
nsor
no
de
m
eas
ur
es
the
inte
rnal
env
ir
onm
ental
data
of
the
r
eagen
t
cabinet
eve
ry
seco
nd
a
nd
c
onti
nuously
tra
ns
m
it
s
the
se
ns
or
data
to
t
h
e
m
ast
er
node
.
T
he
m
ast
er
no
de
identifie
s
t
he
da
ng
e
r
ou
s
sit
uat
ion
t
hro
ugh
t
he
receive
d
se
nso
r
da
ta
an
d
c
ontrols
t
he
dev
ic
e.
In
a
dd
it
io
n,
se
ns
or
data is t
ra
ns
m
itted to
t
he
se
r
ve
r
to
up
date sen
so
r
d
at
a i
n real
tim
e. Fig
ure
1 shows
the st
ruct
ur
e
of the
syst
e
m
.
Figure
1
.
Syst
em
a
rch
it
ect
ur
e
The
data
colle
ct
ion
la
ye
r
c
ollec
ts
data
a
bo
ut
se
nsors
a
nd
de
vices
co
nn
ect
ed
with
sen
so
r
nodes.
The
se
nsors
ut
il
iz
e
tem
per
at
ur
e,
hu
m
idit
y,
gas
(
Hyd
r
og
e
n
Su
lfi
de,
Nitro
gen
Di
ox
i
de,
a
nd
Su
l
fide
Diox
i
de
Gas),
fir
e
dete
ct
ion
,
sm
ok
e
detect
ion,
an
d
m
agn
et
ic
senso
rs
.
T
he
inst
r
um
ent
us
es
a
fan,
an
io
n
cl
us
te
r
,
and
a
la
m
p.
Inform
ation
ab
out
sens
ors
an
d
dev
ic
es
is
m
ea
su
re
d
a
nd
c
ollec
te
d
eve
ry
second
a
nd
t
hen
sent
t
o
the
m
ast
er
node
.
T
he
databas
e
of
the
d
at
a
st
or
a
ge
la
ye
r
up
dates
the
se
ns
or
data
by
co
ntinuo
us
ly
tra
nsm
itti
ng
the
c
ollec
te
d
s
ens
or
data
to
t
he
ser
ver
th
rou
gh
the
m
ast
er n
ode.
T
he
data
pr
ocessin
g
la
ye
r
perf
or
m
s
ta
sk
s
su
c
h
as iden
ti
ficat
io
n
of the
dange
r
ou
s
sit
uation p
erfor
m
ed
by th
e
m
ast
er n
ode
and
a
uto
m
at
ic
con
t
ro
l
of
the
dev
ic
e.
In
ad
diti
on,
t
he
operati
on
of
the
de
vice
a
nd
t
he
real
-
ti
m
e
sta
tus
of
t
he
reag
e
nt
ca
n
be
rem
otely
m
on
it
or
ed
thr
ough the
appli
cat
ion
.
Figure
2
s
how
s
the
op
e
rati
on
of
t
he
de
vice
accor
ding
to
te
m
per
at
ur
e.
The
da
nger
ous
sit
ua
ti
on
due
t
o
tem
per
at
ur
e
is
check
e
d
base
d
on
t
he
ave
ra
ge
value
of
th
e
m
i
nim
u
m
thresho
ld
an
d
the
m
axi
m
u
m
thresh
ol
d
set
by the m
anag
e
r.
First,
if
the
te
m
per
at
ur
e
of
t
he
reag
e
nt
ca
bi
net
is
hi
gh
e
r
t
han
the
ave
rage
valu
e,
disti
nguis
h
w
hethe
r
the
c
urren
t
te
m
per
at
ur
e
is
hi
gh
e
r
or
lo
wer
tha
n
t
he
m
axim
u
m
thresh
ol
d.
I
f
it
is
hi
gher
tha
n
the
m
a
xim
u
m
thres
ho
l
d,
ide
nt
ify
a
da
nger
ous
sit
uati
on
an
d
c
on
ti
nuously
chec
k
t
he
fire
detect
io
n
se
nsor
to
see
if
a
fire
has
occurre
d.
I
n
a
dd
it
io
n,
t
he
fa
n
s
peed
is
set
to
m
axi
m
u
m
(1.
0
m
/s)
a
nd
a
warnin
g
m
essage
is
se
nt
to
the
m
anag
er.
On
t
he
oth
e
r
ha
nd,
if
the
te
m
per
at
ur
e
is
lo
w
er
than
t
he
m
a
xim
u
m
tem
per
at
ur
e
val
ue,
set
the
fan
sp
ee
d
to
the
ba
sic
v
al
ue
(0.
53
m
/s).
Seco
nd,
i
f
t
he
te
m
per
at
ur
e
of
t
he
rea
gen
t
c
abinet
is
l
ow
e
r
t
han
the
a
ve
r
age
val
ue,
che
ck
w
heth
e
r
the
c
urren
t
te
m
per
at
ur
e
is
lowe
r
or
highe
r
tha
n
the
m
in
i
m
u
m
threshol
d.
I
f
it
is
lo
w
er
th
an
the
m
i
nim
u
m
thres
ho
l
d
valu
e,
ide
ntify
it
as
a
da
nger
ous
sit
uation
a
nd
s
et
the
fa
n
s
pee
d
to
m
ini
m
u
m
(0.1
m
/s)
and
sen
d
a
wa
rn
i
ng
m
es
sage
t
o
t
he
m
anag
e
r.
