Indonesi
an
Journa
l
of El
ect
ri
cal Engineer
ing
an
d
Comp
ut
er
Scie
nce
Vo
l.
23
,
No.
2
,
A
ugus
t
20
21
,
pp.
733
~
739
IS
S
N: 25
02
-
4752, DO
I: 10
.11
591/ijeecs
.v
23
.i
2
.
pp
733
-
739
733
Journ
al h
om
e
page
:
http:
//
ij
eecs.i
aesc
or
e.c
om
IoT for sm
ar
t
hom
e syste
m
Puji
Catur
Sis
w
iprapt
ini
,
R
os
ida
Nur
Az
iz
a
,
Iri
an
s
yah
Sa
n
gadj
i,
In
d
ri
an
t
o, Riki
R
uli
A
. Si
reg
ar
,
Grace
Sond
akh
Depa
rtment
o
f
I
nform
at
ic
s,
Insti
tut
Te
knolog
i
P
LN,
Indon
esia
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Ja
n
13
, 2
02
1
Re
vised Ju
n
1,
2021
Accepte
d
J
un
5
, 2
021
Thi
s
pap
er
ex
a
m
ine
s
the
int
egr
at
ion
o
f
sm
art
h
om
e
and
solar
p
ane
l
s
y
st
em
tha
t
is
cont
rol
le
d
and
m
onit
ore
d
using
IoT
(
inter
net
of
th
ings
).
T
o
ena
bl
e
th
e
sm
art
hom
e
sy
st
em t
o
m
onit
or
the
activit
y
withi
n
the
house
and act
accordi
n
g
to
the
cur
ren
t
c
ondit
ions,
it
is
e
quippe
d
with
seve
ral
sensors
,
actua
tors,
an
d
sm
art
appl
ia
n
ces
.
All
of
th
ese
device
s
hav
e
to
b
e
conn
ec
t
ed
to
a
comm
unic
at
io
n
net
work,
so
they
ca
n
comm
unic
a
te
and
provid
e
services
for
the
sm
art
hom
e’
s
inha
bitants.
Th
e
sm
art
hom
e
s
y
stem
was
first
in
troduc
ed
to
pr
ovide
comfort
and
conve
n
ie
nc
e
,
but
later
i
t
should
al
so
addr
ess
m
an
y
othe
r
thi
ngs,
e.
g
.
th
e
i
m
porta
nce
o
f
th
e
eff
ic
i
en
t
use
o
f
ene
rg
y
or
elec
tri
cit
y
and
h
y
brid
use
of
e
ner
g
y
sourc
es.
A
solar
pane
l
is
adde
d
to
the
sm
art
hom
e
prototy
pe
and
i
ts
addi
ti
on
is
studie
d
.
Adapti
v
e
li
n
ea
r
neur
al
net
work
is
implemente
d
in
the
prototy
p
e
as
an
al
gorit
hm
for
pre
dic
t
ing
deci
sions
base
d
o
n
the
cur
ren
t
co
ndit
ions.
Th
e
co
nstruct
ion
of
the
proposed
int
egr
at
ed
s
y
st
em
is
ca
rrie
d
ou
t
t
hrough
seve
ral
proc
edur
es
,
i.e.
the
implement
at
ion
of
the
ada
pt
ive
li
ne
ar
neur
al
net
work
(
AD
ALINE
)
as
the
n
eur
al
net
w
ork
m
et
hod
,
the
design
of
the
proto
t
y
p
e,
and
t
he
te
sting
proc
es
s.
Thi
s
prototy
p
e
int
egr
ates
func
ti
on
al
i
ti
es
of
seve
ral
ho
usehold
appl
i
a
nce
s
int
o
one
appl
icat
ion
cont
rolled
b
y
an
Android
-
base
d
f
ramework.
Ke
yw
or
ds:
ADAL
IN
E
In
te
r
net
of thin
gs
Sensor
Sm
art h
om
e
So
la
r
p
a
nel
This
is an
open
acc
ess arti
cl
e
un
der
the
CC
B
Y
-
SA
l
ic
ense
.
Corres
pond
in
g
Aut
h
or
:
Ri
ki Ruli A. Si
reg
a
r
Dep
a
rtm
ent o
f Info
rm
at
ic
s
In
sti
tut Te
knol
og
i
PL
N
Jl. Lin
gk
a
r
L
ua
r
Ba
rat
D
ur
i
K
os
am
bi
Jakar
ta
Barat,
Ind
on
es
ia
Em
a
il
: riki.ru
li
@it
pln
.ac
.id
1.
INTROD
U
CTION
A
sm
art h
om
e is on
e
of the c
om
pu
ti
ng
and in
form
ation
tech
no
l
og
y a
ppli
cat
ion
s that c
onne
ct
s sev
era
l
sm
art
ob
j
ect
s
a
nd
house
hold
dev
ic
es
or
ap
pl
ia
nces
capa
ble
of
se
ndin
g
in
f
or
m
at
ion
an
d
pro
vid
es
c
onne
ct
ion
s
(for
them
)
to
pro
vid
e
ser
vices
to
it
s
occupa
nt
s
an
d
facil
it
a
te
rem
ote
ho
m
e
co
ntr
ol
[
1]
-
[
3]
.
T
hu
s
,
ho
m
eow
ne
rs
can
c
ontrol
t
he
ir
hom
e
app
li
ances
rem
otely
an
d
m
on
it
or
t
heir
sta
t
us
es.
Re
gardin
g
s
om
e
issues
of
a
doptio
n
and
al
so
the
pote
ntia
l
op
portun
it
ie
s
of
S
HT
(sm
art
ho
m
e
te
chnolo
gies)
,
researc
h
a
nd
de
velo
pm
ent
of
SHT
can
fo
c
us
on
two
a
sp
ect
s
,
t
hat
incl
ud
e
te
chn
ic
al
de
velo
pm
ent
and
ho
w
te
ch
nolo
gy
can
be
a
dopte
d
a
nd
diffuse
d
into
t
he
m
ark
et
or
s
ociet
y.
Firstl
y,
SH
T
ai
m
s
to
im
pr
ov
e
the
qual
i
ty
of
li
fe
at
ho
m
e
by
giv
in
g
m
or
e
conve
nient
se
r
vices
an
d
a
dd
it
ion
al
featu
re
s.
Sec
ondly,
t
his
te
ch
no
l
ogy
is
us
e
d
as
a
n
e
nh
a
ncem
ent
of
a
bu
il
di
ng syst
em
f
or
b
et
te
r
u
s
e of e
nergy an
d
to
im
pr
ov
e
th
e house
util
it
ies.
T
h
e
s
m
a
r
t
h
om
e
s
y
s
t
e
m
i
n
c
l
u
d
e
s
f
e
a
t
u
r
e
s
t
ha
t
a
r
e
v
e
r
y
i
nt
e
l
l
i
g
e
n
t
i
n
t
o
d
a
y
'
s
h
um
a
n
l
i
f
e
a
n
d
i
t
s
m
o
r
e
d
e
t
a
i
l
e
d
o
b
j
e
c
t
i
v
e
s
a
r
e
c
o
n
t
r
o
l
l
i
n
g
h
o
m
e
a
p
p
l
i
a
n
c
e
s
,
s
e
c
u
r
i
n
g
c
o
n
n
e
c
t
i
o
n
c
ha
n
n
e
l
s
b
e
t
w
e
e
n
a
p
p
l
i
c
a
t
i
o
n
a
n
d
t
h
e
e
m
b
e
d
d
e
d
s
y
s
t
e
m
,
s
t
r
e
a
m
i
n
g
r
e
a
l
-
t
i
m
e
v
i
d
e
o
f
r
o
m
w
e
b
c
a
m
e
r
a
o
r
s
e
c
u
r
i
t
y
c
a
m
e
r
a
,
p
r
om
o
t
i
n
g
h
om
e
s
a
f
e
t
y
,
a
n
d
p
r
o
v
i
d
i
n
g
e
n
e
r
g
y
-
e
f
f
i
c
i
e
n
t
f
e
a
t
u
r
e
[
4
]
.
I
n
s
m
a
r
t
h
om
e
a
u
t
om
a
t
i
o
n
,
t
h
e
c
o
n
t
r
o
l
o
f
t
h
e
c
o
n
n
e
c
t
e
d
h
o
u
s
e
h
o
l
d
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.
23
, N
o.
2
,
A
ugust
20
21
:
733
-
739
734
a
p
p
l
i
a
n
c
e
s
c
a
n
b
e
c
a
r
r
i
e
d
o
u
t
u
s
i
n
g
s
m
a
r
t
ph
o
n
e
s
[
5
]
,
[
6
]
.
