Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
V
o
l.
6, N
o
. 4
,
A
ugu
st
2016
, pp
. 16
85
~
1
694
I
S
SN
: 208
8-8
7
0
8
,
D
O
I
:
10.115
91
/ij
ece.v6
i
4.1
021
0
1
685
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJECE
Feasibility of Power Control for
Multi-Channel User in Inter-
Femtocell Network
An
ggun Fi
trian Isn
awati
1,2
,
R
i
s
a
nu
ri
H
i
day
a
t
1
, Selo
Sulistyo
1
,
I W
a
yan Mustik
a
1
1
Departement of
Electr
i
cal
and I
n
forma
tion Tech
nolog
y
,
Gad
j
ah
Mada Univer
sity (UGM), Yog
y
akarta, Indon
esia
2
Telkom School
of Telematics (S
T3
Telkom), Pur
w
okerto,
Indonesia
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Feb 17, 2016
Rev
i
sed
Mar
27
, 20
16
Accepted Apr 14, 2016
The importance of power
contr
o
l feasibil
ity
i
s
c
l
ose
l
y
re
la
t
e
d to t
h
e
di
rect
implementation of femtocell technolog
y
,
which
is related to interfer
e
nce
between
us
ers
a
nd to th
e opt
im
al power us
ag
e to
trans
m
it which
at th
e
end i
t
can save power
and introduce
a long-la
sting b
a
tter
y
.
Th
e power contro
l
feasibility
stud
y done in this paper
is focused
on a centralized femtocell
network
with a m
u
lti-chann
e
l
us
er. The
r
e
sear
ch
m
e
thod is based on the use
of feasible solution algorithms by
obser
ving
the
output of power
vector
that
should be a no
n-negative number.
A non neg
a
tiv
e number in
dicates th
at
power can
be
i
m
p
lem
e
nted.
Th
e res
u
lt
of th
is
s
t
ud
y indi
cat
es
t
h
at a
ll us
ers
can re
ach th
e specif
i
ed SINR tar
g
ets. SINR users will incre
a
se
when ther
e
are ad
dition
a
l
channels for user
groups
. Moreo
v
er, th
e
averag
e power will
also decrease while in
creasing the am
ount of provided chann
e
ls. The greater
number of users
in a user group
,
the less SINR can be ob
tained.
Keyword:
Feasib
ility
Femto
cell
Mu
lti-ch
ann
e
l
Power control
SINR
Copyright ©
201
6 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Angg
un
Fitrian
Isn
a
wati,
Depa
rtem
ent of Electrical and Inform
ation T
echnology,
Gad
j
a
h
M
a
da
Uni
v
ersity
(
U
GM
),
Yo
gy
aka
r
t
a
, In
do
nesi
a.
Em
a
il: an
g
g
u
n
.s3
t
e13
@
m
a
il.u
g
m
.ac.id
1.
INTRODUCTION
Fem
t
ocel
l
t
echnol
ogy
has
an
i
n
creasi
ngl
y
i
m
port
a
nt
r
o
l
e
t
oday
a
n
d
rese
ach i
n
t
h
i
s
fi
el
d i
s
a
bi
g
ch
allen
g
e. It
has sev
e
ral advan
t
ag
es i
n
term
s o
f
it can
b
e
in
stalled
no
t on
ly in
th
e server sid
e
bu
t also
in
t
h
e
sub
s
cri
b
er si
d
e
, so i
t
can
b
e
used t
o
i
m
p
r
o
v
e net
w
o
r
k
per
f
o
r
m
a
nces conce
r
ni
n
g
i
n
do
o
r
co
vera
ge
. The
fem
t
ocel
l
net
w
or
k ca
n al
s
o
b
e
co
nnect
e
d
t
o
br
oa
d
b
an
d
net
w
o
r
k
ser
v
i
ce
p
r
o
v
i
d
e
r
s s
u
c
h
as di
gi
t
a
l
sub
s
cri
b
er
l
i
n
e (DS
L
), c
a
bl
e m
odem
or ra
di
o f
r
e
que
ncy
(R
F)
. Fe
m
t
ocel
l
t
echno
l
ogy
has al
s
o
bene
fi
t
s
i
n
t
e
rm
s of
ove
rc
om
ing the scarcity of s
p
ectrum
fre
que
n
cies. It ca
n s
h
a
r
e s
p
ectrum
with
m
acro
cell network,
b
u
t
stil
l can
avoi
d i
n
t
e
rfe
re
nce am
ong
fe
m
t
ocel
l
users
o
r
bet
w
ee
n fem
t
ocel
l
use
r
s a
n
d
m
acrocel
l
user
s [
1
]
.