Als
o,
if
it
is
higher
tha
n
t
he
m
ini
m
um
thresh
ol
d
va
lue,
set
the
fa
n
sp
ee
d
to the basi
c
(
0.53 m
/s)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Sensor
data
id
enti
fi
cation b
ase
d
re
agent c
abinet
man
ag
e
m
ent system
(
Ch
angsu Ki
m
)
5297
Figure
3
s
ho
ws
the
ope
rati
on
of
the
de
vice
acc
ordi
ng
to
t
he
hu
m
idit
y.
Hu
m
i
dity
co
ntr
ols
the
instr
um
ent
thr
ough
a
total
o
f
five
sit
uatio
ns
. F
irst,
if
t
he
hu
m
idit
y
value
is
ab
ov
e 80%,
set
the
fa
n
s
pe
ed
to
m
axi
m
u
m
(1
.
00
m
/s)
an
d
se
nd
a
warni
ng
m
essage
to
th
e
m
anag
er
t
o
check
the
co
ndit
ion
of
t
he
r
eagen
t
cabinet.
I
f
the
hum
idit
y
is
m
or
e
than
40
%
an
d
le
ss
th
an
80%,
set
the
fan
at
the
m
axi
m
u
m
wind
s
pee
d
(1.00
m
/s)
so
that
the
m
oistur
e
inside
the
rea
gen
t
ca
bin
et
ca
n
esca
pe.
I
f
the
hu
m
idit
y
is
no
t
le
ss
than
30%
an
d
le
ss
than
40%,
the
fa
n
s
peed
is
set
to
a
basi
c
value
(0.53
m
/s)
becau
se
t
he
th
reshol
d
va
lue
is
not
exc
eede
d.
If
the
hum
idity
is
m
or
e
than
10%
a
nd
le
ss
than
30%,
the
fa
n
sp
ee
d
is
set
to
the
m
ini
m
u
m
(0
.1
m
/s).
If
t
he hu
m
idit
y
is l
ess tha
n 1
0%
, th
e
operati
on
of the
fan sto
ps
.
Figure
4
s
how
s
the
ope
rati
on
of
t
he
dev
ic
e
accor
ding
to
hy
drog
e
n
s
ulfide
.
If
the
hydro
gen
su
lfi
de
value
is
m
or
e
t
han
10ppm
,
the
de
vice
act
i
va
te
s
the
fa
n
a
nd
io
n
cl
us
te
r
an
d
l
ock
s
th
e
door
s
o
t
hat
the
r
eagent
cabinet
do
or
c
annot
be
open
ed
f
ro
m
the
outsi
de.
Als
o,
a
warnin
g
m
essa
ge
is
sent
to
th
e
m
anag
er
to
i
nfor
m
the
cu
rr
e
nt
sta
te
of
the
reage
nt
cabinet.
If
the
value
of
hydro
ge
n
su
lfi
de
dec
reases
t
o
le
ss
tha
n
10
pp
m
,
the d
e
vice st
ops the
operati
on
of the
fan an
d
i
on cluster
s a
nd unloc
ks
t
he
re
agen
t ca
bi
net door.
Figure
5
sho
ws
t
he
operat
ion
of
t
he
de
vice
acc
ordin
g
t
o
nitrog
e
n
di
ox
i
de.
Nitr
og
e
n
di
ox
i
de
determ
ines
the
ope
rati
on
of
t
he
de
vice
base
d
on
0.0
3
ppm
.
If
the
th
res
hold
is
ex
ceede
d,
the
dev
ic
e
operate
s
the
fa
n
an
d
io
n
cl
us
te
r
s
to
c
op
e
with
da
ngerous
sit
uatio
ns.
Als
o,
the
de
vice
sto
ps
th
e
dev
ic
es
t
hat
we
r
e
op
e
rated
if it
de
creased
to
le
s
s tha
n 0.03 p
pm
.
Figure
6
s
how
s
the
ope
rati
on
of
the
e
qu
i
pm
ent
acco
rd
i
ng
t
o
t
he
s
ulf
ur
di
ox
i
de
gas.
I
f
the
s
ulf
ur
dioxide
gas
i
s
m
or
e
than
0.02
ppm
,
it
identifie
s
th
e
dange
rous
sit
uation
a
nd
op
e
rates
the
dev
ic
es
.
Also
,
if
the
s
ul
fu
r
di
ox
i
d
e
ga
s
is
re
du
ce
d
t
o
le
ss
than
0.0
2
ppm
,
it
identifie
s
i
t
as
a
saf
e
sit
uation
a
nd
stop
s
the d
e
vices
Figure
2
.
De
vi
ce
o
per
at
io
n
ac
cordin
g
t
o
t
em
per
at
ur
e
Figure
3
.
De
vi
ce
o
per
at
io
n
ac
cordin
g
t
o
h
um
idit
y
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
6
,
Dece
m
ber
201
9
:
5295
-
5303
5298
Figure
4
.
De
vi
ce
o
per
at
io
n
ac
cordin
g
t
o
h
yd
roge
n
s
ulfid
e
Figure
5
.
De
vi
ce
o
per
at
io
n
ac
cordin
g
t
o
n
it
r
og
e
n
d
io
xid
e
Figure
6
.
De
vi
ce
o
per
at
io
n
ac
cordin
g
t
o
s
ulf
ur
d
i
ox
i
de
g
as
Table
1
sho
w
s
a
t
ran
sm
issio
n
f
ram
e
in
w
hich
colle
ct
ed
sen
sor
data
a
re
inte
gr
at
e
d
i
nto
one
a
nd
transm
itted
to
the
m
ast
er
no
de
.
T
he
te
m
per
at
ur
e
value
was
set
to
a
value
of
5
byte
s
beca
us
e
it
can
be
m
easur
ed
up
to
a
m
axi
m
u
m
of
three
di
gits
an
d
incl
ud
e
s u
p
to
the
fir
st
deci
m
al
place.