A
s
w
i
r
e
l
e
s
s
c
om
m
u
n
i
c
a
t
i
o
n
t
e
c
h
n
o
l
o
g
i
e
s
h
a
v
e
d
e
v
e
l
o
p
e
d
r
a
p
i
d
l
y
i
n
r
e
c
e
n
t
y
e
a
r
s
,
i
t
i
s
p
o
s
s
i
b
l
e
t
o
a
c
c
e
s
s
o
r
c
o
n
t
r
o
l
h
o
u
s
e
h
o
l
d
e
q
u
i
pm
e
n
t
r
e
m
o
t
e
l
y
[
7
]
,
[
8]
.
Sm
art
h
om
e
is
on
e
of
the
ap
plica
ti
on
s
of
the
intern
et
of
things
(Io
T
)
,
al
tho
ug
h
IoT
is
al
so
us
ed
in
oth
e
r
areas
su
c
h
as
in
t
ran
s
po
rt
an
d
traf
fic
m
anag
em
ent
[9
]
-
[12].
T
he
im
pl
e
m
e
ntati
on
of
Io
T
f
or
sm
art
hom
es
has
bec
om
e
o
ne
of
the
m
os
t
discuss
e
d
I
oT
-
relat
ed
resea
r
ch
areas
an
d
previ
ou
s
r
esearc
h
al
so
ind
ic
at
e
d
that
there
was
a
gro
wing
nu
m
ber
of
hom
e/
ho
us
e
hold
de
vices
co
nn
ect
e
d
to
the
In
te
r
net
via
sm
artp
hones
[13]
-
[
15
]
.
Also
,
huge
d
at
a
is
gen
erate
d
by
sm
art
ho
m
e
dev
ic
es.
T
here
are
gro
wing
con
ce
r
ns
,
but
pr
e
vious
w
ork
did
no
t
address
en
ough
to
m
anag
e
a
nd
analy
ze
ho
m
e
data.
Sm
art
ho
m
e
cat
egor
iz
at
ion
is
base
d
on
f
oc
us
a
re
as,
s
uch
as
energy,
in
form
ation
an
d
com
m
un
ic
at
i
on,
secu
rity
,
healt
h
,
e
nv
ir
onm
ent,
ho
m
e
entertai
nm
ent,
and
hous
e
hold
a
pp
l
ia
nces
[
16
]
,
[
17]
.
I
n
te
rm
s
of
energy
ef
fici
en
cy
in
sm
art
ho
m
es,
sever
al
is
su
es
s
houl
d
be
ta
ken
into
co
ns
ide
rat
ion
,
i.e
.
ene
rg
y
con
s
um
ption
m
on
it
or
ing
sys
tem
,
ener
gy
usa
ge
m
anag
em
e
nt,
an
d
capa
bili
ty
fo
r
processi
ng d
at
a relat
ed
t
o
the
en
e
r
gy c
onsum
pt
ion
a
rou
nd
the ho
us
e
[6
]
.
2.
RESEA
R
CH MET
HO
D
The uti
li
zat
ion
o
f
t
he
arti
fici
al
n
e
ur
al
syst
em
is broa
der
t
han w
he
n
it
w
a
s f
i
rst intr
oduce
d.
ADAL
I
N
E
has
seve
ral
ad
van
ta
ges,
incl
udin
g
us
i
ng
a
li
near
tra
nsfer
f
un
ct
io
n
instea
d
of
the
ha
rd
-
l
i
m
i
ti
ng
one.
T
hu
s
,
the
ou
t
pu
t
ca
n
va
r
y.
ADAL
IN
E
al
so
res
ponds
t
o
cha
nges
in
it
s
env
i
ronm
ent
wh
e
n
it
is
op
e
r
at
ing
.
It
is
no
t
on
ly
us
e
d
in
la
borat
or
y
ap
plica
ti
ons
that
are
m
or
e
li
kely
to
be
bas
e
d
on
pu
re
sci
ence
but
ca
n
al
so
be
util
iz
ed
in
m
or
e
app
li
cab
le
fiel
ds
.
Usa
bili
ty
will
be
m
or
e
visible
if
the
us
er
r
equ
i
res
the
ap
plica
ti
on
of
a
rtific
ia
l
intel
li
gen
ce
su
ch
as
ex
per
t
syst
e
m
s,
kn
owle
dge
-
base
d
syst
e
m
s,
and
decisi
on
support
syst
e
m
s.
Whe
n
com
par
ed
t
o
sel
f
or
gan
iz
in
g
m
aps
(
SO
M
)
,
ADAL
INE
has
a
dif
fer
e
nt
par
a
dig
m
.
S
OM
bel
ongs
to
th
e
un
s
uper
vise
d
le
arn
i
ng
par
a
di
gm
,
m
eanw
hile,
ADAL
INE
us
es
the
s
up
e
r
vised
le
ar
ning
m
echan
ism
.
T
he
us
e
of
su
pe
r
vised
le
ar
ning
is
widely
i
m
ple
m
ented
on
a
m
or
e
lim
i
t
ed
scal
e,
bu
t
f
or
m
on
it
or
in
g
a
nd
c
ontr
olli
ng
la
rg
er
qu
a
ntit
ie
s
of
sm
art
equ
ipm
ent
si
m
ultaneou
sly
SO
M/
unsupe
rv
ise
d
le
arn
i
ng
m
et
ho
d
is
con
sidere
d
m
or
e
su
it
able.
A
n
ar
ti
fici
al
neu
ral
netw
ork
(AN
N
)
is
a
netw
ork
t
hat
com
pr
ise
s
sm
a
ll
pr
ocessi
ng
unit
s
.
It
is
a
m
od
el
desig
n
e
d
to
i
m
it
at
e
h
um
an
neu
ral
net
works.
It
is
an
adap
ti
ve
syst
e
m
that
can
change
it
s
structu
re
to
so
lv
e
pro
blem
s
based
on
e
xter
nal
or
inter
nal
in
for
m
at
ion
that
flo
ws
th
r
ough
the
netw
ork.
Sim
ply
sta
te
d,
ANN
is
a
non
-
li
near
sta
ti
sti
cal
data
m
od
el
ing
too
l.
that
can
be
us
e
d
to
m
od
el
co
m
ple
x
relat
ion
s
hips
between
in
puts
and
ou
t
pu
ts
for
find
i
ng
patte
r
ns
in
data.
A
rtific
ia
l
neural
netw
orks
a
re
netw
orks
of
s
m
al
l
interconn
ect
ed
processi
ng
un
i
ts,
wh
ic
h
are
m
od
el
ed
base
d
on
ne
ur
al
ne
tworks
.
The
arti
fici
al
neu
ra
l
syst
e
m
i
s
a
l
so
a
n
inf
or
m
at
ion
pr
ocessin
g
syst
e
m
that
has
a
way
of
work
i
ng
an
d
c
har
ac
te
risti
cs
su
ch
as
ne
ur
al
netw
orks
i
n
li
vin
g
thin
gs.
This
was
la
te
r
dev
el
op
e
d
as
a
gen
er
al
iz
at
ion
of
m
a
them
a
ti
c
al
m
od
el
ing
th
at
is
patte
rn
ed
on
the
hu
m
an
co
gn
it
ive
ne
rv
e
.
This
syst
e
m
will
carry
ou
t
der
ivati
ve
le
arn
i
ng
to
achieve
c
onve
r
gen
ce
.
It
can
a
lso
be
sai
d
that
an
a
rtific
ia
l
ner
vous
syst
e
m
is
a
tool
com
m
on
ly
us
ed
an
d
a
pp
li
ed
to
pr
e
dict
an
d
cl
assify
.
The
te
sti
ng
al
gorithm
f
or
ADAL
IN
E
is:
Ob
ta
in
w
ei
gh
t
from
the learn
i
ng pr
ocess.
W
e
igh
t
In
it
ia
li
zat
i
on (w)
.
Fo
r
e
ac
h bip
olar
in
put i
n x ve
ct
or
:
a)
S
et
acti
vation
of the i
nput
un
i
t
(
=
1
,
…
.
,
)
b)
Ca
lc
ulate
n
et
w
ork value
(netv
al
)
f
ro
m
input to
ou
t
pu
t
=
∑
+
(1)
=
(
)
=
=
∑
+
(2)
Wh
e
re:
= n
et
w
ork
v
al
ue
= b
ia
se
d value
= input
data
=
W
ei
ght
valu
e
= outp
ut
To
m
od
el
and
m
at
ch
the
data
patte
rn
s
of
daily
el
ect
rici
t
y
con
s
um
ption
based
on
the
ADALIN
E
m
et
ho
d,
the
da
il
y
ener
gy
c
on
su
m
ption
patte
rn
is
de
fine
d
ba
sed
on
ti
m
e
a
nd
date.