On
e th
i
n
g th
at
stand
s
ou
t to
stu
d
y
on
fem
t
o
cell techn
o
l
og
y is
o
n
its power con
t
ro
l sch
e
m
e
. Po
wer
co
n
t
ro
l sho
w
s
its ro
le in
th
e
d
e
v
e
l
o
p
m
en
t of fem
t
o
cell
n
e
tworks, esp
ecially in
ad
ho
c
natu
re an
d qu
ality o
f
servi
ce [
2
]
.
T
h
e com
m
uni
cat
ion
net
w
or
ks
, eith
er cen
tralized
or d
ecen
t
rali
zed
wan
t
to
tran
sm
it
an
d
receiv
e
d
a
ta in
t
h
e
b
e
st lev
e
l of
QoS (q
u
a
lity of
serv
ice)
ev
en
in
th
e m
i
n
i
m
u
m
p
o
w
er
signals. So
m
e
referen
ces
m
e
nt
i
on t
h
at
cen
t
ralized
power con
t
ro
l in
m
o
b
ile co
mm
u
n
icatio
n
is li
mited
in
th
e p
r
actice i
m
p
l
e
m
e
n
tatio
n
[3
]. Ho
wev
e
r, th
e
co
m
p
arison
with
d
i
st
ri
b
u
t
ed p
o
w
er
co
nt
rol
sc
hem
e
can be
d
one
[
4
]
.
T
h
e i
n
fl
ue
nce
of
nea
r
-
far effect
u
s
i
n
g
freq
u
e
n
c
y
reu
s
e is also
rel
a
ted
to th
e
feasib
ility o
f
signal to
in
terferen
ce
p
l
u
s
no
ise ratio
(SINR) in
a femto
cell n
e
twork
[5
].
A feasib
ility te
st o
f
po
wer con
t
ro
l sch
e
m
e
i
s
d
o
n
e
to
ensure th
at all u
s
ers can
ach
iev
e
SINR targ
ets
in
a fem
t
o
cell n
e
two
r
k
.
Th
e test is no
t on
ly p
e
rform
e
d
u
s
ing
o
n
e
ch
ann
e
l
in
a
g
r
o
u
p
,
bu
t
also
u
s
ing
m
u
ltip
le
channels,
and can be done
either on
a
cen
tralized
or a d
i
st
ribu
ted
n
e
twork
.
If th
e calcu
l
a
ted
p
o
wer tran
sm
it
d
o
e
s no
t m
eet
th
e feasi
b
ility
resu
lts (no
n
-n
eg
ativ
e
p
o
wer
vecto
r) t
h
at still in
n
e
g
a
tiv
e
v
a
lu
e, th
en
h
a
v
e
to
re-
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 4
,
Au
gu
st 2
016
:
16
85
–
1
694
1
686
ad
ju
st th
e
p
a
ra
m
e
ters
[4]
.
R
e
search
o
n
ce
nt
ral
i
zed a
nd
di
st
ri
b
u
t
e
d
po
wer c
o
nt
rol
ha
s bee
n
car
ri
ed
out
by
several stu
d
ies [3]
,
[6]
,
[7]
.
I
n
th
e
pr
ev
i
o
us stud
y [8
], t
h
ere
was a
po
wer con
t
ro
l sch
e
me th
at
u
s
es a feasib
le so
lu
tion
al
g
o
rith
m
on
one c
h
annel
only.