Als
o,
t
he
hu
m
idit
y
is
se
t
to
3
byte
s
bec
ause
it
is
m
easur
e
d
up
to
100%
.
I
n
the
gas,
hydro
ge
n
su
l
fide,
nitroge
n
di
ox
i
de,
a
nd
s
ulf
uro
us
aci
d
ga
s
eac
h
hav
e
a
value
of
4
byte
s.
Ma
gnet
ic
,
fi
re
detect
ion
,
an
d
sm
ok
e
de
te
ct
ion
se
ns
ors
are
set
to
1
byt
e
because
t
hey
i
den
ti
fy
'
0'
or
'
1'
value.
Ta
ble
2
sho
ws
t
he
de
vice
sta
tu
s
dat
a
tra
ns
m
issi
on
fr
am
e.
W
e
us
e
d
fa
n,
ion
cl
us
te
rs
a
nd
la
m
ps
,
re
sp
e
ct
ively
.
Fan
is
set
to
4byt
e
be
cause
it
s
how
s
up
t
o
t
he
se
cond
decim
al
place.
Ion
c
l
us
te
r
a
nd
elec
tric
la
m
p
are set to
1byt
e
to ide
ntify w
he
ther
t
hey oper
a
te
throug
h
'
0'
or '
1'
v
al
ue.
Table
1
.
Senso
r
data tran
sm
is
sion f
ram
e
Sen
so
r
Te
m
p
Hu
m
Gas
Magn
etic
Fla
m
e
S
m
o
k
e
Byte
5
b
y
te
3
b
y
te
1
2
b
y
te
1
b
y
te
1
b
y
te
1
b
y
te
Table
2
.
De
vice
sta
tus tra
ns
m
issi
on
fr
am
e
Dev
ice
Fan
Ion
Clu
ster
La
m
p
Byte
4
b
y
te
1
b
y
te
1
b
y
te
Figure
7
s
how
s
a
flow
c
har
t
of
the
syst
em
.
The
inte
rn
al
env
i
ronm
ent
data
of
the
rea
ge
nt
cabine
t
m
easur
ed
t
hro
ugh
the
se
nsor
s
are
i
ntegr
at
e
d
i
nto
on
e
data
tra
ns
fe
r
f
ram
e
an
d
the
n
tra
ns
fe
rr
e
d
to
the
m
ast
er
node.
T
he
m
ast
er
no
de
ver
i
fies
w
het
her
a
da
nger
ou
s
s
it
uation
occurs
thr
ough
tra
nsm
itted
sen
sor
data.
The
m
ast
er
no
de
ide
ntifie
s
da
ng
e
r
ou
s
sit
uat
ion
s
by
te
m
perat
ur
e,
hum
idity,
an
d
gas
i
f
a
da
ng
e
r
ou
s
sit
uation
occurs
a
nd
aut
om
atical
ly
op
erates
de
vices
t
hat
can
res
pond
to
t
hese
sit
ua
ti
on
s.
It
al
so
con
ti
nu
ousl
y
sends
sens
or
data
a
nd
dev
ic
e
sta
t
us
to
the
ser
ver
a
nd
up
dates
it
in
the
data
base.
Af
te
r
a
dange
rous
sit
uati
on
occurs
and
t
he
de
vice
'
s
op
erati
on
co
m
m
and
is
tran
sm
it
te
d,
the
dev
ic
e
co
ntinuo
usl
y
check
s
w
he
ther
the
dang
erous
sit
uation has
be
en
te
rm
inate
d.
If
the
d
a
nger
ous sit
uatio
n
is
term
inate
d,
it
stop
s
the
operati
on of t
he devic
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Sensor
data
id
enti
fi
cation b
ase
d
re
agent c
abinet
man
ag
e
m
ent system
(
Ch
angsu Ki
m
)
5299
Figure
7
.
Syst
em
flow
ch
a
rt
3.
SY
STE
M
I
M
PLE
MENT
A
TION
This
c
hap
te
r
deals
with
the
i
m
ple
m
entat
i
on
of
a
syst
em
that
identifi
es
da
nger
ous
sit
uations
i
n
reag
e
nt
ca
bine
ts
an
d
aut
om
at
ic
ally
op
e
rates
the
de
vi
ce.
Ta
ble
3
s
hows
the
s
yst
e
m
i
m
ple
mentat
ion
env
i
ronm
ent
.
I
n
the
pro
pose
d
syst
e
m
,
a
co
ntro
l
boar
d
base
d
on
W
i
ndows
CE
6.0
was
use
d
to
che
ck
t
he
real
tim
e
sta
tus
of
the
reag
e
nt
c
abinet.
Gala
xy
S6
is
us
ed
f
or
rem
ote
m
on
it
or
i
ng
a
nd
c
ontrol
in
ap
plica
ti
on
s
.
On
t
he
PC,
I
bu
il
t
the
datab
ase
thr
ough
A
PM
Setu
p.
T
o
m
easur
e
the
inter
nal
en
vir
onm
ent
of
the
r
eagen
t
cabinet,
the
se
ns
ors
us
e
d
te
m
per
at
ur
e,
hu
m
idit
y,
m
ulti
-
channel
gas,
f
ire
detect
io
n,
sm
ok
e
detect
ion,
a
nd
m
agn
et
ic
senso
rs
.
Figure
8
s
how
s
the
m
ai
n
page
of
the
syst
e
m
i
m
ple
m
ented
thr
ough
th
e
W
i
ndows
CE
boa
rd.
Y
ou
c
a
n
m
on
i
tor
the
re
al
-
tim
e
tem
per
at
ur
e
a
nd
hum
idit
y
of
the
re
agen
t
ca
bin
et
,
the
val
ues
f
or
the
gas
sens
or
s,
an
d
check
t
he
sta
tus
of
the
fan,
ion
cl
us
te
rs
an
d
do
or
s
.