T
hen,
it
is
com
par
ed
with
the
act
ual
data
.
The
m
et
ho
do
log
y
f
or
im
plem
enting
A
D
A
LINE
f
or
t
he
prototype
is
s
how
n
in
Fig
ur
e
1,
as
fo
ll
ows.
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
Io
T
for
smart
home
system
(
Puji C
atu
r
S
isw
i
pr
apti
ni
)
735
Figure
1.
A
DALINE
te
sti
ng
m
et
ho
dolo
gy
This
stu
dy
has
identifie
d
t
he
pro
blem
that
le
s
s
i
m
ple
m
entat
i
on
of
an
A
NN
syst
e
m
m
od
el
as
a
m
ob
il
e
syst
e
m
that
all
ow
s
s
et
ti
ng
th
e
ho
m
e
app
li
ances
co
ntr
ol
in
on
e
i
nterf
ace
[18].
N
or
m
al
ly
in
pr
e
vious
re
searc
h
on
ly
h
a
d
aut
om
at
ic
co
ntro
l,
bu
t di
d
not y
et
h
ave a
ny f
unct
ion
for
m
on
it
or
ing
hom
e app
li
ances su
c
h
as a
la
m
p
,
tem
per
at
ur
e
,
ai
r
co
nd
it
io
ning
(
AC
)
,
an
d
te
le
visio
n
(
TV
)
[18]
-
[
21]
.
The
de
sign
e
d
sci
enc
e
and
te
ch
nolo
gy
for
hu
m
anity
(
STH
)
a
ppli
es
ad
aptive
li
nea
r
neural
netw
ork
(
A
DALI
NE
)
as
it
s
m
et
ho
d,
I
oT
,
a
nd
util
iz
es
m
ic
ro
co
ntro
ll
e
rs,
se
ns
ors,
a
nd
seve
ral
oth
e
r
suppo
rting
e
le
ct
ro
nic
de
vic
es
[21]
,
[22].
This
m
ic
ro
co
nt
ro
ll
er
m
et
ho
d
is
the
hard
war
e
us
e
d
to
ob
ta
i
n
a
la
m
p
con
tr
oller
wh
il
e
th
e
ADALINE
m
et
ho
d
is
t
o
get
an
arti
fici
al
nerv
ou
s
s
yst
em
m
od
el
into
a
syst
e
m
[23]
-
[
25
]
.
A
nother
one
is
t
he
ou
tc
om
es
of
t
he
sm
art
dry
er
prot
otype
ca
n
be
inte
gr
at
e
d
with
the
m
od
el
of
a
n
a
rtific
ia
l
nerv
ou
s
sys
tem
us
ing
A
D
ALINE
[
26
]
.
This
prototype
desi
gn
init
ia
te
s
by
inv
entin
g
a
sce
na
rio
de
sig
n
to
discuss
t
he
to
ol
s
us
ed
a
nd
th
e
connecti
on
be
tween
s
of
t
wa
re
an
d
hard
war
e.
S
of
t
war
e
desi
gn
as
sist
s
in
the
flo
w
of
t
he
progr
a
m
to
be
in
ve
nted
by
c
onne
ct
ing
hardware
too
ls,
wh
e
reas
hard
war
e
desig
n
i
s
util
iz
ed
to
place
each
pa
rt
that
will
be
us
ed
[
27
]
,
[28].
Wh
e
n
this
ho
m
e
app
li
anc
es
-
c
on
trol
a
pp
li
cat
io
n
has
bee
n
des
ign
e
d,
t
he
c
on
t
ro
l
syst
em
is
t
hen
te
ste
d
t
o
de
fine
t
he
c
om
pliance
of the a
ppli
cat
i
on that
has bee
n
m
ade [29
]
.
As
see
n
in
Figure
1,
t
he
A
DA
L
I
NE
m
odel
is
util
iz
ed
f
or
t
rainin
g
the
process
of
art
ific
ia
l
neu
ra
l
net
w
orks.
T
he
ta
rg
et
s
an
d
in
pu
ts
are
trai
ne
d
with
a
ne
tw
ork
t
hat
has
be
en
de
velo
ped
to
get
t
he
le
a
rn
i
ng
weig
hts
to
be
util
iz
ed
as
t
he
fun
dam
ental
for
cal
culat
io
ns
on
the
f
ollo
wing
inc
om
ing
tr
ai
nin
g
data
[
30
]
.
If
the
sta
ges
of
norm
al
iz
at
ion
ha
ve
been
im
ple
m
e
nt
ed,
t
he
data
i
s
read
y
to
be
proces
sed
i
nto
t
he
A
D
ALINE
neural
netw
ork
[
3
1
]
.
Wh
e
n
enteri
ng
ta
rg
et
data
a
nd
i
nput
data
a
s
trai
ni
ng
data
,
the
te
sti
ng
is
pe
rfor
m
ed
us
ing
a
neural
netw
ork
too
l
in
the
Ma
tl
ab
app
li
cat
ion
a
nd
beg
i
ns
to
create
netw
orks
by
trai
ni
ng
on
ta
rg
et
dat
a
and
input
data.
Fro
m
a
te
chn
ic
al
po
i
nt
of
vie
w,
sever
al
as
pects
need
to
be
de
velo
ped
nam
ely
the
us
er
interface,
sm
art h
ardwar
e, and t
heir
s
of
tware
platfo
rm
[3
2
].
3.
RESU
LT
S
A
ND
D
IS
C
USS
ION
Figure
2
s
how
s
the
ci
rcu
it
de
sign
of
the
sm
art
ho
m
e
s
yst
e
m
with
a
so
la
r
pan
el
.
T
he
acce
ss
po
i
nt
is
us
e
d
for
inter
netw
ork
an
d
Ra
sp
be
rr
y
Pi
serv
e
s
as
the
web
s
er
ver.
Th
e
m
echan
ism
of
s
olar
pa
nel
s
is
by
conve
rting
so
l
ar
e
nergy
int
o
el
ect
ric
current
,
al
so
kn
own
a
s
ph
otovo
lt
ai
c
syst
e
m
s.
The
ba
tt
ery
is
us
e
d
t
o
st
or
e
the curre
nt ele
ct
rici
ty
an
d
s
upply i
t t
o
t
he
house
hold a
ppli
ances as
the alt
ern
at
ive
po
wer so
ur
ce
in
th
e e
ven
t
of
a
power
outa
ge
.
The
Sm
artph
one
is
us
e
d
as
the
cl
ie
nt
that
can
con
tr
ol
hom
e
app
li
ances
su
ch
as
ch
oo
se
TV
Chan
nels,
tu
r
n
the
li
gh
ts
on
a
nd
off
,
a
nd
cha
ng
e
t
he
sp
e
ed
of
fan
or
ai
r
co
nd
it
io
ner
.
The
process
of
t
he
cl
ie
nt
is
by
retrievi
ng
data
to
ope
n
the
HTML
5
F
ra
m
ewo
r
k
f
ro
m
the
Ra
spbe
rr
y
a
s
a
com
pu
te
r
s
erv
e
r,
bo
t
h
of
wh
ic
h
are
c
onnected
to
the
acce
ss
point.
H
y
p
erT
ex
t
m
ark
up
la
nguage
(
HTML
)
is
u
sed
as
cont
ent
s
writt
en
in
it
ar
e
ea
si
l
y
ac
c
essed
b
y
th
e
brows
ers
and
it
offe
rs
m
ore
eff
ic
ie
n
t
codi
ng
.
And
then
the
A
ndr
oid
cl
ie
nt
processes
AC/Fa
n,
th
e
tur
n
on
a
nd
off
li
gh
ts,
T
V
c
ha
nn
el
by
pressi
ng
t
he
O
N
bu
tt
on
on
the
i
nter
face
an
d
se
nding
t
his
com
m
a
nd
t
o
the
acce
ss
poi
nt
then
receive
d
by
the
ser
ve
r.
A
s
the
se
rv
e
r
gi
ves
the
c
om
m
and
to
the
Ether
net
Sh
ie
l
d
the
n
passes
it
to
the
m
ic
ro
co
ntr
oller,
t
he
c
omm
a
nd
is
recei
ved
by
the
relay
an
d
the
li
gh
t
will
tur
n
on.
T
he
c
ircuit
requires
on
e
c
hip
of
Ra
s
pb
e
r
ry
as
the
m
ic
ro
con
t
ro
ll
er.