Fundam
entally, the fea
s
ibility test s
hould not only
be
done on
one group
of users
with
a
sin
g
l
e ch
ann
e
l b
u
t
also
can
b
e
do
ne on
the u
s
ers
b
y
u
s
i
n
g
m
u
lti-ch
ann
e
ls th
at will b
e
d
i
v
i
d
e
d
in
to
sev
e
ral
g
r
ou
p
s
o
f
users. In
th
is p
a
p
e
r, th
e cen
tralized
po
wer con
t
ro
l sch
e
m
e
fo
r
m
u
lti-ch
an
n
e
l
u
s
ers in
fem
t
o
cell
n
e
two
r
k
is st
ud
ied
.
Th
e feasib
ility test o
f
m
u
l
ti-ch
ann
e
l
u
s
er is to
b
e
i
m
p
o
r
tan
t
to
be do
n
e
b
ecau
s
e it is
asso
ciated
w
i
t
h
f
e
m
t
o
cell n
e
tw
or
k
topo
logy an
d
also
asso
ciated
w
ith
ch
ar
acter
istics of
f
e
m
t
o
cell
u
s
er
s wh
o
u
s
e t
h
e
d
i
v
i
sion
o
f
RBs as t
h
e tran
sm
issio
n
ch
ann
e
l.
Feasi
b
ility test is in
ten
d
e
d
fo
r each fem
t
o
cell u
s
er to
b
e
ab
le to
ach
i
ev
e th
e targ
et SINR.
Or
ga
ni
zat
i
on o
f
t
h
i
s
pape
r i
s
st
ruct
u
r
e
d
as fol
l
o
w
s
. The sy
st
em
m
odel
of t
h
i
s
st
udy
i
s
descri
be
d i
n
Sectio
n
2
,
and
Sectio
n
3
exp
l
ain
s
th
e
feasib
i
lity o
f
po
we
r co
n
t
ro
l
for th
e
m
u
l
ti-ch
ann
e
l
u
s
er in
i
n
ter-femto
cell
n
e
two
r
k
.
Sectio
n 4 shows t
h
e sim
u
lat
i
o
n
resu
lts, an
d th
e con
c
lu
si
o
n
is presen
ted in
secti
o
n 5.
2.
SYSTE
M
MO
DEL
The st
udie
d
fe
mtocell network in this
pape
r
consists
of se
veral FAPs as
a
receiver and
se
veral FUEs
as a tran
sm
it
te
r.
Du
e to th
e
use of th
e ch
annels th
at sh
ou
ld
n
o
t
b
e
t
h
e sam
e
in
a fem
t
o
cell, th
en
u
s
e th
e
same
chan
nel
can
on
l
y
be do
ne by
t
h
e user i
n
t
h
e di
ffe
re
nt
fem
t
o
cel
l
.
In t
h
e fe
m
t
ocel
l
net
w
or
k, eve
r
y
use
r
i
s
onl
y
affected b
y
i
n
t
e
rferen
c
e fro
m
ano
t
h
e
r fem
t
o
cell u
s
er, as illu
strated in
Figu
re 1.
Fig
u
re
1
.
System
m
o
d
e
l o
f
m
u
lti ch
an
nel u
s
er in
in
ter-fem
t
o
cell n
e
twork
We fo
cu
s on th
e fem
t
o
cel
l u
s
ers (fem
t
o
u
s
ers)
or seco
nd
ary u
s
ers (SU) in
cog
n
itiv
e
rad
i
o
technology. T
h
ey are refe
rre
d to as the seconda
ry use
r
(SU) beca
use the
users are
us
ing
a channel that
is not
b
e
ing
u
tilized
b
y
th
e prim
ar
y u
s
er (PU),
wh
ich
in
th
is
case, is th
e macro
cell u
s
er (macro
u
s
er).