The
s
ta
tus
of
t
he
de
vice
ind
ic
at
es
that
the
de
vice
is
in
op
e
rati
on
wh
e
n
the
T
oggle B
utton i
s e
nab
le
d
a
nd the
de
vic
e is in
op
e
rab
le
if it
is d
isa
bled
Figure
9
(a
)
sho
ws
the
m
ai
n
pa
ge
of
the
A
ndr
oid
a
pp
li
cat
io
n.
The
m
ai
n
page
consi
sts
of
te
m
per
at
ur
e,
hu
m
idit
y,
gas,
an
d
i
ns
tr
um
e
nt
sta
tus
.
Tem
per
at
ur
e
an
d
hu
m
idit
y
ind
ic
at
e
the
te
m
per
at
ur
e
a
nd
hum
idit
y
m
easur
ed
i
n
t
he
insi
de
of
t
he
rea
ge
nt
in
rea
l
tim
e,
an
d
t
he
ga
s
m
on
it
ors
hydro
ge
n
s
ulfi
de,
nitr
ogen
dio
xi
de,
and
su
l
fur
dioxide.
The
de
vi
ce
ind
ic
at
es
th
e
curre
nt
oper
at
ing
sta
te
.
Also
,
it
c
on
sist
s
of
c
ontr
ol
pa
ge
f
or
rem
ote
con
t
ro
l
of
the
rea
ge
nt
cabi
net,
an
d
a
vi
e
w
pa
ge
w
he
re
y
ou
ca
n
vi
ew
i
nfor
m
at
ion
a
bout
the
da
ng
e
r
ou
s
sit
uation
i
n
the
r
ea
gen
t ca
bin
e
t.
Figure
9
(
b)
s
hows
the
view
pag
e
of
the
Android
a
ppli
cat
i
on.
T
he
vie
w
pag
e
al
l
ow
s
y
ou
to
m
on
it
or
m
essages
se
nt
w
hen
a
dang
erous
sit
uation
occ
ur
s
in
t
he
rea
gen
t
ca
bi
ne
t.
A
m
essage
ab
ou
t
a
da
ng
erous
sit
uation
i
nd
ic
at
es
inf
or
m
at
i
on
ab
out
the
ye
ar,
m
on
th,
day,
hour,
a
nd
da
ng
e
r
ou
s
sit
uation
that
oc
curred
.
Me
ssages
a
re a
lso a
uto
m
at
ic
a
l
ly
d
el
et
ed
afte
r
thr
ee
m
on
ths
of sto
rag
e
.
Figure
9
(c
)
shows
the
co
ntr
ol
pag
e
of
t
he
A
ndr
o
id
a
ppli
cat
ion
.
T
he
m
anag
er
can
set
the
tem
per
at
ur
e
and hum
idit
y a
nd r
em
otely
co
ntr
ol the s
ta
tu
s
of
t
he device
. Yo
u
ca
n
al
s
o
s
et
the d
e
sired
wind s
peed w
he
n
yo
u
tur
n
on
t
he
fa
n.
A
fter
the
s
et
ti
ng
is
c
om
plete
d,
y
ou
cl
ic
k
t
he
Set
butt
on
to
c
ontrol
the
reag
e
n
t
ca
bin
et
accor
ding t
o
th
e v
al
ue
s set
by
the m
anag
er
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
6
,
Dece
m
ber
201
9
:
5295
-
5303
5300
Table
3
.
De
velop
m
ent
e
nvir
onm
ents f
r
om
m
achine
Ty
p
e
Co
m
p
o
sitio
n
Too
ls
Edit Plu
s, And
roid
Stud
io
,
APM
Setu
p
7
,
W
in
d
o
ws CE 6
.0, Ar
d
u
in
o
,
FCM
PC
W
in
d
o
w
7
Enterp
r
ise
K
6
4
b
it
,
Intel
i5
-
4460
,
8
GB
RAM
Sen
so
r
Te
m
p
e
rature,
Hu
m
id
ity
,
Multich
an
n
el Gas,
F
la
m
e
,
S
m
o
k
e,
M
ag
n
etic
Dev
ice
Ardu
in
o
UNO
,
W
i
n
d
o
w
CE
,
Galax
y
S6
Figure
8
.
A
ppli
cat
ion
m
ai
n
p
a
ge
(a)
(b)
(c)
Figure
9
.
A
ndr
oid
a
ppli
cat
ion
, (
a
)
Ma
in
p
a
ge
, (b)
Vie
w page
, (
c
)
C
on
tr
ol
pa
ge
4.
REVIEW
This
sect
io
n
de
al
s
with
the
com
par
ison
be
tween
t
he
exi
sti
ng
syst
em
a
nd
t
he
propos
ed
syst
em
.
The
data
set
f
or
the
e
xperim
ent
is
us
ed
i
n
the
existi
ng
syst
e
m
and
the
pro
po
se
d
syst
e
m
in
the
sa
m
e
way.
The
e
xp
e
rim
e
nt
co
nf
i
rm
s
ho
w
m
uch
tim
e
it
ta
kes
f
or
t
he
dev
ic
e
t
o
operate
wh
e
n
t
he
value
of
hy
droge
n
su
lfide
is
out
of
t
he
t
hr
es
hold.
T
he
num
ber
of
e
xpe
rim
ents
is
20
eac
h.
Tabl
e
4
s
hows
the
existi
ng
syst
e
m
and
the d
at
a
set
of t
he pr
opos
e
d sy
stem
.
Table
4
.
