Indonesi
an
Journa
l
of El
ect
ri
cal Engineer
ing
an
d
Comp
ut
er
Scie
nce
Vo
l.
12
,
No.
3
,
Decem
ber
201
8
, p
p.
1171~
1178
IS
S
N: 25
02
-
4752, DO
I:
10
.11
591/ijeecs
.v1
2
.i
3
.pp
1171
-
1178
1171
Journ
al h
om
e
page
:
http:
//
ia
es
core.c
om/j
ourn
als/i
ndex.
ph
p/ij
eecs
A Desi
gn
Co
nsid
eration f
or Plan
ar In
verte
d Fract
al Antenn
a to
Mini
mize Length
-
Dep
endent Sp
ec
ific Abso
rption
Rat
e
Ak
r
amu
l
Haq
ue, Shei
kh Ali
mur Raz
i, Nu
r M
ohamm
ad
, Md. Sh
amsul
A
ri
fin,
Qua
z
i
Delw
ar
Hossain
Depa
rtment
o
f
E
le
c
tr
ical
and El
e
ct
roni
c
Eng
ineer
ing,
Ch
it
t
agong University
of
En
gine
er
ing
and
T
e
chnol
og
y
(CUE
T)
,
Chit
ta
gong
-
4349
,
Bang
la
d
esh
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
J
ul
1
,
201
8
Re
vised
A
ug
2
1
, 2
01
8
Accepte
d
Se
p
2
1
, 201
8
Thi
s
pap
er
pre
sents
a
num
erica
l
soluti
on
to
m
ini
m
iz
e
el
e
c
tromagnet
i
c
rad
iation
from
a
Plana
r
Inve
rt
e
d
Frac
tal
Anten
na
(PIF
A)
used
in
c
el
lu
la
r
phone.
The
PIF
A
is
sim
ula
te
d
using
a
sem
ic
onduct
or
subs
tra
t
e
havi
ng
a
die
l
ec
tr
ic
constant
of
3.
38
.
The
h
ei
ght
of
the
diele
ct
ri
c
subs
tra
t
e
is 0.
813
m
m
.
The
design
ed
an
te
nna
is
sim
ula
t
ed
at
a
bro
ad
ra
nge
of
m
ic
rowa
ve
fre
que
n
c
y
spec
trum
used
i
n
ce
l
lul
ar
communic
a
ti
on.
A
50
-
ohm
probe
of
0.
5
m
m
rad
ius
per
pendicula
r
to
the
ground
subs
tra
t
e
pl
ate
is
use
d
as
a
fe
edi
ng
m
edi
um
.
T
h
e
ant
enn
a
per
form
anc
e
is
ev
al
ua
ted
for
three
d
iffe
r
ent
l
engt
hs
ke
eping
al
l
other
par
amete
rs
const
ant
.
Sim
ula
ti
on
r
esult
s show
tha
t
the
in
te
nd
ed
PIF
A ha
ving a
le
ngth
of
20
m
m
ca
n
be
used
eff
ectivel
y
to
r
e
duce
th
e
Speci
f
i
c
Abs
orpti
on
Rat
e
(
SA
R)
of
rad
iation.
Moreo
ver
,
the
r
eflec
t
io
n
coe
fficie
n
t
was
found
to
be
m
ini
m
al
0.
156
9
at
20
m
m
ant
enn
a
l
engt
h
which
is
d
eterm
ine
d
b
y
cha
ra
cteri
sti
c
impedance
relati
on
.
The
r
efo
re
,
thi
s
inve
stigation
of
m
ini
m
iz
ing
the
r
adi
a
ti
on
absorpti
on
c
an
b
e
c
onsidere
d
dur
in
g
the
implement
at
ion
ph
ase
of
var
ious
cellular
an
te
nn
as
to
av
oid
rad
ia
t
ion
-
re
l
at
ed
he
al
th
ha
za
r
ds.
Ke
yw
or
d
s
:
Characte
risti
cs
i
m
ped
anc
e
PI
F
A
a
nten
na
Ra
diati
on
m
inim
iz
at
ion
Re
flect
ion
c
o
-
e
ff
ic
ie
nt
Sp
eci
fic a
bsor
ption rate
Copyright
©
201
8
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Nur
M
oh
am
m
a
d
,
Dep
a
rtm
ent o
f El
ect
ri
cal
an
d
Ele
ct
ro
nic
Eng
ineerin
g,
Chit
ta
gong Un
iversity
of E
ngineerin
g
a
nd T
echnolo
gy
(CU
ET),
Chit
ta
gong
-
43
49, Ba
ng
la
des
h
.
Em
a
il
:
nu
r.
m
oh
am
m
ad@
cuet
.ac.bd
1.
INTROD
U
CTION
The
Ele
ct
r
o
M
agn
et
ic
Ra
diati
on
(EMR)
,
in
the
lo
ng
r
un,
de
te
rior
at
es
hum
an
a
nd
pla
nts
he
al
th
if
th
e
rad
ia
ti
on
e
xce
eds
a
th
reshol
d
le
ve
l.
Hen
ce
,
it
is
re
quire
d
to
c
ontrol
th
e
EMR
poll
uti
on
to
ens
ure
healt
h
safety
[1
]
.
The
EMR
pro
paga
te
s
as
el
ect
ric
and
m
agn
et
ic
fiel
ds
thr
ough
a
ny
m
ediu
m
.
A
ccordin
g
t
o
[
2],
th
e
EMR
po
ll
utio
n
trigg
e
rs
t
he
dr
op
in
the
popu
la
ti
on
gro
wth
r
at
e
of
s
pa
rrows
an
d
bee
s,
in
exch
a
nge
of
w
hich
Colo
ny
Coll
ap
se
Disorde
r
(C
CD)
be
gins.
It h
as
bee
n
re
port
ed
that
unde
r
the
EMR
expos
ed
en
vir
on
m
ent,
bats
and
s
parrows
s
how
a
disincli
nation
in
thei
r
beh
a
vio
u
r
.