The
Ra
sp
be
rr
y
ser
ves
as
the
se
rver
of
data
sent
by
th
e
sm
artph
on
e
cl
ie
nt,
t
hen
f
orwa
rd
s
the
com
m
a
nd
to
the
de
vic
es
via
t
he
acce
ss
point.
Acces
s
points
w
ork
as
an
interm
ediary b
et
ween
softwa
r
e an
d har
dware
(devices
)
in
th
e proces
s
of
c
ontr
olli
ng
t
he
a
ppli
ances
.
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.
23
, N
o.
2
,
A
ugust
20
21
:
733
-
739
7
36
Figure
2
.
The
c
ircuit
d
esi
gn
of the sm
art h
om
e syst
e
m
The
pr
ocess
of
this
prot
otyp
e
is
to
m
od
el
and
m
at
ch
the
data
patte
r
ns
of
daily
consum
ption
of
el
ect
r
cal
energ
y
based
on
t
he
ada
ptive
li
ne
ar
arti
fici
al
ne
ur
al
netw
ork
m
e
tho
d
so
th
at
the
de
finiti
on
of
el
ect
rical
energy
d
ai
ly
con
s
um
pt
ion
patte
r
ns
base
d
,
the
n
c
om
par
e
with
th
e
act
ual
data.
Table
s
1
a
nd
2
sh
ows
the
behavi
or
of
the
el
ect
rical
load
on
the
tr
ansm
issi
on
li
ne
fo
r
el
ect
rici
ty
con
s
um
ption
in
the
area
of
Java
,
Ba
li
,
and
Ma
dura
.
T
his
data was
ob
s
er
ved
e
ver
y hou
r
daily
for
on
e
m
on
th
,
in D
ecem
ber
.
Th
us
,
t
he
dim
ensio
n
of the
vecto
r
-
m
at
rix
us
e
d for
the
sim
ulati
on
is 31
days
by
24 ho
ur
s
, or
31
x
24.
Table
1.
A
ver
a
ge
m
on
thly
po
wer co
nsum
pti
on
m
egaw
at
t (
M
W
)
in
Jav
a
, B
al
i, Ma
dura
(
12:30
AM
–
6:0
0
AM
)
HOUR
DAT
E
1
2
:3
0
AM
1
:0
0
AM
1
:3
0
AM
2
:0
0
AM
2
:3
0
AM
3
:0
0
AM
3
:3
0
AM
4
:0
0
AM
4
:3
0
AM
5
:0
0
AM
5
:3
0
AM
6
:0
0
AM
1
1
2
,12
4
1
2
,08
6
1
1
,86
2
1
1
,67
2
1
1
,48
7
1
1
,38
1
1
1
,34
3
1
1
,35
7
1
1
,45
1
1
1
,52
0
1
1
,25
4
1
0
,46
8
2
1
0
,86
7
1
0
,74
1
1
0
,65
9
1
0
,46
6
1
0
,39
9
1
0
,42
2
1
0
,45
4
1
0
,55
8
1
0
,79
3
1
1
,01
9
1
0
,95
1
1
0
,66
0
3
1
1
,59
6
1
1
,37
4
1
1
,28
5
1
1
,16
9
1
0
,97
3
1
1
,01
2
1
1
,00
3
1
1
,05
7
1
1
,32
0
1
1
,48
5
1
1
,30
0
1
0
,90
7
4
1
1
,91
5
1
1
,71
6
1
1
,62
9
1
1
,56
7
1
1
,44
6
1
1
,43
2
1
1
,28
1
1
1
,43
4
1
1
,80
9
1
2
,16
0
1
2
,27
1
1
1
,90
8
5
1
2
,96
0
1
2
,78
3
1
2
,68
2
1
2
,52
7
1
2
,54
4
1
2
,38
1
1
2
,35
1
1
2
,45
6
1
2
,74
1
1
3
,12
9
1
3
,24
2
1
2
,75
2
6
1
3
,30
4
1
3
,11
9
1
3
,02
0
1
2
,83
3
1
2
,78
3
1
3
,03
5
1
2
,85
4
1
2
,91
4
1
3
,39
3
1
3
,68
6
1
3
,88
5
1
3
,67
1
7
1
3
,48
4
1
3
,26
9
1
3
,13
1
1
2
,95
2
1
2
,89
0
1
2
,71
3
1
2
,72
3
1
3
,06
4
1
3
,14
4
1
3
,48
4
1
3
,65
1
1
3
,37
1
8
1
3
,25
5
1
3
,14
1
1
2
,94
9
1
2
,84
5
1
2
,76
1
1
2
,64
4
1
2
,61
1
1
2
,66
0
1
2
,87
0
1
3
,29
3
1
3
,47
7
1
3
,26
2
9
1
3
,30
5
1
3
,05
8
1
2
,89
6
1
2
,72
9
1
2
,40
6
1
2
,24
8
1
2
,19
1
1
2
,25
5
1
2
,52
8
1
3
,08
2
1
3
,11
6
1
2
,77
0
10
1
2
,32
6
1
2
,11
0
1
2
,06
5
1
2
,05
6
1
1
,86
0
1
1
,86
2
1
1
,80
6
1
1
,82
7
1
2
,14
9
1
2
,28
3
1
2
,40
3
1
1
,81
6
11
1
2
,16
0
1
1
,99
5
1
1
,89
2
1
1
,80
6
1
1
,65
9
1
1
,59
1
1
1
,57
0
1
1
,70
6
1
1
,92
1
1
2
,58
7
1
2
,75
4
1
2
,63
0
12
1
3
,42
1
1
3
,12
0
1
3
,10
3
1
2
,77
6
1
2
,76
9
1
2
,63
3
1
2
,57
1
1
2
,72
9
1
3
,10
0
1
3
,51
3
1
3
,61
5
1
2
,98
0
13
1
3
,38
3
1
3
,03
6
1
2
,98
6
1
2
,88
6
1
2
,63
4
1
2
,61
0
1
2
,38
0
1
2
,53
6
1
2
,94
4
1
3
,26
4
1
3
,67
3
1
3
,48
5
14
1
3
,37
5
1
3
,13
3
1
3
,05
1
1
2
,89
4
1
2
,71
6
1
2
,62
7
1
2
,54
4
1
2
,62
5
1
3
,00
7
1
3
,43
1
1
3
,52
0
1
3
,51
1
15
1
2
,92
2
1
2
,84
0
1
2
,67
5
1
2
,48
4
1
2
,43
7
1
2
,34
0
1
2
,24
5
1
2
,38
2
1
2
,64
7
1
3
,15
8
1
3
,50
4
1
3
,31
5
16
1
3
,28
2
1
3
,08
2
1
2
,86
5
1
2
,66
1
1
2
,56
3
1
2
,53
0
1
2
,40
2
1
2
,42
8
1
2
,86
2
1
3
,23
3
1
3
,35
8
1
2
,93
3
17
1
2
,38
4
1
2
,25
9
1
2
,12
4
1
1
,99
6
1
1
,82
4
1
1
,75
8
1
1
,53
5
1
1
,63
8
1
1
,89
9
1
2
,20
6
1
2
,24
7
1
1
,71
8
18
1
1
,84
5
1
1
,66
0
1
1
,56
4
1
1
,48
4
1
1
,23
3
1
1
,27
9
1
1
,21
6
1
1
,21
7
1
1
,61
0
1
2
,16
7
1
2
,33
7
1
2
,29
5
19
1
3
,22
4
1
2
,98
6
1
2
,79
3
1
2
,83
1
1
2
,63
9
1
2
,65
2
1
2
,51
0
1
2
,58
0
1
2
,83
8
1
3
,31
3
1
3
,55
9
1
3
,24
5
20
1
3
,18
3
1
2
,99
1
1
2
,98
9
1
2
,83
4
1
2
,67
5
1
2
,48
7
1
2
,45
6
1
2
,51
9
1
2
,73
7
1
3
,28
8
1
3
,67
5
1
3
,46
3
21
1
3
,26
2
1
3
,13
9
1
2
,81
7
1
2
,60
9
1
2
,56
1
1
2
,50
8
1
2
,38
3
1
2
,41
4
1
2
,80
5
1
3
,23
1
1
3
,33
4
1
3
,37
4
22
1
3
,29
2
1
3
,14
2
1
2
,98
8
1
2
,82
9
1
2
,80
1
1
2
,61
8
1
2
,57
2
1
2
,50
1
1
2
,96
4
1
3
,28
4
1
3
,70
4
1
3
,32
0
23
1
3
,24
0
1
3
,04
5
1
2
,81
5
1
2
,71
0
1
2
,60
1
1
2
,59
8
1
2
,35
3
1
2
,46
7
1
2
,73
0
1
3
,15
2
1
3
,46
1
1
2
,92
1
24
1
2
,85
9
1
2
,48
5
1
2
,35
6
1
2
,24
0
1
2
,25
2
1
2
,02
9
1
1
,95
7
1
1
,93
8
1
2
,33
4
1
2
,41
9
1
2
,67
3
1
2
,01
8
25
1
2
,43
3
1
2
,24
3
1
2
,13
9
1
1
,92
5
1
1
,73
9
1
1
,74
0
1
1
,74
5
1
1
,78
6
1
2
,09
7
1
2
,58
3
1
2
,88
9
1
2
,56
9
26
1
3
,30
4
1
3
,07
2
1
2
,94
1
1
2
,91
4
1
2
,71
3
1
2
,63
8
1
2
,61
2
1
2
,58
9
1
3
,03
5
1
3
,41
7
1
3
,75
5
1
3
,38
8
27
1
3
,34
9
1
3
,12
1
1
3
,05
9
1
2
,79
4
1
2
,87
2
1
2
,67
4
1
2
,64
8
1
2
,75
6
1
2
,97
4
1
3
,53
1
1
3
,93
1
1
3
,43
7
28
1
3
,35
1
1
3
,25
2
1
3
,17
6
1
2
,93
0
1
2
,80
7
1
2
,64
4
1
2
,68
9
1
2
,67
3
1
3
,08
7
1
3
,33
6
1
3
,74
6
1
3
,42
6
29
1
3
,63
3
1
3
,42
1
1
3
,34
7
1
3
,21
0
1
3
,01
6
1
2
,80
6
1
2
,78
3
1
2
,91
8
1
3
,26
8
1
3
,56
7
1
3
,88
1
1
3
,64
4
30
1
3
,53
1
1
3
,30
8
1
3
,07
1
1
2
,97
0
1
2
,77
7
1
2
,68
4
1
2
,71
2
1
2
,81
4
1
3
,06
0
1
3
,31
4
1
3
,56
6
1
3
,04
2
31
1
2
,48
6
1
2
,30
9
1
1
,98
6
1
2
,03
7
1
1
,92
9
1
1
,65
4
1
1
,66
8
1
1
,83
7
1
1
,95
5
1
2
,30
6
1
2
,43
9
1
1
,99
0
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
Io
T
for
smart
home
system
(
Puji C
atu
r
S
isw
i
pr
apti
ni
)
737
Table
2
.