Syst
em
exp
erim
ent
dataset
Dataset
Descripti
o
n
Hy
d
rog
en
Sulf
id
e
1
to 2
0
pp
m
Co
u
n
t
The n
u
m
b
e
r
o
f
exp
eri
m
en
ts
was 2
0
As
a
res
ult
of
the
ex
per
im
ent
in
Fig
ur
e
10
,
wh
e
n
the
e
xisti
ng
syst
em
was
app
li
e
d,
the
dange
rous
sit
uation
occ
urre
d
a
nd
the
tim
e
req
uire
d
for
t
he
dev
i
ce
to
op
e
rate
was
a
bo
ut
4
seco
nds
on
aver
a
ge.
These
res
ults
m
ay
be
m
or
e
tim
e
con
su
m
ing
due
t
o
the
si
tuati
on
of
t
he
m
anag
er
,
bec
a
us
e
it
can
be
di
ff
ere
nt
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Sensor
data
id
enti
fi
cation b
ase
d
re
agent c
abinet
man
ag
e
m
ent system
(
Ch
angsu Ki
m
)
5301
dep
e
ndin
g
on
how
qu
ic
kly
the
m
anag
er
re
sp
on
ds
to
t
he
dange
rous
sit
ua
ti
on
.
T
he
pro
po
s
ed
syst
em
t
ook
a
n
aver
a
ge
of
a
bout
2.2
seco
nd
s
unti
l
a
da
ng
e
r
ou
s
sit
ua
ti
on
occurre
d
an
d
t
he
de
vice
becam
e
act
ive.
Un
li
ke
the
e
xisti
n
g
syst
em
,
t
he
syst
em
autom
atical
ly
sel
e
ct
s
an
d
operat
es
the
de
vice
in
t
he
syst
em
withou
t
m
anu
al
c
on
t
rol
by
t
he
m
anag
er
.
As
a
res
ul
t,
it
is
c
onfir
m
ed
that
it
is
po
s
sible
t
o
c
ope
m
or
e
quic
kl
y
tha
n
the m
anu
al
con
trol
by the m
anag
er
li
ke
the
exi
sti
ng
syst
em
.
Figure
10
.
E
xp
erim
ental
resu
lt
s o
f
syst
em
w
i
thout da
nger
pr
iority
appli
ed
5.
CONCL
US
I
O
N
The
e
xisti
ng
s
yst
e
m
transm
its
the
i
ntern
al
env
i
ronm
ent
da
ta
of
the
rea
gen
t
ca
bin
et
t
o
the
ser
ve
r,
identifie
s
the
da
ng
e
r
ou
s
sit
uat
ion
,
a
nd
se
nds
a
noti
ficat
ion
m
essage
to
t
he
m
anag
er.
T
he
m
anag
er
ca
n
i
de
ntify
the
da
ng
e
rous
sit
uation
th
rou
gh
the
noti
ficat
ion
m
essage
a
nd
op
e
rate
t
he
dev
ic
e
th
rou
gh
the
c
on
tr
ol
bo
ard
or
app
li
cat
io
n
t
o
cop
e
with
t
his
sit
uatio
n
[17
-
20]
.
As
de
scrib
ed
a
bove
,
w
he
n
t
he
m
anag
e
r
m
anu
al
ly
co
nt
ro
ls
the d
e
vices,
th
ere
is
a
pr
ob
le
m
that
it
is
di
ffi
cult
to
quic
kly
co
pe
w
it
h
a
d
a
ng
e
r
ou
s
sit
uati
on w
he
n
the
m
anag
e
r
is
no
t
present.
Also
,
since
the
ser
ver
pe
rform
s
all
the
oper
at
ion
s
t
hat
occ
ur
i
n
a
plural
it
y
of
rea
gen
t
ca
bin
et
s
,
the
ser
ve
r op
e
r
at
ion
s
pee
d
m
a
y becom
e slowe
r
as t
he nu
m
ber
of
rea
gen
t c
abinets i
ncr
eas
es
[21
-
25]
.
In
ord
er
t
o
s
ol
ve
t
hese
pro
blem
s,
this
pap
e
r
pro
po
ses
a
syst
e
m
that
identif
ie
s
da
ng
e
r
ou
s
sit
uations
i
n
a
reage
nt
ca
bine
t
and
aut
om
atical
ly
op
erates
the
de
vice.
In
the
rea
ge
nt
ca
bin
et
,
se
nsor
dat
a
an
d
dev
ic
e
st
at
us
are
c
ollec
te
d
t
hro
ugh
t
he
se
ns
or
no
de
a
nd
tra
ns
m
i
tt
ed
to
t
he
m
ast
er
n
ode.
T
he
m
a
ste
r
node
ide
nt
ifie
s
the
da
ng
e
r
ou
s
sit
uation
th
rough
the
tra
nsm
itted
sens
or
data
an
d
the
n
op
e
rates
the
dev
ic
e
a
uto
m
a
ti
cal
ly
.
In
a
dd
it
io
n,
by
c
on
ti
nu
ou
sly
transm
it
t
ing
se
ns
or
data
a
nd
dev
ic
e
sta
tus
t
o
t
he
ser
ve
r,
the
real
-
ti
m
e
sta
tus
of
the
rea
ge
nt
ca
bin
et
ca
n
be
c
onfirm
ed
f
ro
m
the
a
ppli
cat
ion
or
co
ntr
ol
boar
d.
A
ppli
cat
i
on
s
ca
n
m
on
it
or
an
d
con
t
ro
l
the
rea
gen
t
ca
bin
et
st
at
us
in
r
eal
ti
m
e
throu
gh
rem
ote
m
on
it
or
in
g
and
c
ontrol.