For
exam
ple,
in
th
e
US
A
,
dead
ly
crash
es
of
bir
ds
wit
h
com
m
un
ic
at
ion
m
ast
s
cause
t
he
extinct
io
n
of
m
il
li
on
s
of
bird
s
eac
h
ye
ar
[
2].
Furthe
rm
or
e
,
the
EMR
ha
m
per
s
br
ai
n
act
ivit
ie
s,
cause
s
hea
da
ches,
tri
gg
e
r
s
m
e
m
or
y
loss
as
well
as
can
cause
ca
ncer
i
n
so
m
e
cases.
It
m
a
y
init
ia
te
sensiti
ve
prob
le
m
s
li
k
e
congen
it
al
di
sabili
ty
and
low
sp
e
rm
cou
nt
[3
]
.
The
EMR
absorp
ti
on
de
pends
on
f
or
m
s
and
dim
ension
s
of
the
organ,
diele
ct
ric
pro
per
ti
es
of
res
pecti
ve
body
ti
ssu
e
.
T
he
a
bs
or
ption
a
m
ou
nt
is
ind
exe
d
by
the
te
rm
sp
ecif
ic
abs
orption
rate
(SAR).
T
he
SA
R
value
fo
r
a
rad
i
o
de
vice
is
diff
e
r
ent
f
or
diff
e
re
nt
body
par
ts.
T
he
S
A
R
is
def
ined
as
the
power,
W
abs
orbed
due
to
ex
posure
to
an
EMR
by
pe
r
kg
of
body
ti
ssue
[
4]
.
It
de
pe
nd
s
on
th
e
el
ect
ric
f
ie
ld
intensit
y,
fr
e
qu
e
ncy
of
s
upply,
pola
risa
ti
on
of
the
fiel
d
(f
a
r
and
nea
r
fiel
d)
chara
ct
erist
ic
s.
A
gain
,
no
t
only
the
fiel
ds
t
hat
are
em
it
ti
n
g
th
rou
gh
the
m
ob
il
e
is
resp
on
si
ble
for
the
SA
R
,
but
al
so
ref
le
ct
e
d
an
d
scat
te
red
EMR
fr
om
per
ip
her
al
de
vice
s
con
tri
bu
te
to
a
hig
h
value
of
the
SA
R
[
5].
Re
ga
rd
i
ng
t
his,
a
n
inv
e
sti
gation
w
as
co
nducted
by
Pack
an
d
Le
e,
w
ho
fig
ur
e
d
ou
t
th
e
S
AR
valu
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
12
, N
o.
3
,
Dece
m
ber
2
01
8
:
1171
–
1178
1172
for
seve
ral
siz
es
and
s
hap
es
of
he
ad
.
They
exam
ined
by
us
ing
fi
nite
difference
ti
m
e
dom
ai
n
m
e
tho
d
f
or
a
broa
d
ra
nge
of
fr
e
qu
e
ncy
rang
ing
from
8
35 t
o 1,765 M
Hz. [
6].
In
va
rio
us
res
earch
,
a
uthors
f
ound
the
s
ources
of
EMR
an
d
c
orres
pond
i
ng
he
al
th
i
ssu
es
[
7
-
11
].
Ak
i
hir
o
Tat
e
no
a
nd
K
ens
uk
e
in
t
heir
st
udy
fou
nd
that
m
ob
il
e
phon
es
an
d
ta
blet
co
m
pu
te
rs
are
t
wo
of
t
he
pr
im
e
so
ur
ces
of
the
EMR
poll
ution
[10].
The
po
wer
le
ve
l
of
the
EMR
so
urce
an
d
th
e
distance
bet
ween
th
e
so
urce
a
nd
rec
ei
ver
are
t
he
two
si
gn
i
ficant
va
riables
t
hat
def
ine
t
he
po
wer
densi
ty
at
a
sp
eci
fic
spo
t.
The
consi
ste
nt
gro
wth
of
the
m
ob
il
e
c
om
pan
y
as
well
as
it
s
us
e
r,
the
E
MR
po
ll
uti
on
has
i
nd
ee
d
be
com
e
a
sign
ific
a
nt c
oncern f
or hum
anity
an
d
the
e
nviro
nm
ent.
To
m
ini
m
ise
t
he
ad
ve
rse
ef
f
ect
s
of
t
he
ra
di
at
ion
,
sci
entis
ts
and
resea
rchers
are
pro
posing
dif
fer
e
nt
te
chn
iq
ues
a
nd
m
et
ho
ds
.
In
[7
]
,
Jal
al
J.
Ham
ad
Ami
n
pr
ese
nted
a
re
so
luti
on
to
im
plem
ent
op
ti
cal
fibr
e
com
m
un
ic
at
ion
,
rep
la
ci
ng
c
urre
nt
m
ic
ro
wav
e
com
m
un
ic
at
ion
,
wh
ic
h
is
costly
for
de
ve
lop
in
g
c
ount
ri
es.
Th
e
auth
or
al
s
o
s
ha
red
so
m
e
ti
ps
an
d
tric
ks
f
or
us
in
g
cel
l
-
ph
on
e
,
wh
ic
h
ca
n
re
duce
the
pro
bab
le
detri
m
ental
eff
ect
s
of the E
MR
, how
e
ver,
ignore
d
t
o
way
out to
reduce t
he
am
ou
nt
of ra
diate
d
EMR
.
Yu
s
uf
Abd
ulra
hm
an
Sam
bo
an
d
et
al
.
in [9] i
nv
est
igate
d
m
ulti
ple
asp
ect
s inclu
di
ng
S
AR s
hield
ing
,
Be
am
f
orm
ing
proces
ses
along
with the
instal
la
ti
on
of m
assive MIMO
t
hat c
an
le
sse
n
the
a
m
ou
nt of r
a
dia
ti
on
.
Re
centl
y,
a
Planar
Inver
te
d
F
r
act
al
An
te
nna
(P
I
FA)
is
re
ported
in
the
li
te
r
at
ur
e
to
c
on
t
rol
the
EMR
le
vel
[
12]
.
Th
e
PI
F
A
is
a
so
rt
of
li
near
Inver
te
d
F
-
A
nten
na,
a
m
on
op
ole
anten
na
us
ed
in
wir
el
ess
com
m
un
ic
at
ion
[
13]
.