Aver
age m
on
thly
powe
r
c
onsu
m
ption
m
egaw
at
t (
M
W
)
in
Jav
a
, B
al
i,
Ma
dura
(
6
:30
AM
–
12
:0
0
P
M)
HOUR
DAT
E
6
:3
0
AM
7
:0
0
AM
7
:3
0
AM
8
:0
0
AM
8
:3
0
AM
9
:0
0
AM
9
:3
0
AM
1
0
:0
0
AM
1
0
:3
0
AM
1
1
:0
0
AM
1
1
:3
0
AM
1
2
:0
0
PM
1
9
,84
8
9
,44
5
9
,66
3
9
,64
9
9
,61
0
9
,58
3
9
,63
4
9
,97
2
1
0
,01
4
1
0
,10
6
1
0
,07
2
9
,88
9
2
1
0
,35
7
1
0
,38
3
1
0
,68
7
1
0
,99
7
1
1
,33
8
1
1
,69
1
1
1
,80
8
1
2
,18
4
1
2
,09
5
1
2
,12
0
1
2
,12
9
1
1
,96
3
3
1
0
,75
6
1
0
,43
7
1
0
,69
7
1
0
,70
6
1
1
,03
7
1
1
,15
7
1
1
,29
0
1
1
,50
3
1
1
,59
4
1
1
,69
7
1
1
,63
6
1
1
,58
0
4
1
1
,78
7
1
2
,07
1
1
2
,67
3
1
3
,41
2
1
4
,14
6
1
4
,46
7
1
4
,67
7
1
4
,76
8
1
5
,12
3
1
5
,39
8
1
5
,22
7
1
4
,71
9
5
1
2
,43
3
1
2
,58
5
1
3
,18
5
1
3
,73
4
1
4
,40
2
1
4
,32
9
1
4
,92
5
1
5
,18
8
1
5
,33
5
1
5
,42
9
1
5
,34
8
1
4
,67
1
6
1
3
,25
8
1
3
,19
3
1
3
,80
7
1
4
,20
7
1
4
,74
4
1
4
,97
9
1
5
,15
7
1
5
,45
9
1
5
,41
8
1
5
,57
5
1
5
,41
6
1
4
,90
6
7
1
3
,01
5
1
2
,99
4
1
3
,68
7
1
4
,00
8
1
4
,69
0
1
4
,97
4
1
5
,16
8
1
5
,18
6
1
5
,39
9
1
5
,36
3
1
5
,40
7
1
5
,27
8
8
1
2
,90
8
1
2
,92
2
1
3
,45
3
1
4
,16
0
1
4
,63
1
1
4
,94
7
1
5
,11
4
1
5
,32
7
1
5
,44
7
1
5
,29
3
1
4
,75
6
1
3
,76
0
9
1
2
,32
7
1
2
,32
4
1
2
,43
6
1
2
,98
8
1
3
,49
0
1
3
,68
7
1
3
,89
3
1
3
,94
6
1
4
,11
3
1
4
,13
1
1
4
,08
9
1
3
,53
1
10
1
1
,48
6
1
1
,27
2
1
1
,11
6
1
1
,41
8
1
1
,60
2
1
1
,67
5
1
1
,71
5
1
1
,85
5
1
1
,96
6
1
2
,01
2
1
1
,90
3
1
1
,63
0
11
1
2
,19
0
1
2
,21
9
1
3
,11
7
1
3
,55
4
1
4
,35
8
1
4
,53
0
1
4
,99
7
1
5
,09
5
1
5
,30
1
1
5
,25
7
1
5
,24
8
1
4
,66
4
12
1
2
,93
6
1
2
,79
8
1
3
,68
2
1
4
,05
6
1
4
,57
5
1
4
,90
3
1
4
,95
1
1
5
,11
7
1
5
,44
1
1
5
,35
4
1
5
,29
6
1
4
,42
7
13
1
3
,16
3
1
3
,06
6
1
3
,56
0
1
4
,06
1
1
4
,76
6
1
4
,94
3
1
5
,06
7
1
5
,19
9
1
5
,24
3
1
4
,94
7
1
4
,94
9
1
4
,12
6
14
1
3
,15
9
1
2
,94
4
1
3
,54
4
1
4
,19
7
1
4
,59
3
1
4
,95
0
1
5
,04
1
1
5
,18
6
1
5
,31
1
1
5
,38
0
1
5
,23
2
1
4
,59
8
15
1
2
,95
4
1
2
,79
5
1
3
,46
4
1
3
,95
3
1
4
,26
1
1
4
,69
5
1
4
,56
8
1
4
,85
2
1
5
,04
6
1
5
,04
8
1
4
,44
3
1
3
,60
8
16
1
2
,49
2
1
2
,11
7
1
2
,59
1
1
2
,99
0
1
3
,37
0
1
3
,51
1
1
3
,56
5
1
3
,92
1
1
4
,12
6
1
4
,16
5
1
4
,00
7
1
3
,37
3
17
1
1
,41
2
1
1
,15
2
1
1
,25
0
1
1
,35
5
1
1
,61
4
1
1
,67
2
1
1
,89
7
1
1
,83
4
1
1
,86
1
1
1
,92
1
1
1
,85
3
1
1
,58
9
18
1
2
,00
9
1
2
,09
8
1
2
,81
0
1
3
,45
4
1
3
,94
5
1
4
,19
9
1
4
,39
7
1
4
,62
2
1
4
,77
8
1
4
,87
2
1
4
,76
7
1
4
,34
8
19
1
3
,01
7
1
3
,04
3
1
3
,49
1
1
4
,00
9
1
4
,53
0
1
4
,66
0
1
4
,86
9
1
5
,04
0
1
5
,13
1
1
5
,21
6
1
5
,04
2
1
4
,39
8
20
1
3
,07
5
1
2
,96
2
1
3
,65
6
1
3
,98
4
1
4
,63
3
1
4
,89
6
1
4
,98
2
1
5
,31
5
1
5
,34
6
1
5
,29
2
1
5
,34
3
1
4
,63
6
21
1
3
,22
2
1
3
,07
0
1
3
,46
5
1
4
,00
5
1
4
,59
6
1
4
,95
2
1
5
,05
2
1
5
,10
0
1
5
,40
1
1
5
,35
1
1
5
,19
3
1
4
,45
6
22
1
2
,99
3
1
2
,97
4
1
3
,46
1
1
4
,09
7
1
4
,36
6
1
4
,89
3
1
5
,01
3
1
5
,29
0
1
5
,47
9
1
5
,43
0
1
4
,87
0
1
3
,90
7
23
1
2
,55
5
1
2
,18
0
1
2
,55
8
1
3
,15
8
1
3
,54
6
1
3
,77
6
1
3
,99
4
1
4
,14
0
1
4
,42
6
1
4
,50
6
1
4
,11
8
1
3
,75
8
24
1
1
,58
5
1
1
,39
3
1
1
,55
8
1
1
,63
6
1
1
,89
6
1
1
,93
3
1
2
,16
8
1
2
,36
2
1
2
,42
0
1
2
,50
9
1
2
,29
3
1
2
,06
8
25
1
2
,48
8
1
2
,39
7
1
3
,15
3
1
3
,77
5
1
4
,45
3
1
4
,72
8
1
4
,86
2
1
4
,94
6
1
5
,30
0
1
5
,26
7
1
5
,16
6
1
4
,47
5
26
1
3
,25
7
1
3
,22
2
1
3
,64
6
1
4
,59
9
1
5
,06
4
1
5
,08
5
1
5
,09
9
1
5
,21
0
1
5
,65
3
1
5
,35
9
1
5
,22
4
1
4
,75
5
27
1
3
,03
7
1
3
,04
1
1
3
,67
3
1
4
,30
4
1
4
,85
7
1
5
,26
5
1
5
,29
8
1
5
,57
2
1
5
,60
5
1
5
,64
4
1
5
,65
6
1
4
,99
1
28
1
3
,20
7
1
3
,27
6
1
3
,64
1
1
4
,22
3
1
4
,75
3
1
5
,05
7
1
5
,17
7
1
5
,52
1
1
5
,63
1
1
5
,46
3
1
5
,38
4
1
4
,86
4
29
1
3
,28
5
1
3
,13
8
1
3
,75
9
1
4
,28
4
1
4
,83
5