F
uture
resea
rch
will
be
able
to
reduce
the
tim
e
consum
ed
in
ide
nt
ific
at
ion
of
da
ng
e
r
ou
s
sit
ua
t
ion
or
de
vice
operati
on
c
om
m
and
by im
ple
m
entin
g rea
ge
nt cab
i
net m
od
ule t
ha
t i
ntegr
at
es
con
trol
bo
a
rd, se
nsor
no
de
a
nd m
ast
er no
de
int
o on
e
.
ACKN
OWLE
DGE
MENTS
This
resea
rch
was
sup
ported
by
T
he
Lea
di
ng
H
um
an
R
eso
ur
ce
T
raini
ng
P
r
ogram
of
Re
gional
Ne
o
industry
th
rou
gh
t
he
N
at
ion
a
l
Re
search
F
oundat
io
n
of
K
or
ea
(N
R
F)
fun
ded
by
the
Mi
nistry
of
Scie
nc
e,
ICT
and
fu
t
ur
e
Pla
nn
i
ng(
No.
2016H
1D5
A
1911
091).
This
work
was
sup
ported
by
t
he
rese
arch
gr
a
nt
of
Paic
hai
Un
i
ver
sit
y i
n 2
019
.
REFERE
NCE
S
[1]
J.
O.
W
estga
rd
,
"U
seful
m
ea
s
ure
s
and
m
odels
for
anal
y
tica
l
qua
li
t
y
m
ana
g
ement
in
m
edi
c
al
la
bor
at
or
ie
s
,
"
Cli
nic
a
l
Ch
emist
ry
and
Labor
ato
ry
Me
di
ci
n
e
(
CCL
M)
,
vol.
54
,
no
.
2,
pp.
223
-
233
,
2016
.
[2]
H.
I.
Jung
and
K.
Y.
Chung,
"K
nowledge
-
base
d
dietar
y
nu
tri
ti
on
r
ec
om
m
enda
ti
on
for
ob
e
se
m
ana
gement
,
"
Information
Tec
hnology
and
Ma
nageme
nt
,
vol
.
1
7,
no
.
1
,
pp
.
29
-
42
,
2016
.
[3]
C.
I.
Cu
cor
anu,
"La
bora
tor
y
Inf
orm
at
ion
S
y
ste
m
s
Mana
gement
and
Opera
t
ions,
"
Cli
nic
s
in
lab
oratory
medi
ci
n
e
,
vol.
36
,
no
.
1
,
,
p
p.
51
-
56
,
2016
.
[4]
R.
Roos,
K.
Br
own
and
T.
Tor
abi
,
"LUS
–
a
ta
b
le
t
-
b
ase
d
NF
C
s
y
stem
to
fa
ci
l
itate
insta
n
t
d
ata
base
a
rch
iv
al
of
la
bora
tor
y
assessment"
,
Australas
ian
Journal
of
E
ngine
ering
Education
,
vol. 21
,
n
o.
2
,
pp
.
74
-
80
,
2016
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
6
,
Dece
m
ber
201
9
:
5295
-
5303
5302
[5]
A.
Mera
be
t,
K
.
T.
Ahm
ed,
H.
I
bra
him,
R.
Begu
ena
ne
and
A.
M.
Ghias,
"
Ene
rg
y
m
ana
gement
an
d
cont
rol
s
y
st
em
for
la
bor
at
or
y
s
ca
l
e
m
ic
rogrid
b
ase
d
wind
-
PV
-
bat
t
er
y
,
"
I
EE
E
Tr
ansacti
ons
on
S
ustainabl
e
Ene
r
gy
,
vo
l.
8
,
no
.
1,
pp.
145
-
154
,
20
17
.
[6]
H.
W
.
Kim
,
E.
G.
Lee
and
D.
W
.
Kw
on,
"Che
m
ic
al
la
bora
tor
y
safe
t
y
m
an
agem
ent
service
usi
ng
IoT
sensors
a
nd
open
AP
Is
,
"
in
I
nformation
and
Comm
unic
ati
ons (
ICIC)
,
2017
Int
ern
ational
Confe
ren
ce on, 2017
.
[7]
K.
Cao,
H.
L
ee
a
nd
J.
Jung
,
"
Tas
k
Mana
gement
S
y
stem
Acc
ord
ing
to
Chang
es
in
the
Situation
Based
on
IoT
,
"
Journal
of
I
nfor
mation
Proc
essing
Syste
m
,
vol
.
1
3
no.
6
,
pp
.
1
459
-
1
466
,
201
7
.
[8]
K.
C
ao,
I
.
K
ang,
H.
Choi
and
H.
Jung,
"
Re
age
nt
Cabi
net
Man
agem
ent
S
y
s
te
m
Us
ing
Dang
er
Prio
rity
”
,
Journal
of
Information
and
Comm
unic
ati
on
Conve
rgenc
e En
gine
ering
,
vol
.
1
5
,
no
.
4
,
pp
.
2
27
-
23
1
,
201
7
.
[9]
Y.
Evc
hin
a,
J.
Puttone
n
,
A.
Dvo
c
y
anc
h
ikova
and
J.
L.
M
.
L
astra
,
“
Conte
xt
-
awa
r
e
knowledge
-
base
d
m
iddl
ewa
re
fo
r
sele
c
ti
ve
infor
m
at
ion
d
el
iv
er
y
in
d
ata
-
intensi
ve
m
onit
or
ing
s
y
stem
,”
Engi
n
ee
ring
Applications
of
Art
if
i
cial
Inte
lligen
ce
,
vol
.
43,
pp.
111
-
126
,
2015
.
[10]
N.
J.
Cho
and
Y.
G.
Ji
,
“
Indoor
I
oT
Monitor
ing
S
y
stem
b
ase
d
on
Visible
Li
ght
Co
m
m
unic
at
ion
usi
ng
Sm
ar
t
Phone
,
”
Journal
of
the K
orean
Instit
u
te
o
f
Tel
emati
cs
and
Elec
troni
cs
,
vo
l
.