It
is
m
anu
fact
ur
e
d
by
set
ti
ng
a
m
et
al
l
ic
patch
over
a
be
d
of
diele
ct
ric,
wh
ic
h
is
su
pp
or
te
d
on
a
gro
und
plate
.
The
fee
di
ng
m
et
hod
of
the
m
et
al
li
c
patch
is
done
ei
the
r
by
add
in
g
m
ic
ro
strip
transm
issi
on
li
ne
or
by
a
prob
e
.
Easy
fa
bri
cat
ion
,
l
ow
c
os
t
a
nd
sm
al
l
weig
h
a
re
the
reas
ons
beh
i
nd
th
e
m
assive u
si
ng
scenari
o of PI
F
A
in
the m
ob
il
e industry.
Re
searche
rs
a
r
e
inv
est
igati
ng
ways
to
reduc
e
the
am
ou
nt
of
EMR
ra
diati
on
f
ro
m
the
cel
lular
phones
by inc
orporati
ng
new d
esi
gn s
trat
egies of t
he
PI
F
A. S
om
e recen
t st
udie
s reveal
ed
t
hat th
e d
et
rim
ental
i
m
pacts
of
EMR
ra
diati
on
from
the
PI
FA
co
uld
be
m
ini
m
ise
d
by
red
uci
ng
t
he
SAR
[1
4
-
16
]
.
I
n
[
8],
Sang
-
il
K
w
ak
an
d
et
al
.
sh
owe
d
a
process
of
desig
ning
P
I
FA
with
the
arti
fici
al
m
agn
et
ic
condu
ct
or.
The
SA
R
re
du
ct
io
n
m
echan
ism
wa
s
introduc
ed
c
on
si
der
i
ng
a
wide
band
c
ode
div
isi
on
m
ul
ti
ple
acce
ss
ban
ds,
an
d
a
43
%
SAR
reducti
on
was
ob
ta
ine
d
at
the
centre
f
reque
nc
y.
The
sam
e
auth
or
s
in
thei
r
earli
er
pu
blica
ti
on
[11],
pro
po
s
ed
two
ty
pe
s
of
el
ect
ro
m
agn
et
ic
band
ga
pe
m
od
el
f
or
S
AR
re
du
ct
io
n
of
th
e
EMR
rad
ia
ti
on
.
H
oweve
r,
bot
h
the
stud
ie
s
disre
ga
rd
e
d
the
e
ff
ect
of
a
nten
na
le
ngth
on
the
S
A
R
reducti
on.
T
o
fill
this
ga
p
,
this
pa
per
p
r
op
os
es
a
PI
F
A,
an
d
it
s
le
ng
t
h
is
cal
ibr
at
ed
to
fin
d
the
dep
e
nd
e
nce
of
SA
R
on
the
le
ng
t
h.
T
he
SA
R
reducti
on
em
pirical
m
et
ho
d
is
discuss
e
d,
a
nd
t
he
SA
R
val
ues
a
re
m
easur
ed
f
or
th
ree
diff
e
r
ent
PI
F
A
le
ng
th,
an
d
the
ob
ta
ined
resu
lt
s
a
re c
ompare
d
.
So
m
e
of
t
he
o
t
her
va
riables
s
uch
as
shorti
ng
pin
wi
dth
,
pr
ob
e
fee
d
ra
dius,
the
dista
nce
betwee
n
t
he
sh
ort
in
g
pin
a
nd
the
pro
be,
the
substrat
e
m
at
erial
and
the
ins
ulator
he
igh
t
ha
ve
le
ss
sign
ific
a
nce
on
th
e
SA
R
[17]. T
he
reb
y i
m
pact o
f
these
var
ia
bles
is d
is
reg
a
rd
e
d i
n
this st
ud
y.
In
t
he
nex
t
sec
ti
on
m
et
ho
dolo
gy
f
or
t
his
re
se
arch
is
outl
ine
d
f
ollo
wed
by
the
res
ults
an
d
discu
ssio
n
in
S
ect
io
n 3. T
he
c
on
cl
ud
i
ng
rem
ark
s ar
e
pr
ov
i
ded in
S
ect
ion 4
at the e
nd.
2.
METHO
DOL
OGY
Table
1
e
nlist
s
f
reque
ncy
ba
nd
use
d
in
c
el
lular
com
m
un
ic
at
ion
.
As
s
een
in
the
ta
bl
e
a
bro
a
d
sp
ect
r
um
of
th
e
high
-
fr
e
quen
cy
el
ect
ro
m
ag
netic
wav
e
are
us
e
d
f
or
dif
fe
ren
t
gen
e
rati
on
of
c
omm
un
i
cat
ion
te
chnolo
gy
[
18]
.
High
e
r
the
frequ
e
ncy,
the
m
or
e
th
e
pro
ba
bili
ty
of
energy
bein
g
abs
orbe
d
by
body
ti
ssu
e
.
A
s
the
fr
e
quen
cy
of
rad
ia
ti
on
in
crease,
it
s
io
nizat
ion
po
wer
a
lso
increa
ses.
Ce
ll
ph
one
a
ntenn
a
s
are
desi
gn
e
d
t
o
transm
it
d
at
a at var
i
ou
s
freq
ue
ncies
dep
e
nding o
n
t
he
c
omm
un
ic
at
ion
g
e
ner
at
io
n.
Table
1.
T
he
fr
equ
e
ncy
range
of v
a
rio
us
tel
e
com
g
ener
at
io
n.
Gen
eration
Frequ
en
cy
Ran
g
e
2G
869
-
8
9
0
M
Hz (
C
DMA
),
93
5
-
9
6
0
MHz
(GS
M90
0
),
18
0
5
–
1
8
8
0
M
Hz (GS
M18
0
0
)
3G
2110
–
2
1
7
0
M
Hz
4G
8
0
0
M
Hz,
2
1
0
0
-
2
6
0
0
M
Hz
Du
e
t
o
res
our
ce
lim
it
at
ion
,
the
w
ork
is
carried
out
by
the
Adva
nce
d
Desi
gn
Syst
e
m
(A
DS)
si
m
ulati
on
platfo
rm
,
becau
se
of
it
s
accu
racy
and
easy
-
to
-
use
interface.