1
5
,10
2
1
5
,29
7
1
5
,47
6
1
5
,71
6
1
5
,73
6
1
5
,04
6
1
4
,22
1
30
1
2
,37
4
1
2
,18
7
1
2
,66
5
1
2
,92
7
1
3
,24
5
1
3
,52
4
1
3
,66
2
1
3
,89
5
1
4
,06
5
1
3
,89
0
1
3
,90
5
1
3
,59
8
31
1
1
,81
3
1
1
,50
2
1
1
,46
1
1
1
,59
6
1
1
,79
2
1
1
,86
7
1
1
,80
8
1
1
,95
1
1
2
,09
2
1
2
,00
5
1
1
,90
1
1
1
,66
9
Figure
3
sho
w
s
the
ge
ne
ral
24
-
hour
el
ect
ri
ci
ty
con
su
m
ption
patte
r
n
in
t
he
area
of
Ja
va
,
Ba
li
,
and
Ma
dura.
That
patte
rn
the
n
is
no
rm
al
iz
ed
to
get
a
bette
r
resu
lt
.
Data
nor
m
al
iz
ation
is
conve
rting
the
act
ual
value
i
nto
a cer
ta
in v
al
ue
whic
h
ca
n
the
n be
use
d
for det
erm
i
ning the
arti
fic
ia
l neural
n
et
w
ork
m
od
el
.
Figure
3. 24
-
hour ele
ct
rici
ty
co
nsum
ption
pa
tt
ern
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.
23
, N
o.
2
,
A
ugust
20
21
:
733
-
739
738
In
this
case
,
t
he
data
will
be
norm
al
iz
ed
into
0
(ze
ro)
or
1
(
one)
.
T
he
res
ult
of
the
data
no
rm
aliza
ti
on
pr
ese
nted
in
t
he
g
e
ner
al
patte
rn of
daily
el
ect
rical
b
eha
vior
as seen
in Fi
gu
re
4.
Figure
4. The
gen
e
ral
patte
rn of
daily
elec
tric
al
b
eha
vi
or
e
ve
ry ho
ur
a
fter
norm
al
iz
a
ti
on
In
arti
fici
al
ne
ur
al
net
works
base
d
on
s
uper
vised
le
ar
ning
par
a
dig
m
s,
in
gen
e
ral,
3
(t
hree)
m
a
tric
es
are
need
e
d.
T
hese
m
at
rices
functi
on
as
tra
ining
data
that
re
qu
i
res
trai
ni
ng
m
at
rices
and
ta
r
get
m
at
r
ic
e
s.
Anothe
r
m
at
ri
x
is
a
sim
ulatio
n
m
at
rix.
Gen
erall
y
dep
ic
te
d
as
sho
wn
in
F
igure
4.
T
hese
three
m
at
rices
will
be
ta
ken
res
pecti
ve
ly
based
on
hourl
y
loa
d
patte
rn
s
an
d
daily
load
patte
r
ns
.
I
n
the
case
of
lo
ad
behavi
or
p
a
t
te
rn
s
,
the
trai
ni
ng
m
at
rix
is
a
31
x
2
m
at
rix
for
lo
ad
beh
a
vior
ba
sed
on
pe
r
30
m
inu
te
s
in
24
hours.
T
he
re
is
al
so
a
24
x
2
m
at
rix
fo
r
l
oad
beh
a
vior
base
d
on
weekday
s
an
d
ho
l
iday
s.
T
he
trai
nin
g
m
at
ri
x
is
ta
ke
n
f
r
om
the
la
rg
est
value
a
nd the
sm
allest
value
fro
m
the
nor
m
al
iz
ed
load beh
a
vior
da
ta
m
at
rix
colu
m
n.
4.
CONCL
US
I
O
N
This
r
esearc
h
was
unde
rta
ken
to
desi
gn
an
d
e
valuate
that
ne
ural
netw
ork
al
gor
it
h
m
nam
el
y
ADAL
IN
E
ca
n
be
im
ple
m
ented
to
co
ntr
ol
hous
e
hold
ap
pl
ia
nces
as
an
integ
rated
sm
art
ho
m
e
syst
e
m
with
a
so
la
r
pa
nel.
W
hen
the
syst
e
m
w
orks
a
uto
m
at
ic
al
ly
based
on
the
rea
d
in
gs
of
t
he
e
xisti
ng
sens
or
s
,
ADA
LINE
is
us
e
d
acc
ording
to
t
hat
co
ndit
ion
in
t
his
stu
dy.
T
his
syst
e
m
can
be
co
ntr
olled
via
A
ndr
oid
a
ppli
cat
ions
that
act
as
re
m
ote
c
on
t
ro
ls
w
he
n
in
m
anu
al
m
od
e.
To
ena
ble
the
op
erati
on
of
this
prototype,
on
e
m
ic
ro
co
ntr
oller
is
require
d
as
c
on
t
ro
ll
er
a
nd
s
erv
e
r.
Th
e acc
ess
point
is nee
ded
to
c
ontrol all
the
m
edia
c
onnected
to
t
he
sm
art
syst
e
m
.
In
te
gr
at
ing
a
ndro
i
d
-
base
d
com
pone
nts
an
d
a
ppli
cat
ion
s
is
e
xpect
ed
to
inc
rea
se
the
fle
xib
il
it
y
(f
or
us
ers
)
in
c
ontr
olli
ng
ho
m
e
app
li
ance
s.
Ov
e
r
an
d
a
bove
th
at
,
the
aut
om
a
ti
c
con
tr
oller
of
ho
us
eh
old
de
vice
s
base
d
on
se
nsor
rea
dings
is
exp
ect
e
d
to
in
crease
the
ef
fici
ency
of
el
ect
rici
ty
us
age.
Ba
sed
on
te
sti
ng
a
nd
fin
dings
from
t
he
res
ults
of
th
is
stud
y,
that
t
he
IoT
sm
art
ho
m
e
m
od
el
wit
h
the
ADAL
I
NE
m
e
tho
d
ca
n
be
a
m
od
el
reco
m
m
end
at
ion,
as
a
proposal
to
ob
ta
in
a
n
ef
fic
ie
nt
and
be
nefi
ci
al
el
ect
ric
it
y
us
age
patte
r
n
to
be
i
m
ple
m
ented
in a
sm
art h
om
e
.
REFERE
NCE
S
[1]
V.
Ric
qu
ebour
g,
D.
Menga
,
D.
Durand,
B.
Mar
hic
,
L.
Dela
ho
c
he
.
and
C.
Loge,
“
The
Sm
art
H
om
e
Conce
pt
:
our
imm
edi
at
e
futur
e,
”
2006
1ST IE
EE
Inte
rnatio
nal
Confe
renc
e
on
E
-
Learning
in
Industrial
El
ectronics
,
2006,
pp.