Based
on
Fi
gu
re 1
,
i
t
can be expl
ai
ned t
h
at
fo
r cha
n
nel
1 (C
h
1
), wh
en
th
e tran
sm
itter
FUE
1,1
(SU Tran
sm
it
ter, SUTX
1,1
)
sends a si
gnal
to the
receive
r FAP
1
(SU R
eceiver, SURX
1
), the i
n
terference
occ
u
rs
on the
receive
r FAP
2
(SUR
X
2
). Du
e to
th
is in
terferen
c
e, tran
sm
it
ter FUE
2,2
(S
U
T
X
2,2
) h
a
s to
raise th
e tran
smit p
o
w
er so
it can
ach
iev
e
targ
et
SINR
. Th
is con
d
ition
also
app
lies to ch
ann
e
l 2
(Ch
2
) an
d
will b
e
su
stained
,
so th
at it
req
u
i
res
fo
r po
we
r
c
ont
rol.
In t
h
e case
of
inter-fem
t
ocell netw
ork,
each fem
t
ocell user would be
a
f
fe
cted by inte
rfe
rence
from
nei
g
hb
o
r
i
n
g fe
m
t
ocel
l
users,
whi
c
h i
s
ca
use
d
by
t
h
e u
s
e
of the sam
e
cha
nnel. As
seen i
n
Figure 1, the
use
r
FUE
1,1
and FUE
2,2
usi
ng t
h
e
sam
e
channel
,
C
h
1
, th
u
s
in
terfere with
each o
t
her. Si
m
i
l
a
r
l
y
,
user FUE
2,1
who
will b
e
affected
b
y
in
te
rference fro
m
u
s
er
FUE
1,2
bec
a
use
bot
h
use t
h
e sa
m
e
chan
nel
,
C
h
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Fea
s
ib
ility o
f
Po
wer C
o
n
t
ro
l
fo
r Mu
lti-Chan
n
e
l
User in
Inter-Femt
o
cell Netwo
r
k (Ang
gu
n Fitria
n Isnawa
t
i
)
1
687
In t
h
i
s
pa
per
,
f
i
ve pai
r
s
o
f
use
r
s we
re s
p
rea
d
ra
n
d
o
m
l
y
by
t
h
e user
di
st
ance
set
by
fem
t
ocel
l
net
w
o
r
k
co
nfigu
r
ation
s
with
in th
e m
acro
cell c
ove
ra
g
e
, as
sh
o
w
n
i
n
Fi
gu
re
2.
F
r
om
t
h
e sy
st
em
m
odel
,
t
h
e
r
e a
r
e
som
e
pairs
of use
r
t
r
ansm
itter and receive
r. The
solid lines
re
present t
h
e co
m
m
unication pat
h
,
while the
dashe
d
lin
es in
d
i
cate th
e in
terferen
ce p
a
th
, as shown
in
Fi
g
u
re
1.
Li
nk
gai
n
bet
w
een t
h
e F
U
E
i
and t
h
e
FA
P i
i
s
rep
r
ese
n
t
e
d by
. Th
e
g
a
in
o
f
th
e link
b
e
tween
th
e
FUE
j
and
th
e
FAP i is
represen
ted
b
y
. F
o
r a
sim
p
le
n
o
t
ation
,
it is assu
m
e
d
th
at th
e in
terferen
ce and
noise are com
b
ined and
declare
d
as
. Let
is th
e
t
r
ansm
i
ssi
on p
o
we
r
of F
U
E i
and
i
s
t
h
e
po
w
e
r o
f
ot
her
FU
Es. T
h
ere
f
o
r
e, the
expected SINR of use
r
i
ca
n
be defi
ned
as [
9
]
:
(1)
whe
r
e
is ave
r
a
g
e
of noise level and
is the
sa
me for all recei
vers
.
Fig
u
r
e
2
.
Fem
t
o
cell u
s
er
topolo
g
y
3.
FEASIBILIT
Y O
F
PO
WE
R
CO
NTROL
FO
R
MULTI
-C
HA
NNEL
USER
Int
e
r
f
ere
n
ce
m
a
nagem
e
nt
schem
e
i
n
fem
t
ocel
l
net
w
or
k
i
s
di
vi
de
d i
n
t
o
c
o
-t
i
e
r a
n
d c
r
os
s-t
i
e
r
interfe
rence
.