Fig
ur
e
1
s
hows
th
e
diag
ram
of
a
PIF
A
wh
e
re
L1
an
d
L2
de
no
te
the
le
ng
th
a
nd
wi
dt
h
of
the
PIFA
resp
ect
ively
,
W
is
the
widt
h
of
the
shortin
g
pin
s
,
D
is
fo
r
the
se
par
at
io
n
of
fee
d
an
d
shortin
g
pin
,
an
d
h
is
t
he
distance
of
PI
F
A
ab
ov
e
th
e
gr
ou
nd
plane
.
The
anten
na
hei
gh
t
,
h
is
fill
ed
with
an
ai
r
substr
at
e
(
εr
=
1
.
0).
The
sho
rting
pl
at
e
con
sist
s
of
a
ver
ti
cal
condu
ct
in
g
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
A D
esi
gn Co
nsi
dera
ti
on f
or
P
lanar
In
ve
rte
d Fra
ct
al
Ante
nna
to
Mini
mize
Length…
(
Akr
amul
Haq
ue
)
1173
strip,
a
nd
it
is
us
e
d
not
only
t
o
co
nnect
betw
een
the
patc
h
a
nd
gro
und
but
al
so
to
sup
port
the
whole
ant
enn
a
.
The 5
0Ω
c
oax
i
al
p
r
obe
has
a
r
adius
of
0.5m
m
an
d
is fe
d
in
the ce
ntre
li
ne
of the
r
ect
a
ngular pat
ch
.
Figure
1. Plane
r
in
ver
te
d
F
an
t
enn
a
w
it
h a s
hortin
g plat
e
To
reduce
the
SA
R,
the
PIF
A
le
ngth
L
1
ha
s
bee
n
m
od
if
ie
d.
Hen
c
e,
t
hi
s
pa
per
rec
om
m
end
s
the
m
et
ho
d
of
le
ngth
re
duct
ion
to
m
ini
m
ise
SA
R
.
Th
e
desig
ne
d
P
IFA
has
be
en
pr
i
nted
over
a
Ro
ge
rs'
boar
d
a
nd
fed
b
y a 5
0
-
ohm
p
ro
be of
0.5
m
m
r
adius.
Th
e p
robe w
a
s p
e
rp
e
ndic
ul
ar to
the g
r
ound p
la
t
e. Tab
le
2
sho
ws
th
e
desig
n
s
pecific
at
ion
s
of
the
P
I
FA
.
T
he
Ro
ge
r
s
boar
d
i
nd
ic
at
ed
in
A
DS
as
RO4003C
has
a
diele
ct
ric
c
onsta
nt,
ε
of
3.38.
The
thickne
ss
of
th
e
copper
plate
is
0.
5
m
m
.
The
area
of
the
ant
enn
a
is
L
1×
L
2
,
a
nd
th
e
area
of
th
e
gro
und
plate
is
Lg×
Wg
.
T
he
ai
r
(ε
=
1.0
0)
i
s
consi
der
e
d
as
the
insu
l
at
io
n
la
ye
r.
A
sho
rting
st
rip
co
nne
ct
s
the
patch
a
nte
nn
a
to
the
gr
ound as
w
el
l as s
uppo
r
ts t
he whole
a
nt
enn
a.
Table
2.
Propo
sed
P
IFA a
nte
nn
a
s
pecifica
ti
on
s
SL.
No
Para
m
eters
Sp
ecif
icatio
n
s
1.
Sh
ap
e
Rectan
g
u
lar
2.
Frequ
en
cy
of
op
eration
1
.39
-
1
.5GHz
3.
Dielectri
c con
stan
t of
the su
b
strate
3
.38
(
Ro
g
ers RO4
0
0
3
C)
4.
Heig
h
t
o
f
dielectric sub
strate
0
.81
3
m
m
5.
Feed
in
g
M
eth
o
d
p
rob
e f
eed
The
operati
ng
fr
e
quency
of
the
P
IFA
c
an
be
rou
gh
l
y
determ
ined
by
the
f
ollo
wing
set
of
equ
at
io
ns (
1
-
3)
[
17
]
.
1
=
4
=
1
4
×
√
(1)
2
=
4
√
2
+
1
(2)
=
4
(
1
+
2
)
(3)
Her
e
,
λ
d
is
t
he
wav
el
e
ng
t
h
inside
t
he
subs
trat
e,
εr
is
the
relat
ive
pe
rm
i
tt
ivit
y,
L1
and
L2
are
t
he
le
ng
th
and
widt
h
of t
he
P
IFA
resp
ect
ively
,
a
nd
c
is t
he velo
ci
ty
o
f
the
EM
wav
e
.
The
fo
ll
ow
i
ng
set
of
e
quat
ion
s
pro
vid
e
s
the
detai
l
in
vestigat
io
n
on
the
SA
R
f
or
a
nten
nas
us
e
d
in
W
irel
e
ss
Com
m
un
ic
a
ti
on
[
19
]
.
T
he
SA
R
has
a
n
inv
e
rse
propo
rtion
al
relat
io
n
with
the
ret
urn
lo
s
s
wh
ic
h
is
de
fin
ed
as
the
lo
gari
th
m
of
the
rati
o
of
the
incide
nt
powe
r,
Pi
to
the
re
flect
ed
powe
r,
Pr
as
show
n
i
n
Eq
uation
(
4).
(
)
=
10
log
10
(4
)
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
12
, N
o.
3
,
Dece
m
ber
2
01
8
:
1171
–
1178
1174
The
le
sser
the
ref
le
ct
ed
po
we
r
the
higher
is
the
return
loss,
and
the
lowe
r
is
the
SA
R.
Th
e
ref
le
ct
io
n
coeffic
ie
nt
al
so
carries
ex
cel
le
nt
inform
at
io
n
ab
ou
t
the
S
AR
value
.