54
,
no
.
4
,
pp.
3
5
-
43
,
2017
.
[11]
H.
J.
Lee,
H.
W
.
Choi
and
H.
K.
Jung,
“
Rea
gen
t
Cabi
net
Ha
za
rd
Situa
ti
on
Ide
n
ti
f
ic
a
ti
on
S
y
stem
Util
izing
Mul
ti
p
l
e
Sensor
Data
,
”
J
ournal
of
the
K
orea
Instit
u
te
o
f
Information
an
d
Comm
unic
ation
Engi
n
ee
ring
,
vol.
22
,
no
.
1
,
pp.
63
-
68
,
2018
.
[12]
S.
J.
L
ee
and
J
.
G.
Hong,
“
Mea
suring
th
e
Gree
nhouse
Gas
Emi
ss
ion
Reduction
and
Man
age
m
e
nt
S
y
s
te
m
Us
in
g
Blue
toot
h
Senso
r
Node
,
”
Journa
l
of
In
formation
and
Comm
unicatio
n
Con
ve
rge
nce
Engi
n
ee
ring
,
vo
l.
17
,
no.
5
,
pp.
1095
-
1100
,
2014
.
[13]
Y.
J An,
D.
H.
K
im,
J.
H.
Lee an
d
B.
J.
Lee, “
Indoor
Envi
ronm
ent Cont
rol
S
y
stem
Util
izing
th
e
In
t
ern
et of
Th
ings
,
”
Journal
of
the K
IECS
,
vo
l. 12, n
o.
4
,
pp
.
645
-
65
0
,
2017
.
[14]
W
.
Y.
Le
e
,
H.
M.
Ko,
J.
H.
Yu
and
B.
Sim
,
“
An
Im
ple
m
ent
at
io
n
of
Sm
art
Dorm
it
or
y
S
y
s
te
m
base
d
on
Inte
rn
et
of
Thi
ngs
,
”
Journa
l
of
Korean
Inst
i
tut
e
of
In
telli
g
en
t
Syst
ems
”,
vo
l.
26,
no
.
4
,
pp
.
29
5
-
300
,
2016
.
[15]
M
.
J.
Song,
“
A
Stud
y
on
Busine
ss
T
y
pes
of
IoT
-
base
d
Sm
art
ho
m
e:
Based
on
th
e
Th
eor
y
of
Pl
atform
T
y
polog
y
,
”
The
Journal
o
f t
he
Insti
tut
e
of
In
te
rnet
,
Broadcasti
ng
and
Comm
unic
ati
on
,
vo
l. 16, no. 2, pp. 27
-
40
,
2016
.
[16]
N.
J.
Cho
and
Y,
G.
Ji,
“
Anal
y
sis
of
Safe
t
y
Mana
gement
Co
ndit
ion
&
A
cc
id
ent
T
y
p
e
in
Do
m
esti
c
and
Fore
ign
La
bora
to
y
,
”
Jou
rnal
of the
Engo
nomics
Societ
y
o
f
Korea
,
vol
.
35
,
no.
2
,
pp
.
97
-
10
9
,
2016
.
[17]
X.
L
ie,
J.
Li
an
d
X.
Li,
“
Stud
y
of
D
y
n
amic
Ri
sk
Mana
gement
S
y
st
em
for
Fla
m
m
abl
e
and
Ex
posive
Dang
eou
s
Chemica
ls Stor
a
ge
Are
a
,
”
Journ
al
of
Los
s P
rev
e
nti
on
in
th
e Proc
ess Ind
ustries
,
v
ol.
49
,
pp
.
983
-
9
88
,
2017
.
[18]
Y.
H.
Cho,
“
Smart
W
indows
and
Doors
Plat
form
for
Providing
Optimize
d
Inne
r
Envi
ronm
ent
”
,
Journal
o
f
Kore
an
Inti
tute
o
f
In
te
l
ligent
S
yste
ms
,
vo
l.
25
,
no
.
5
,
pp
.
4
64
-
469
,
2015
.
[19]
J.
H.
Kim
,
“
Dev
el
opm
ent
of
U
-
Buil
ding
Fire
Sa
fety
M
ana
gemen
t
S
y
s
te
m
,
”
F
ire
Sci
en
ce
and
Eng
ine
ering
,
vo
l.
30
,
no.
1
,
pp
.
74
-
80
,
2016
.
[20]
C.
S.
O
h,
M.
S
.
Seo,
J.
H.
Le
e
,
S.
H.
Kim
,
Y.
D
.
Kim
and
H.
J.
Park,
“
Indoor
A
ir
Qualit
y
Monit
o
ring
S
y
stem
in
the
IoT
Envi
ro
n
m
ent
,
”
The
Journal
of
Korean
In
stit
ute
of
Comm
unic
ati
on
and
In
formation
Scien
ce
,
vol
.
40,
no
.
5
,
pp.
886
-
891
,
20
15
.
[21]
J
.
W
.
P
ark
,
D
.
S.
Kim
and
N.
K.
Joo,
“
Indoo
r
Envi
ronm
ent
Monitori
ng
and
Control
l
ing
S
y
stem
Design
an
d
Im
ple
m
ent
at
ion
base
d
on
In
te
r
e
nt
of
Thi
ngs
,
”
J
ournal
of
th
e
K
orea
Instit
ut
e
of
Information
an
d
Comm
unic
ati
o
n
Engi
ne
ering
,
vol
.
20
,
no
.
2
,
pp
.
3
67
-
374
,
2017
.
[22]
Y.
S.
Moon,
J.
W
.