F
r
om
the distribution
of s
u
ch s
c
hem
e
s,
the inter-fem
t
ocell netwo
r
k incl
ude
s the co-tier sc
hem
e
of uplink
interfere
n
ce
so as t
o
be able t
o
use both C
ode
Di
vision M
u
ltiple Access
(CDMA) a
n
d Orthogona
l
Fre
que
ncy Division
M
u
ltiple Access (OFDMA).
I
n
th
e
O
F
D
M
A
f
e
m
t
o
cell n
e
tw
or
k, th
e study o
f
po
wer
con
t
ro
l is also
d
i
stin
gu
ish
e
d
b
y
th
e ch
ann
e
l
usa
g
es,
which
is a single-c
ha
nnel
or a
m
u
ltiple-c
h
anne
l,
where
each
user
can allocate t
r
a
n
sm
it powe
r
through
m
u
l
tip
le frequen
c
y b
a
n
d
s [1
0]. In
th
e m
a
tter o
f
OFDMA
fe
m
t
ocel
l
,
i
t
depend
s o
n
t
h
e s
u
bcha
n
n
el
s al
l
o
cat
i
on.
In
LTE
fem
t
ocell, sub-c
h
annels known a
s
Resource
Bl
ocks (RBs
).
If the
res
o
urce
bloc
ks
(RBs) were
exha
ust
e
d i
n
f
e
m
t
ocel
l
s
, fem
t
ocel
l
s
wo
ul
d
use ot
her s
u
b
-
chan
nel
s
[
11]
and i
f
am
ong
user
uses t
h
e d
i
ffere
nt
sub
-
c
h
an
nel
s
, t
h
e i
n
t
e
r
f
ere
n
ce
wo
ul
d
be a
v
o
i
ded
[1
2]
. S
u
b
c
han
n
el
s al
l
o
c
a
t
i
on ca
n be i
m
pl
em
ent
e
d b
y
sub
-
ch
ann
e
l selectio
n algo
rith
m
such as E
x
perie
n
ce-Weighted
Attraction (E
WA)
al
go
ri
t
h
m
[13]
an
d
ga
m
e
t
h
eory
m
e
t
hod [
14]
,
[
1
5
]
.
Besid
e
s con
s
idering
th
e in
terferen
ce
factors, th
e co
m
p
lex
ity in
th
e im
ple
m
entation also affects the
d
ecision
o
f
its ap
p
licatio
n. In
practice, the centralized network re
qui
res a cent
r
al
st
at
i
on and
gl
oba
l
in
fo
rm
atio
n
of all u
s
er chann
e
ls.
Similarly
,
th
e cen
t
ralized
p
o
wer co
n
t
ro
l will n
e
ed co
m
p
reh
e
n
s
iv
e
co
n
t
ro
l
si
gnal
s
i
n
t
h
e
net
w
or
k,
so i
t
i
s
har
d
t
o
be
carri
e
d
out
,
p
a
rt
i
c
ul
arl
y
f
o
r
t
h
e ad
-h
oc
wi
r
e
l
e
ss net
w
or
k.
Th
e
com
p
l
e
xi
t
y
of t
h
e cent
r
al
i
zed
net
w
or
ks
occ
u
r
r
beca
use t
h
e pr
ocess
of c
o
m
put
i
ng at
t
h
e base st
at
i
on
i
s
t
o
o
m
u
ch
, wh
ich
is related
to
t
h
e
p
r
o
cessing o
f
all in
fo
rm
atio
n
fro
m
th
e env
i
ron
m
en
t and
th
e
u
s
ers [16
]
.
Th
erefo
r
e, t
h
e
i
m
p
l
e
m
en
tatio
n
of d
i
stri
bu
ted
system
s
th
at u
s
e on
ly lo
cal
in
fo
rm
atio
n
wo
u
l
d
b
e
m
o
re realistic
to
b
e
app
lied
[1
7
]
.