It
is
the
rati
o
of
th
e
ref
le
ct
ed
volt
age
to
the
inci
den
t
vo
lt
age,
w
hic
h
is
al
so
relat
ed
t
o
the
a
nten
na
i
m
ped
ance
ZL
and
cha
racteri
s
t
ic
s
i
m
ped
ance
Zo
as
giv
e
n by E
qu
at
ion
(
5)
.
|
|
=
=
−
0
+
0
(5
)
The
S
AR
in
te
rm
s
of
the
el
ec
tric
al
condu
ct
i
vity
(S
/m
)
of
ti
ssu
e
(σ
),
t
he
densi
ty
of
ti
ss
ue
in
Kg
/m
3
(
)
an
d
the
inte
nsi
ty
o
f
the
elec
tric
f
ie
ld
(E
)
is
pr
ese
nted
b
y
E
q
uatio
n
(
6)
.
SAR
=
|
|
2
2
(6
)
The
norm
al
iz
e
d
S
AR
is
giv
e
n
by E
q. (
7).
.
=
(
1
−
|
|
2
)
(
1
−
ɳ
.
2
)
(7
)
Her
e
,
S
ARorig
inal
m
eans
the
SA
R
value
t
ha
t
is
cal
culat
ed
us
in
g
E
q
u
at
ion
(6)
m
or
eo
ve
r,
ɳr
a
d
is
t
he
rad
ia
ti
on e
ff
ic
ie
ncy.
3.
RESU
LT
S
AND DI
SCUS
S
ION
This
sect
io
n
presents
the
sim
ulati
on
res
ults
to
analy
se
t
he
perform
ance
of
the
desi
gn
e
d
PI
F
A.
The
le
ng
th
L
1
is
c
hange
d
from
30
m
m
to
20
m
m
with
a
5
m
m
equ
al
dec
re
m
ent
kee
ping
t
he
widt
h
L
2
a
nd
oth
e
r
par
am
et
ers
co
nst
ant.
Se
ve
ral
te
st
tria
ls
ha
ve
been
perform
e
d
to
ac
hieve
an
en
ha
nced
f
requen
cy
m
at
ch
at
ea
c
h
ste
p.
S
horte
nin
g
the
patch
a
nten
na
to
the
gro
und
helps
to
m
at
ch
the
im
ped
ance.
Howev
e
r,
it
m
ak
es
t
he
rad
ia
ti
on
im
pe
dan
ce
reacti
ve
.
The
reb
y,
a
c
apacit
ive
to
p
loadi
ng
is
us
e
d
to
m
ini
m
ise
t
his
ef
fect.
Fig
ur
e
2
sh
ows
t
he
2D
and
3D
vie
w
of
the
de
sig
ned
PI
F
A
ante
nn
a
for
30
m
m
le
ng
th.
For
bre
vity
,
si
m
il
ar
views
for
oth
e
r
tw
o
le
ng
t
hs
ha
ve no
t
be
en pr
ovide
d
in
this pa
per.
(a)
(b)
Figure
2. The
desig
ne
d
P
IFA
an
te
nna
f
or
L1
=
30m
m
,
(a)
2D
view,
(
b) 3D
v
ie
w
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
A D
esi
gn Co
nsi
dera
ti
on f
or
P
lanar
In
ve
rte
d Fra
ct
al
Ante
nna
to
Mini
mize
Length…
(
Akr
amul
Haq
ue
)
1175
(a)
(b)
Figure
3. The
re
tur
n
los
s
prof
i
le
o
f
PIFA
an
te
nn
a
for
L
1
=
30 m
m
(
a) Mag
ni
tud
e,
and
(
b)
phase a
ngle
sp
ect
r
um
Figure
3
prese
nts
the
sim
ulatio
n
resu
lt
.
T
he
m
agn
it
ud
e
a
nd
the
ph
a
se
an
gle
sp
ect
r
um
s
of
retu
rn
l
os
s
for
a
broa
d
range
of
fr
e
quenc
y
are
sh
own
in
Figu
r
e
3(
a
)
an
d
3(b
)
res
pecti
vely
.
The
act
ua
l
m
agn
it
ud
e
of
th
e
return
loss
is
the
ne
gative
of
the
Y
-
axis
va
lues
of
the
m
a
gn
it
ude
s
pectr
um
.
As
sh
own
in
Figu
r
e
3(
a
),
the
m
agn
it
ud
e
of
the
return
loss
is
fo
und
the
m
axi
m
u
m
of
14.43
6d
B
at
1.2
16
GH
z
.
Fig
ur
e
3(b)
s
how
s
the
corres
pondin
g
the
phase
s
hift
of
ar
ound
-
20
de
gr
ee
.
T
he
l
ine
im
ped
ance
ZL
is
cal
c
ula
te
d
f
ro
m
the
S
m
it
h
char
t
s
how
n
i
n Fi
gure
4. The
re
flect
ion
c
oe
ff
i
ci
ent is cal
cu
la
te
d
to
be
|
Γ|=0.175
3 usin
g
E
quat
ion
(
5).
Figure
4.
The
Sm
it
h
char
t
of
PI
F
A
a
nten
na f
or L1 =
30 m
m
(a)
(b)
Figure
5. The
re
tur
n
los
s
prof
i
le
o
f
PIFA
an
te
nn
a
for
L
1
=
25 m
m
(
a) Mag
ni
tud
e,
and
(
b)
phase a
ngle
sp
ect
r
um
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
12
, N
o.
3
,
Dece
m
ber
2
01
8
:
1171
–
1178
1176
The
m
agn
it
ude
and
ph
ase
a
ng
le
s
pectr
ums
of
ret
urn
los
s
for
the
25
m
m
le
ng
th
of
the
PI
F
A
are
sh
ow
n
in
Fig
ure
5.
As
sho
w
n
in
Fig
ur
e
5(a),
the
m
agn
it
ud
e
of
the
return
loss
is
f
ound
the
m
axi
m
um
of
15.56
6
dB
at
1.
314
GH
z
.
Fig
ur
e
5(
b)
sh
ows
the
cor
res
pondin
g
phase
sh
i
ft
of
ar
ound
zero
de
grees.