Jung,
S.
P.
Choi,
T
.
H.
Ki
m
,
B.
H.
Le
e
,
J.
J.
Kim
and
H.
L.
Choi
,
“
Rea
l
-
Ti
m
e
Man
age
m
ent
S
y
stem
of
Re
ef
er
Con
ta
in
er
ba
sed
on
IoT
,
”
Jo
urnal
of
the
Ko
rea
Insti
tu
te
of
Information
and
Comm
unic
ati
o
n
Engi
ne
ering
,
vol
19,
nio. 9, pp. 2
093
-
2099
,
2015
.
[23]
C
.
C.
Shih
,
R.
S
.
hong
and
S.
K
.
Lee,
“
Inve
st
igation
of
L
ab
Fire
Preve
nti
on
Ma
nage
m
ent
S
y
s
tem
of
Com
bini
ng
Root
C
ause
Anal
y
s
is
and
Anal
y
t
ic
Hie
rar
ch
y
Proce
ss
with
E
ven
t
Tr
ee
Anal
y
sis
,
”
Mac
hema
ti
cal
Proble
ms
i
n
Engi
ne
ering
201
6.
[24]
S
.
Bruckne
r
,
J.
L.
Mar
enda
z
an
d
T.
Me
y
e
r,
“
Us
ing
Ver
y
Tox
ic
or
Espec
i
al
l
y
H
a
za
rdous
Chemic
al
Subs
ta
nc
es
in
a
Resea
rch
and
T
e
ac
hing
Instit
u
ti
o
n
,
”
Sa
fety
S
cienc
e
,
vo
l. 88, pp. 1
-
15
,
2016
.
[25]
J.
B
ack
,
“
Secur
e
Pre
-
aut
hen
ti
c
ati
on
Schemes
for
Fast
Handoff
in
Prox
y
Mobile
IPv6
,
”
Journal
o
f
I
nformation
and
Comm
unic
ati
on
Conve
rgenc
e En
gee
ring
,
vol
.
14
,
no.
2,
pp.
89
-
96
,
2016
.
BIOGR
AP
H
I
ES
OF
A
UTH
ORS
Ch
ang
su
Ki
m
,
h
e
r
ecei
ved
h
is
B.
S.,
M.S.
,
and
Ph.D.
d
egr
ee
s
fr
om
the
Depa
r
tment
of
Com
pute
r
Engi
ne
eri
ng
of
Paic
ha
i
Univer
si
t
y
,
Kore
a,
in
19
96,
1998,
and
2
002,
respe
ct
iv
ely
.
From
2005
to
2012,
h
e
worke
d
for
the
Depa
r
t
m
ent
of
In
te
rn
et
at
Chungwoon
Univer
sit
y
as
a
profe
ss
or.
Sinc
e
2013,
he
has
wo
rke
d
in
th
e
De
p
art
m
ent
of
Com
pute
r
Enginee
r
in
g
a
t
Paic
h
ai
Uni
ver
sit
y
,
wher
e
he
now
works
as
a
pro
fessor.
His
cur
ren
t
res
ea
rch
intere
sts
i
ncl
ude
m
ult
ime
dia
do
cument
arc
hi
te
c
ture m
odel
ing
,
web
2.
0
,
a
nd
the sem
ant
i
c web.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Sensor
data
id
enti
fi
cation b
ase
d
re
agent c
abinet
man
ag
e
m
ent system
(
Ch
angsu Ki
m
)
5303
Hongy
oul
Ki
m
,
h
e
r
ec
e
ive
d
th
e
B.
S.
d
egr
e
e
in
2
015
from
th
e
De
par
tment
of
Soci
al
W
el
far
e
o
f
Honam
Univer
si
t
y
and
M.S.
d
eg
ree
in
2019
f
ro
m
the
Depa
rtme
nt
of
Com
pute
r
Engi
ne
eri
ng
of
Paic
ha
i
Univer
si
t
y
,
He
is
cur
r
entl
y
a
Doc
torate
c
ourse
in
Dep
artm
ent
of
Com
pute
r
Engi
n
ee
r
ing
of
Paic
ha
i
Univ
ersity
.
Kore
a.
Si
nce
1999,
h
e
ha
s
worked
at
Kw
ang
ji
n
Co
.
,
L
td.
where
he
now
works
as
a
rep
re
senta
ti
v
e
Dir
ec
t
or.
His
cur
ren
t
r
ese
arc
h
intere
sts
inc
lud
e
m
ult
imedia
do
cument
arc
hi
te
c
ture m
odel
ing
,
Io
T
and
A
I.
Hoe
k
y
un
g
Jun
g
,
h
e
re
ce
iv
ed
the
M.S.
degr
e
e
in
1987
and
Ph.
D.
degr
ee
in
1993
from
the
Dep
art
m
ent
of
Com
pute
r
En
gine
er
ing
of
Kw
angwoon
Univer
sit
y
,
Kore
a. From
1994
to
1995
,
he
work
ed
fo
r
E
TRI
as
a
rese
archer.
Sinc
e
199
4
,
he
h
as
worked
i
n
th
e
Depa
r
tment
of
Com
pute
r
Engi
ne
eri
ng
at
Paic
ha
i
Unive
rsit
y
,
wher
e
h
e
no
w
works
as
a
p
rofe
ss
or.
His
cu
rre
nt
r
ese
ar
ch
int
er
ests
inc
lud
e
m
ult
imedia
document
ar
ch
it
e
ct
ure
m
ode
ling,
informat
io
n
proc
essing,
informati
on
r
et
r
i
eva
l
,
m
ac
h
ine l
e
arn
ing, bi
gd
at
a
,
and
IoT
.
Evaluation Warning : The document was created with Spire.PDF for Python.