Ho
wev
e
r, th
ese two
d
i
fferen
t sch
e
m
e
s
requ
ire
feasib
il
ity test at
th
e ti
me b
e
fore it will b
e
im
ple
m
ented, as
seen
in Figure 3.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 4
,
Au
gu
st 2
016
:
16
85
–
1
694
1
688
Fig
u
re
3
.
Feasi
b
ility test fo
r cen
t
ralized
and
d
i
stribu
ted
sche
m
e
[16
]
3.
1.
Fea
s
ibility in
Centra
lized Po
wer Co
ntro
l
If th
ere is transmissio
n
ch
ann
e
l with
a fre
quency K, K<N, so that e
xpect
ed SINR of the user i from
t
h
e eq
uat
i
o
n
(1
) ca
n
be re
de
fi
ned
as
[8]
,
[9]
:
(2)
whe
r
e
is average of noise leve
l and sam
e
for all receivers.
is th
e in
terferen
c
e fun
c
tio
n
that sh
ows t
h
e
in
terferen
c
e that o
ccurs
b
e
tween
th
e
user i
an
d j, an
d is
d
e
fin
e
d
as:
On t
h
e physical layer, powe
r control can
minimi
ze the interfe
re
nce and for a
feasibl
e
syste
m
, the
resu
lts
o
f
all
u
s
ers m
u
st meet th
e
d
e
termin
ed
SIR.
Users can
adj
u
st th
eir tran
sm
i
t
p
o
wer level to
t
h
e
d
i
stribu
ted
power con
t
ro
l to
mak
e
sure th
at
all u
s
ers
wh
o s
h
are t
h
e sam
e
channel m
eet
the re
quirem
ents of the
targ
et SIR (
to
the expecte
d
re
ceiver.
To
fu
lfill th
e
req
u
i
rem
e
n
t
s o
f
Qo
S
for th
e
sp
ecified co
mmu
n
i
cation
s
i
n
fe
m
t
o
cell n
e
two
r
k
,
we m
u
st
be
c
o
nsidere
d
making sure
th
at
the
SINR of each user
al
way
s
be
great
e
r
t
h
an t
h
e
t
h
resh
ol
d S
I
NR
:
(3)
Fo
r
u
s
ers who sh
are th
e same frequ
en
cy chan
n
e
l, tr
an
sm
is
sio
n
power wil
l
affect th
e q
u
ality
o
f
th
e
l
i
nk a
n
d
t
h
e
t
e
m
p
erat
ure
noi
s
e
i
n
a
pa
rt
i
c
ul
ar c
h
an
nel
.
Ther
efo
r
e, th
e pur
po
se
of
p
o
wer co
n
t
ro
l is t
o
set th
e
tran
sm
it p
o
w
er o
f
all u
s
ers i
n
i
m
p
r
ov
ing
th
e q
u
a
lity o
f
link th
at allo
w
s
u
s
er group
s to
tran
sm
it d
a
ta th
ro
ugh
th
e sam
e
ch
ann
e
l to
reach
th
e b
it error
rate (BER) targ
et.
Feasib
ility o
f
po
wer con
t
ro
l is b
a
sed
up
on
t
h
e fo
llo
wi
n
g
equ
a
tio
n [9
]:
(
4
)
If gi
ven t
h
e t
a
rget
SIR
ɣ
*, for achie
ving the feas
ible condition, it i
s
n
ecessary to
m
eet
the
r
e
qu
ir
em
en
ts of
non
-n
eg
ativ
e po
w
e
r
v
ector
P*
, w
ith
is no
rm
alized
lin
k g
a
in m
a
trix
an
d satisfied
th
e equ
a
tion
b
e
lo
w:
(
5
)
and
is no
rm
ali
zed
n
o
i
se v
ect
o
r
th
at fu
lfilled th
is eq
u
a
tion
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Fea
s
ib
ility o
f
Po
wer C
o
n
t
ro
l
fo
r Mu
lti-Chan
n
e
l
User in
Inter-Femt
o
cell Netwo
r
k (Ang
gu
n Fitria
n Isnawa
t
i
)
1
689
(
6)
Fo
r
u
s
ers who sh
are th
e same frequ
en
cy chan
n
e
l,
th
e transmissio
n
p
o
wer will affect th
e q
u
a
lity o
f
the link a
n
d the te
m
p
erature
disturba
n
ce i
n
a certain c
h
a
n
nel. T
h
e
purpose
of
th
e pow
er
co
n
t
ro
l is adju
sting
th
e tran
sm
i
t
p
o
w
er in all u
s
ers for im
p
r
o
v
in
g
t
h
e
q
u
a
lity o
f
relatio
n
s
h
i
p
s
t
h
at allow th
e u
s
ers
group
t
o
tran
sm
it th
ro
ugh
th
e sam
e
ch
an
n
e
ls so
it can
meet the specified BER.