T
he
li
ne
i
m
ped
ance
ZL
is
ob
ta
ine
d
f
r
om
the
S
m
ith
cha
rt
show
n
in
Fig
ure
6,
an
d
the
co
rresp
onding
re
flect
io
n
coeffic
ie
nt is
f
ound to
b
e
|Γ|
= 0
.
1661.
Figure
6.
T
he S
m
it
h
char
t
of P
IF
A
an
te
nna f
or
L
1
=
25 m
m
(a)
(b)
Figure
7. The
re
tur
n
los
s
prof
i
le
o
f
PIFA
an
te
nn
a
for
L
1
=
20 m
m
(
a) Mag
ni
tud
e,
and
(
b)
phase a
ngle
sp
ect
r
um
Figure
8. The
Sm
it
h
char
t
of
PI
F
A
a
nten
na f
or L
1
=
20 m
m
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
A D
esi
gn Co
nsi
dera
ti
on f
or
P
lanar
In
ve
rte
d Fra
ct
al
Ante
nna
to
Mini
mize
Length…
(
Akr
amul
Haq
ue
)
1177
Si
m
il
arly
,
the
m
agn
it
ud
e
an
d
phase
a
ng
le
s
pectr
um
s
of
th
e
retu
rn
l
os
s
f
or
t
he
20
m
m
le
ng
th
of
th
e
PI
F
A
are
sho
wn
in
Fi
gure
7.
Acc
ordi
ng
t
o
the
Fig
ur
e
7(a),
the
m
agn
i
tud
e
of
the
ret
urn
loss
is
f
ound
the
m
axi
m
u
m
of
15.
686
dB
at
1.437 G
Hz
.
Fig
ure
7(b
)
s
hows
t
he
co
rr
e
spo
nd
i
ng
phase
s
hift
of
a
rou
nd
-
20
degree
.
The
li
ne
im
ped
ance
ZL
is
obt
ai
ned
f
r
om
the
Sm
it
h
char
t
show
n
in
Fi
gure
8,
a
nd
t
he
co
rresp
onding
re
fl
ect
ion
coeffic
ie
nt is
f
ound to
b
e
|Γ|
= 0
.
1569.
In
s
umm
ary,
the
return
loss
is
fo
un
d
the
m
ax
i
m
u
m
,
and
the
ref
le
ct
io
n
coeffic
ie
nt
is
fo
und
m
ini
m
u
m
wh
e
n
t
he
a
nten
na
le
ngth
is
20
m
m
.
The
m
easur
em
ent
of
anten
na
i
m
ped
a
nce
by
var
yi
ng
the
le
ng
t
h
of
PI
F
A
is
li
ste
d
in
Table
3.
As
seen
i
n
the
ta
ble,
the
ante
nn
a
im
ped
ance
appro
ac
hes
to
prob
e
im
ped
a
nce
as
the
ante
nn
a
le
n
gt
h
dec
reases
pro
vid
in
g
a
bet
te
r
i
m
ped
a
nce
m
at
ching
at
20
m
m
.
This
sig
ni
fies
the
r
obust
ness
of
the
des
ign
e
d
P
IFA
f
or
the
le
ng
th
abo
ve.
Table
3.
Mea
s
ur
em
ent o
f
an
t
enn
a
im
ped
anc
e b
y
var
yi
ng t
he
leng
t
h of PI
F
A
Leng
th
of
PI
FA
an
ten
n
a (
m
m
)
An
ten
n
a
I
m
p
ed
an
ce
(
Z
L
)
1
30
7
1
.25
2
25
6
9
.92
3
20
6
8
.62
In
su
m
m
ary,
the
return
l
os
s
increase
s
as
t
he
a
nten
na
le
ngth
L1
dec
rea
ses.
Im
pr
ov
e
d
retu
rn
loss
sign
ifie
s
t
he
dim
inishing
of
ref
le
ct
ed
powe
r
acco
rd
i
ng
t
o
the
Eq
u
at
io
n
(4).
T
his
m
ea
ns
that,
for
a
certai
n
a
m
ou
nt
of
rec
ei
ved
powe
r,
t
he
re
quire
d
a
m
ou
nt
of
inci
den
t
powe
r
is
reduce
d
,
w
hich
in
t
urn
le
ss
ens
the
a
m
ou
nt
of
tra
nsm
it
te
d
po
we
r. Ther
e
by the el
ect
ric fiel
d
is r
edu
ce
d.
C
onse
qu
e
ntly
, th
e S
AR r
e
du
ces ac
cordin
g
to E
q
u
at
io
n
(
6).
Fr
om
an
oth
e
r
po
i
nt o
f
vie
w,
i
t
is
obse
rv
e
d
t
ha
t
the r
eflect
io
n
c
oeffici
ent dec
reases w
it
h
the r
ed
uctio
n
in
anten
na
le
ngth
w
hich
im
plies
that,
the
refl
ect
ed
vo
lt
age
al
so
reduces
.
As
the
re
flect
ed
volt
age
re
duces,
the
incident
vo
lt
a
ge
requirem
ent
is
reduced
f
or
a
f
ixe
d
am
ou
nt
of
the
in
duce
d
volt
age
in
t
he
receiver
a
nte
nn
a
accor
ding
to
Eq
u
at
ion
(5).
As
a
res
ult,
the
el
ect
ric
fiel
d
intensit
y
of
a
nten
na
de
creases.
As
seen
i
n
Eq
u
at
ion
(6),
f
or
a
set
value
of
ti
ss
ue
co
nd
uctivit
y
and
de
ns
it
y,
the
S
AR
al
so
dec
reases
.
Eq
u
at
io
n
(
7)
sh
ow
s
t
hat the
norm
alized
S
AR
dim
i
nish
es
as
well
.
Th
ough
the
ph
ase
dif
fer
e
nce
betwee
n
inci
de
nt
an
d
ref
le
ct
e
d
wa
ves
cha
ng
es,
ho
wev
e
r,
t
his
does
not
carry
on
any
sign
i
ficant
i
m
pact
on
SA
R
value.
T
he
Sm
i
th
Charts
sho
w
th
at
sign
ific
ant
c
hanges
in
the
l
eng
t
h
do not a
f
fect t
he
an
te
nna im
ped
ance
noti
ceab
ly
.