23
-
28,
doi
:
10
.
1109
/ICE
LI
E.
2006
.
3
47206
.
[2]
F.
K.
Aldric
h,
“
Sm
art
Ho
m
es :
Past,
Present
and
Futures,
”
in
Book
chap
:
Inside
the
Smar
t
Hom
e
,
R.
Harpe
r,
London,
UK
:
S
p
r
i
n
g
e
r
,
2003,
pp.
17
-
39,
doi: 10.
1
007/1
-
85233
-
854
-
7_2.
[3]
M.
R.
Alam,
M.
B.
I.
Reaz,
and
M.
A.
M.
Ali,
“
A
Revi
ew
of
Sm
art
Hom
es
—
P
ast,
Present
,
and
Future,
”
in
I
EEE
Tr
ansacti
ons
on
Syste
ms
,
Man
,
a
nd
Cybe
rnetics,
Part
C
(
Appl
ic
ati
ons
and
Re
vi
ew
s)
,
vol.
42,
no.
6,
pp.
1190
-
1203,
Nov.
2012,
doi:
10.
1109/T
SM
CC.2012.
2189204
.
[4]
V.
Jy
o
thi,
M.
G.
Krishna,
B.
Ra
vee
ndra
n
adh,
an
d
D.
Rupal
in
,
“
IOT
Based
Sm
ar
t
Hom
e
Sy
stem
Te
chno
logi
es,
”
Inte
rnational
Jo
urn
al
of Engi
ne
e
ring R
ese
arch
a
nd
Dev
e
lopment
,
vol.
13,
no.
2,
p
p.
31
-
7
,
2017
.
[5]
L.
C.
De
Silva
,
C.
Morikawa
,
and
I.
M.
Petra
,
“
Stat
e
of
the
art
of
sm
art
ho
me
s,”
Engi
n
ee
rin
g
Appl
ic
at
ions
o
f
Arti
ficial Int
el
l
ig
enc
e
,
vo
l. 25, no
.
7
,
pp
.
1313
-
13
21
2012,
doi: 1
0.
1016/j
.
enga
ppa
i.2012.05.
002
.
[6]
E.
Kam
el
and
A.
M.
Mem
ari
,
“
State
-
of
-
the
-
Ar
t
Revi
ew
of
En
er
g
y
Sm
art
Hom
es,
”
J
.
Archit.
En
g.
,
vol
.
25
,
no.
1,
2019,
doi
:
10
.
10
61/(ASCE)AE.
1
943
-
5568.
00003
37.
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
Io
T
for
smart
home
system
(
Puji C
atu
r
S
isw
i
pr
apti
ni
)
739
[7]
V.
D.
Vaid
y
a
a
nd
P.
Vishw
akarm
a,
“
A
Com
par
at
iv
e
Anal
y
sis
on
Sm
art
Hom
e
S
y
stem
to
Con
t
rol,
Monitor
an
d
Secur
e
Hom
e,
b
ase
d
on
t
ec
hnolo
gie
s li
k
e
GS
M,
I
OT,
Bluetooth a
nd
PIC Micr
oco
ntrol
ler wit
h
Zi
g
Bee
Modula
ti
on
,
”
2018
Inte
rnati
onal
Confe
ren
c
e
on
Smar
t
Cit
y
a
nd
Eme
rging
Technol
o
gy
(
ICSCET)
,
2018,
pp.
1
-
4,
doi:
10
.
1109/IC
SC
ET
.
2018.
853
7381
.
[8]
A.
Le
oni,
et
.
a
l
.
,
“
Ene
rg
y
h
arv
esti
ng
opti
m
izat
ion
for
buil
t
-
in
power
rep
la
c
e
m
ent
of
el
ec
tro
nic
m
ult
isensor
y
arc
hi
te
c
ture,”
A
EU
-
Inte
rnatio
nal
Journal
of
El
e
ct
ronics
and
Comm
unic
ati
o
ns
,
vol
.
107,
p
p.
170
-
176,
201
9,
doi:
10
.
1016/j.a
e
ue.
2019.
05
.
002
[9]
Indria
nto
,
M.
N.
I.
Sus
ant
i,
R.
R.
A.
Sirega
r,
P.
Putri
J.,
and
Y.
Purw
ant
o
,
“
Sm
art
ta
xi
sec
uri
t
y
s
ystem
design
with
int
ern
et
of
thi
ng
s
(IoT
),
”
TEL
K
OMNIKA
Tele
c
omm
unic
ati
on
C
omput
ing
Elec
tr
on
ic
s
and
Contr
ol,
vol
.
17
,
no.
3,
p
p
.
1250
-
1255
,
2019,
doi
:
10
.
12
928/t
el
kom
nik
a.v17i3.
10167.
[10]
M.
Abina
y
a
and
R.
U.
Devi,
“
Inte
lligent
Veh
icle
Control
Us
ing
W
ire
le
ss
Emb
edde
d
S
y
stem
i
n
Tra
nsportation
S
y
stem
Based
On
GS
M
and
GP
S
Technol
og
y
,
”
I
nte
rnational
Jou
rnal
of
Comput
e
r
Sci
en
ce
and
M
obil
e
Computing
,
vol.
3
,
no
.
9
,
pp
.
244
-
258,
2014
.
[11]
D.
Hart
an
ti
,
R
.
N.
Aziza,
and
P.
C.
Sis
wipra
pti
n
i
,
“
Optimiza
t
ion
of
sm
art
tra
ffi
c
l
ight
s
to
pre
v
ent
tra
ffi
c
cong
esti
o
n
using
fuz
z
y
log
ic
,
”
TEL
KOMNIKA
Telecomm
unic
ati
on
Computing
Elec
tronic
s
and
Control
,
vol.
17,
no.
1
,
pp.
320
-
32
7
,
20
19,
doi
:
10
.
1292
8/t
el
kom
nik
a
.
v1
7i1.
10129.
[12]
P.
C.
Sis
wipra
pt
ini
,
W
.
H.
Mart
ono,
and
D.
Har
ta
nti,
“
Reduc
ing
a
conge
stion
wi
th
int
roduc
e
th
e
gre
ed
y
a
lgori
th
m
on
tra
ff
ic l
igh
t c
ontrol
,
”
J. P
hys
.
Conf.
S
er.
,
vol
.
9
74,
no
.
1
,
2018
,
doi
:10
.
1088/17
42
-
6596/974/
1/0
12013.
[13]
J.
Y.
Kim
,
H
.
L
e
e,
J.
Son,
and
J.
Park,
“
Sm
art
hom
e
web
of
objec
ts
-
base
d
IoT
m
a
nage
m
ent
m
odel
and
m
et
hods
for
hom
e
dat
a
m
ining,”
2015
17th
Asia
-
Pacific
Ne
t
work
Operation
s
and
Manage
m
ent
S
ymposium
(AP
NOMS)
,
2015,
pp.
327
-
331
,
doi
:
10.
1109
/APN
O
MS
.
2015.
7275448.
[14]
S.
G
hosh,
“Sm
a
rt
hom
es:
Archi
te
ct
ur
al
and
enginee
ring
design
i
m
per
at
ive
s
for
sm
art
ci
t
y
buil
d
i
ng
code
s,”
2018
Tech
.
for Smart
-
Cit
y
Ene
rgy
Se
c
urity
and
Powe
r
(
ICSESP
)
,
2018,
pp.
1
-
4,
doi: 10.1109/ICS
ESP
.
2018.
8376676
.
[15]
D.
E
.
Paul
and
A.
Vijay
a
n
,
“
Sm
art
Ene
rg
y
M
et
er
Us
ing
Andr
oid
Appli
ca
t
ion
And
Gs
m
Netw
ork,
”
I
nt.
J
.
En
g.
Comput.
Sc
i.
,
vo
l.
5
,
no
.
3
,
pp
.
16
058
-
16063,
201
6,
doi
:
10
.
18535
/i
jecs/v5i3
.
36.
[16]
B.
L
ashka
ri
,
Y.
Chen,
and
P.
M
usile
k,
“
Ene
rg
y
m
ana
gement
for
sm
art
hom
es
-
stat
e
of
th
e
ar
t,”
Appl
ie
d
Sc
ie
nc
e
s
(
Swit
zerland)
,
vol.
9
,
no
.
17
.
201
9,
doi
:
10
.
3390/app9173459.
[17]
M.
Sam
art
h
Pandit
,
M.
Snehamandhre
,
and
M.
Megha
nanicha
l
,
“
S
m
art
Ene
rg
y
Mete
r
using
int
ern
e
t
of
Thi
ngs
(IoT
),
”
V
ishwak
arm
a
Journal
of
Engi
ne
ering
R
ese
arch
,
vo
l. 1, no. 2,
pp.