B
a
sed
on
eq
ua
t
i
on (
4
) a
b
o
u
t
a feasi
b
l
e
s
o
l
u
t
i
on i
n
si
n
g
l
e
-c
han
n
el
user
g
r
ou
p,
t
h
e
n
t
h
e
e
quat
i
o
n ca
n
b
e
d
e
v
e
l
o
p
e
d
i
n
to
a feasi
b
le so
lu
tion
to th
e
m
u
l
ti-ch
ann
e
l
u
s
er as
fo
llows:
(7
)
whe
r
e
K i
s
t
o
t
a
l
num
ber
of
ch
annel
,
k
i
s
c
h
a
nnel
n
u
m
b
er, a
n
d
i
s
fu
nct
i
o
n
of
use
r
i
n
t
e
rfe
r
e
nce.
3.
2.
Fea
s
ibility in
Distri
buted Po
wer Co
ntro
l
Feasib
ility test
for th
e
d
i
stribu
ted
p
o
wer con
t
ro
l (D
PC) is sho
w
n
b
y
testin
g
th
e conv
erg
e
n
ce
of an
i
t
e
rat
i
v
e al
go
ri
t
h
m
.
The eq
ua
t
i
on
used t
o
ve
ri
fy
t
h
e c
o
n
v
e
r
gence
i
n
t
r
od
uc
ed by
Yat
e
s [
6
]
.
Po
wer c
o
nt
r
o
l
i
n
d
i
str
i
bu
ted
sche
m
e
r
e
f
e
r
s
t
o
th
e stud
y of
Y
a
tes [6
], wh
i
c
h
pro
v
e
s th
e
co
nv
er
g
e
n
ce
of
th
e
D
P
C algo
r
ith
m
base
d
on
st
an
d
a
rd
i
n
t
e
r
f
ere
n
c
e
f
unct
i
o
n:
(
8
)
wh
ere I is th
e
in
terferen
c
e fun
c
tio
n th
at d
e
fin
e
s power
ve
ct
or i
n
t
h
e
nex
t
i
t
e
rat
i
on. I
n
t
e
rfe
rence
f
unct
i
on i
s
nam
e
d st
anda
r
d
i
f
i
t
m
e
et
s al
l o
f
a
n
o
n
-
ne
gat
i
ve p
o
w
er
vect
or:
1
)
Positiv
ity
2)
M
o
not
oni
ci
t
y
3
)
Scalab
ility
If t
h
e sy
st
em
is feasi
b
l
e
, t
h
e
n
t
h
e po
wer
vec
t
or o
f
st
an
dar
d
i
n
t
e
rfere
nce f
unct
i
o
n w
o
ul
d
con
v
e
r
ge t
o
m
i
nim
u
m
sol
u
t
i
on
po
wer
ve
ct
or
P*
, w
h
i
c
h i
s
i
n
di
cat
ed
by
a
no
n-
ne
g
a
t
i
v
e p
o
we
r
v
ect
or a
n
d
ge
o
m
et
ri
c
conve
r
ge
nce s
p
eed.