4.
CONCL
US
I
O
N
In
this
pap
e
r,
we
ha
ve
analy
zed
the
le
ngth
dep
e
ndent
ret
urn
los
s,
re
flect
ion
c
oeffici
ent
and
s
pecific
abs
orption
rate
of
EMR
by
va
ryi
ng
t
he
P
IFA
le
ngth
.
T
he
PI
F
A
is
desig
ne
d
a
nd
sim
ulated
in
A
DS
pla
tfor
m
.
The
stu
dy
revea
le
d
that
the
return
loss
inc
reases
w
hile
t
he
ref
le
ct
io
n
coeffic
ie
nt
decre
ases,
as
the
anten
na
le
ng
th
, decrea
s
es. A
ls
o, a
decre
ase in t
he
a
ntenn
a
len
gth
yi
el
ds
a
d
ec
rease i
n
the
S
AR
valu
e.
The
key
lim
it
a
t
ion
of this len
gt
h
re
du
ct
io
n
is
that
the wo
rk
i
ng fre
qu
e
ncy of
the an
te
nn
a
ch
ang
e
s w
it
h
decr
easi
ng
le
ngth
w
hich
s
hould
be
ta
ke
n
into
acco
unt
duri
ng
the
de
sign
process.
T
his
pro
blem
c
an
be
m
ini
m
iz
ed
by
two
m
et
ho
ds
:
on
e
is
to
i
ncr
e
ase
the
ai
r
ga
p
betwee
n
the
gro
und
plate
a
nd
a
nten
na,
a
nd
the
sec
ond o
ne
is t
o
c
hange t
he
m
at
erial
o
f
ins
ulati
on
lay
er.
REFERE
NCE
S
[1]
.
Nehru
V.
"
Glob
al
W
ire
le
ss
Spiderwe
b:
The
Invis
ibl
e
Thr
ea
t
Pos
ed
b
y
W
ire
le
ss
Radi
a
ti
on
"
.
1st
ed.
Gur
gaon,
India
:
Partridge
Publ
is
hing,
Ind
ia
,
201
6.
[2]
.
Alber
t
M.
Manv
il
le
II
.
"
Im
pac
ts
to
Birds
and
Bats
Due
to
Coll
isions
and
El
ectroc
uti
ons
from
So
me
Ta
l
l
Struct
ur
es
in
the
Unite
d
Sta
te
s:
W
ire
s,
Tow
ers,
Turb
ine
s,
an
d
Solar
Arra
y
s
—
Stat
e
of
the
Art
in
Address
ing
the
Problems
"
.
I
n:
Proble
mati
c
Wildlif
e:
A
Cros
s
-
Discipl
inary
Ap
proach
.
Angelic
i
F.
Arlingt
on
,
VA,
US
A:
Spri
nger
,
Ch
am;
2016:
415
–
42.
DO
I:
10
.
1007/978
-
3
-
319
-
22246
-
2.
[3]
.
Anon.
"
EMF
Hea
lt
h
Eff
e
ct
s
"
.
,
2018
.
[Onlin
e]
.
Avail
ab
le:
htt
ps://
ww
w.safes
pac
epr
ot
ec
t
ion.com
/e
m
f
-
hea
lt
h
-
risks/emf
-
hea
lt
h
-
eff
ects/c
el
l
-
tower
s/.
[A
cc
essed:
Mar.
21
,
2018
]
.
[4]
.
Choi
D,
Shin
C,
Kim
N,
Shin
H
.
"
Design
and
S
AR
Anal
y
sis
of
Broadba
nd
PIF
A
with
Tr
ipl
e
B
and
"
.
Progress
I
n
El
e
ct
rom
agnet
i
c
s
Re
search
Symposium
.
Hangz
hou,
Chin
a
.
PIERS
,
onl
ine
2005;
1:
290
–
3.
DO
I:
10.
2529/PIERS0412122116
50.
[5]
.
Hoque
AK
M
F,
Hos
sain
MS
,
M
oll
ah
AS
,
Akram
uzz
aman
M.
"
A
Study
on
Spe
ci
fi
c
Abs
orpti
on
Rat
e
(SA
R)
Du
e
to
Non
-
Ioniz
ing
R
adi
a
ti
on
from
W
ire
le
ss
/T
e
leco
m
m
unic
at
ion
in
Bangl
ad
esh
"
.
A
meric
an
Journa
l
of
Ph
ysic
s
an
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.
12
, N
o.
3
,
Dece
m
ber
2
01
8
:
1171
–
1178
1178
Appl
ic
a
ti
ons.
20
13;
1(3):
104
–
10
.
DO
I:
10
.
11648
/j
.
aj
pa
.
20130103
.
18.
[6]
.
Le
e
A,
Choi
H
,
et
al.
"
Eff
e
ct
of
Hea
d
Size
for
M
obil
e
Phone
Exp
osure
on
EM
Ab
sorption
"
.
Asia
-
P
aci
fic
Mi
crowav
e
Confe
renc
e
(
APMC)
.
Ta
ip
ei
,
T
aiw
an.
I
EE
E
.
2001
;
384
–
7.
DO
I:
1
0.
1109/APM
C.
2001.
985668.
[7]
.
Jala
l
J
.
Ham
ad Am
ee
n
.
"
Reduct
ion
of
Ce
ll
Ph
on
e
Elec
tromagneti
c
Radiat
ion
Eff
e
ct
on
Hum
an
Bo
d
y
"
.
Inte
rnat
ion
al
Journal
of
Scien
ce
s.
2014
;
3(Mar
ch)
:
19
–
23
.
[8]
.
Kw
ak
S
il
,
Sim
D,
Kw
on
JH
,
Yoon
YJ
.
"
Design
of
PIF
A
W
it
hMeta
m
ateri
a
ls
for
Bod
y
-
SA
R
Red
uct
ion
inWea
rab
le
Applic
a
ti
ons
"
.
IEE
E
Tr
ansact
ions
on
El
e
ctr
omagneti
c
Co
mpatibi
lity
.
20
17;
59(1):
29
7
–
300.