222
-
229,
2017.
[18]
P.
C.
Sis
wipra
p
ti
ni,
R.
Nur
Az
iz
a
,
I
.
B
.
Sang
a
dji
,
I.
Indri
ant
o,
and
R
.
R
.
Sir
e
gar
,
“
Autom
at
e
d
Sm
art
Hom
e
Control
le
r
Bas
e
d
on
Adapti
ve
Li
nea
r
Neura
l
Network,
”
2019
7th
Inte
rnational
Confe
ren
ce
on
Control,
Me
chat
roni
cs
an
d
Aut
omat
ion
(
ICCMA)
,
2019,
p
p.
423
-
427
,
doi
: 10.1109/
ICCMA
46720.
2019.
898
8733
.
[19]
Y.
S.
Ta
k,
J.
Kim
,
and
E.
Hw
ang,
“
Hier
arc
hi
ca
l
quer
y
ing
sche
m
e
of
hum
an
m
oti
ons
for
s
m
art
ho
m
e
envi
ronm
ent
,
”
Eng.
Appl.
Art
if.
Intell.
,
vo
l. 25,
no.
7
,
pp
.
1301
-
1312,
2012
,
doi
:
10.
101
6/j.e
ng
ap
pai
.
2012
.
03.
020
.
[20]
S.
Z.
Re
y
h
ani
a
nd
M.
Mahda
vi,
“
Us
er
aut
hent
icati
on
using
neur
al
net
work
in
s
m
art
hom
e
net
works
,
”
Int.
J.
Sm
art
Hom
e
,
vol
.
1
,
no
.
2
,
pp
.
147
-
154
,
2007.
[21]
A.
Bhanot
and
S.
Bad
la
n
,
“
Sm
art
Hom
e
S
y
stem
Design
base
d
on
Artificial
Neur
a
l
Networks,”
Pr
oce
ed
ings
of
the
World
Congress
on
Engi
n
ee
ring
and
Computer
S
ci
en
ce 2011
,
vol
.
1
,
2011
.
[22]
R.
R.
A.
Siregar,
H.
Sikum
bang,
I.
B.
M.
San
gadj
i
,
and
Indri
ant
o,
“
KW
h
Me
te
r
Sm
art
Car
d
Model
Toke
n
For
El
e
ct
ri
ca
l
En
erg
y
Moni
tori
ng
,
”
MATEC
We
b
Co
nf.
,
vol
.
218
,
201
8,
doi
:
10
.
1051/
m
at
ec
conf
/2018
21803002.
[23]
A.
Q.
L
eon
,
“
5
al
gorit
hm
s
to
tr
ai
n
a
neur
al
n
etw
ork
,
”
Dat
a
Sci
enc
e
and
Ma
chine
L
ea
rning
Blo
g,
2020.
[Online]
Avail
ab
le
:
htt
ps:
//
ww
w.
neur
al
d
esigne
r.com
/bl
og/
5_al
gorit
hm
s_to
_tra
in_
a_ne
ura
l_
net
work
[24]
T.
Bolukb
asi,
J.
W
ang,
O.
Deke
l
,
and
V.
Sali
gr
ama,
“
Adapti
v
e
Ne
ura
l
Networks
fo
r
Eff
icient
Inf
erence
,
”
ICML'
17:
Proce
ed
ings o
f
t
he
34th
Int
ernational
Conf
ere
nce on
Mac
h
ine Lea
rning
,
vol
.
7
0
,
2
017,
pp
.
527
-
53
6
.
[25]
R.
Begg
and
R.
Hass
an,
“
Artifi
ci
a
l
Neura
l
Ne
t
works
in
S
m
art
Hom
es,
”
J.
C.
Augus
to,
C.
D.
Nugent,
Designi
ng
Smar
t
Hom
es
,
L
ondon,
UK
:
Spri
nger
,
2016
,
pp.
1
46
-
164,
doi
:
1
0
.
1
0
0
7
/
1
1
7
8
8
4
8
5
_
9
.
[26]
D.
T
.
Vie
t,
N.
H.
Hieu
,
and
N.
M.
Khoa
,
“
A
Method
for
Mo
nit
oring
Vol
ta
g
e
Disturbances
B
ase
d
on
Discr
ete
W
ave
le
t
Tr
ansform
and
Adapti
ve
Li
nea
r
Neura
l
Network
A Met
hod
for
Monitori
ng
Volta
ge
Disturbanc
es
Based
on
Discre
te,”
Int
.
R
ev
i
ew
of
E
le
c
trical
Eng
ine
er
ing
,
vol.
11
,
no
.
3
,
pp
.
314
-
322
,
2016
,
doi: 10.
15866
/i
r
ee
.
v11
i3.
8344
.
[27]
H.
Le
e
and
T.
Kim
,
“
Prototy
p
e
of
IoT
ena
ble
d
sm
art
fac
tor
y
,
”
ICIC
Ex
press
Lett
.
Part
B
Ap
pl.
,
vo
l
.
7,
no.
4,
pp.
955
-
960
,
20
16.
[28]
B
.
K
a
n
g
,
S
.
K
i
m
,
M
.
C
h
o
i
,
K
.
C
h
o
,
S
.
J
a
n
g
,
a
n
d
S
.
P
a
r
k
,
“
A
n
a
l
ys
i
s
o
f
T
yp
e
s
a
n
d
I
m
p
o
r
t
a
n
c
e
o
f
S
e
n
s
o
r
s
i
n
S
m
a
rt
H
o
m
e
S
e
r
v
i
c
e
s
,
”
2
0
1
6
I
E
E
E
1
8
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
H
i
g
h
P
e
r
f
o
r
m
a
n
c
e
C
o
m
p
u
t
i
n
g
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s
;
I
E
E
E
1
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
m
a
r
t
C
i
t
y
;
I
E
E
E
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
D
a
t
a
S
c
i
e
n
c
e
a
n
d
S
y
s
t
e
m
s
(
H
P
C
C
/
S
m
a
r
t
C
i
t
y
/
D
S
S
)
,
2
0
1
6
,
p
p
.
1
3
8
8
-
1
3
8
9
,
d
o
i
:
1
0
.
1
1
0
9
/
H
P
C
C
-
S
m
a
r
t
C
i
t
y
-
D
S
S
.
2
0
1
6
.
0
1
9
6
.
[29]
J.
C.
Augus
to
and
C.
D.
Nugent,
Designing
Smar
t
Hom
es
T
he
Rol
e
of
Artific
ia
l
Inte
l
li
g
en
ce
,
London
,
UK
:
Springer
,
2006
.
[30]
H
.
D
.
M
e
h
r
,
H
.
P
o
l
a
t
,
a
n
d
A
.
C
e
t
i
n
,
“
R
e
s
i
d
e
n
t
a
c
t
i
v
i
t
y
r
e
c
o
g
n
i
t
i
o
n
i
n
s
m
a
r
t
h
o
m
e
s
b
y
u
s
i
n
g
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
,
”
2
0
1
6
4
t
h
I
n
t
.
I
s
t
a
n
b
u
l
S
m
a
r
t
G
r
i
d
C
o
n
g
r
e
s
s
a
n
d
F
a
i
r
(
I
C
S
G
)
,
2
0
1
6
,
p
p
.
1
-
5
,
d
o
i
:
1
0
.
1
1
0
9
/
S
G
C
F
.
2
0
1
6
.
7
4
9
2
4
2
8
.
[3
1
]
D.
T
.
Vie
t,
N.
H.
Hieu
,
and
N.
M.
Khoa
,
“
A
Method
for
Mo
nit
oring
Vol
ta
g
e
Disturbances
B
ase
d
on
Discr
ete
W
ave
le
t
Tra
nsfo
rm
and
Adapti
v
e
Li
ne
ar
Neur
al
Network,
”
In
t.
R
ev
.
Elec
tr.
Eng
.
,
vol.
11
,
no.
3,
p
p
.
314
-
322
,
201
6,
doi:
10
.
15866/i
r
ee
.
v11
i3.
8344
.
[3
2
]
W
.
Ji
and
E.
H.
W
.
Chan,
“
Cri
tical
Fact
ors
Inf
lu
enc
ing
the
Adoption
of
Sm
art
H
om
e
Ene
rg
y
Tec
hnolog
y
in
Ch
in
a:
A Guangdong
Province Ca
se
Stu
d
y
,
”
Ene
rgi
es
,
v
ol.
12
,
no
.
21
,
20
19,
doi
:
10
.
3390
/e
n12214180
.
Evaluation Warning : The document was created with Spire.PDF for Python.