DO
I:
10.
1109/TEMC.
2016.
2593493.
[9]
.
Sam
bo
Y
A,
Heli
ot
F,
Im
ran
M
A.
"
A
survey
a
nd
tut
oria
l
of
e
l
ec
tromagne
ti
c
r
adi
a
ti
on
and
red
uct
ion
in
m
obile
comm
unic
at
ion
sy
st
ems
"
.
IEE
E
Comm
unications
Surve
ys
and
Tutor
ial
s
.
2015;
17(2):
790
–
802.
DO
I:
10.
1109/COMS
T.
2014.
2364136
.
[10]
.
Ta
t
eno
A,
T
anaka
K,
e
t
a
l.
"
C
om
par
ison
of
SA
R
in
Hum
an
Bod
y
Ra
diated
from
Mobile
Phone
and
Ta
bl
e
t
Com
pute
r".
Inter
nati
onal
Sympo
sium
on
El
e
ct
ro
magneti
c
Compatibi
l
it
y
,
Tok
y
o
(EMC’14/T
ok
y
o
).
Tok
y
o
,
Jap
an:
IEE
E
.
2014
;
186
–
9.
[11]
.
Kw
ak
S,
Si
m
D
U,
Kw
on
JH
,
Y
un
JH
.
"
SA
R
re
duct
ion
of
PIF
A
with
EBG
stru
ct
ure
s
for
m
obil
e
appl
i
ca
t
ions
"
.
IEI
CE
Tr
ansactions
on
Comm
unic
ati
ons.
2009;
E
92
-
B(11):
3550
–
3.
DO
I:
10
.
1587
/t
ran
scom
.
E92
.
B
.
3550.
[12]
.
Chat
tha
HT
,
Huang
Y,
Ishfaq
M
K,
Bo
y
es
SJ
.
"
A
Com
pre
hensive
Para
m
et
ric
Study
of
Plana
r
Inve
rte
d
-
F
Antenna
"
.
Wirel
ess E
ng
ineering
and
Te
chn
ology
.
2012;
03(
01):
1
–
11.
DO
I: 10.4236/
wet
.
201
2.
31001.
[13]
.
Douglas
H.
W
ern
er,
Ra
j
Mitt
r
a.
"
Frac
ta
l
-
Shap
ed
Antenna
s
"
.
In:
F
rontie
rs
in
El
e
ctr
omagneti
cs.
1st
ed.
New
Je
rs
ey,
USA
.
W
ile
y
-
IEEE
Press
.
2000:
81
4.
DO
I:
10
.
1109
/9780470544686.
ch2.
[14]
.
Yous
af
J,
Jung H,
Kim
K,
Nah
W
.
"
Design,
Anal
y
s
is,
and Equi
val
en
t
Circ
u
it
M
odel
ing
of
Dua
l
-
Band
PIF
A Us
in
g
a
Stub
for
Perfor
m
anc
e
Enha
nc
e
m
ent
"
.
Journal
of
El
ectrom
agnet
ic
Engi
n
ee
ring
and
Sci
ence.
20
16;
16(3):
169
–
81.
DO
I:
10.
5515/J
KIEE
S.2016.
16
.
3.
169.
[15]
.
Rat
hwa
RR
,
Ko
ta
k
NA
,
Ta
rp
ar
a
N.
"
Design
of
a
wideba
nd
Plana
r
Inve
rt
ed
F
-
Antenna
(
PIFA
)
for
W
ire
le
ss
Com
m
unic
at
ion
Devic
es
"
.
In
te
rn
ati
onal
Journ
al
of
Scientific
Re
s
earc
h
in
S
ci
en
c
e,
Engi
ne
ering
and
Technol
ogy
.
2018;
4(2):
128
–
32.
[16]
.
Rana
S,
Tha
ku
r
A,
e
t
a
l.
"
A
wideba
nd
planar
inv
ert
ed
F
antenna
for
wire
le
ss
comm
unic
a
ti
on
d
evi
c
es
"
.
Inte
rnational
C
onfe
renc
e
on
Adv
anc
es
in
C
omputing,
Com
municat
ion
,
&
Au
tomati
on
(
ICACCA
)
(
Spring
)
.
Dehradun,
India
:
IEEE
.
2016;
DO
I:
10.
1109/IC
ACCA
.
2016.
7578875.
[17]
.
S.
Pala
niv
el
Ra
nja
n,
Dr
R.
Su
kane
sh,
S.
Vij
a
y
pra
s
at
h.
"
Perfo
rm
anc
e
Evalua
tion
of
Mobile
Phone
Radiati
o
n
Minim
iz
at
ion
t
hrough
Charact
eri
sti
c
Im
peda
n
ce
Me
asure
m
en
t
for
He
al
th
-
c
are
Appl
ic
a
ti
on
s
"
.
2012
IEEE
Inte
rnational
Confe
renc
e
o
n
Adv
an
ce
d
Comm
unic
ation
Control
and
Comp
uti
ng
Te
chnologies
(
ICACCCT
)
.
Ramanat
hapur
am,
I
ndia
:
IEEE. 201
2;
1
–
4.
DO
I:
10
.
1109/ICACCCT.
2012.
6320722
.
[18]
.
W
ong
KL,
Le
e
GY
,
Chiou
T
W
.
"
A
low
-
profil
e
pla
n
a
r
m
onopole
ant
enn
a
for
m
ult
iba
nd
oper
at
ion
of
m
obil
e
handse
ts
"
.
IE
EE
Tr
ansactions
on
Antennas
and
Propagati
on.
2003;
51(1):
121
–
5.
D
OI:
10.
1109/T
AP
.
20
03.
809044.
[19]
.
Shi
Y,
Sun
H,
Li
ang
C
-
H
.
"
SA
R
Stud
y
of
Antenna
s
in
W
ireless
Com
m
unic
at
ion
Te
rm
inals
"
.
Mic
rowav
e
an
d
Optic
al
Technology
Let
te
rs
.
201
4;
56(10):
2361
–
5.
DO
I:
10
.
1002
/mop.
Evaluation Warning : The document was created with Spire.PDF for Python.