B
a
sed
on
t
h
e a
l
go
ri
t
h
m
Foschi
ni
-M
i
l
j
a
n
i
c
[
18]
ca
n
be c
o
n
c
l
ude
d t
h
at
a
s
l
o
n
g
as t
h
ere
i
s
a feasi
b
l
e
v
ector P*
th
en
th
e DPC iteratio
n algo
rith
m
o
f
th
e link
in the n
e
two
r
k
will always resu
lt in
rap
i
d conv
erg
e
n
ce
to
ward
s P*
.
The alg
o
rith
m
can
b
e
written
as
fo
llows:
(9
)
wh
ere
k
is a
d
i
screte ti
m
e
in
dex
,
ɣ
i is m
i
n
i
mu
m
SINR thresh
o
l
d
for lin
k
i
and
is SINR
user
i
.
If t
h
e vec
t
or
P*
do
no
t satisfy th
e eq
u
a
tion
, th
en
th
is alg
o
r
ith
m
will
g
e
n
e
rate po
wer to
th
e en
tire tran
sm
it
ter
becom
e
s infi
nite (infea
sible).
Fo
sch
i
n
i
-Milj
an
ic (FM) al
g
o
rith
m
sh
o
w
ed
t
h
at th
e sch
e
m
e
o
f
DPC will b
e
conv
erg
e
n
c
e to
ward
s a
feasib
le co
nd
itio
n wh
en
feasible so
lu
tion
ex
is
ts [19
]
.
Iteration
sch
e
m
e
p
r
opo
sed
b
y
FM:
(10)
whe
r
e
β
is a
pa
ram
e
ter that is chosen to ac
hi
eve c
o
nve
r
ge
nce.
Because the
J
acobia
n
iteration produ
ces sl
ow
convergence so as
to
ge
t better faste
r
conve
r
ge
nce
usi
n
g i
t
e
rat
i
v
e
al
gori
t
h
m
s
Fosc
hi
ni
M
i
l
j
an
i
c
[20]
an
d al
so sh
o
w
an al
t
e
rnat
i
v
e t
o
pr
ove t
h
e co
nve
rge
n
ce
besi
des
Yat
e
s
[
6
]
.
Furt
herm
ore,
DPC can
be done by using the
gam
e
theory m
e
thod a
nd
get
the converge
nce condition
can
b
e
don
e based
on
an
iterativ
e alg
o
rithm o
f
p
o
wer co
n
t
ro
l d
e
riv
e
d fro
m
u
tilit
y f
u
n
c
tion
,
wh
ich is th
e
d
i
fferen
tial v
a
lu
e
o
f
th
e
u
tility
fu
n
c
tion
t
o
wa
rd
th
e power tran
sm
it fu
n
c
tion
[21
]
-[23
].
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 4
,
Au
gu
st 2
016
:
16
85
–
1
694
1
690
4.
R
E
SU
LTS AN
D ANA
LY
SIS
B
a
sed o
n
t
h
e f
e
m
t
ocel
l
user t
o
p
o
l
o
gy
i
n
Fi
g
u
re
2, t
h
e
per
f
o
rm
ance o
f
t
h
e pr
o
pose
d
p
o
w
er c
o
nt
r
o
l
schem
e
is checked. As
sum
e
d that group of five
t
r
a
n
sm
itter-receiver pairs
of s
econda
r
y users is
com
m
uni
cat
i
ng
wi
t
h
eac
h
ot
h
e
r i
n
a
40
0 s
q
u
a
re m
e
t
e
rs area
.
Th
is
p
a
rt
will sh
ow so
m
e
o
f
t
h
e resu
lts
o
f
the feasib
ility test o
f
fem
t
o
cell
n
e
two
r
k
s
t
h
at asso
ciated
t
o
p
o
wer co
n
t
ro
l
b
a
sed
o
n
th
e
network top
o
l
o
gy in
Fi
g
u
re
2
.
Feasib
ility test
of th
e system
is do
n
e
to
en
sure th
at
th
e syste
m
is feasib
le with
the p
o
wer v
ect
o
r
is n
o
n
-
n
e
g
a
tive, wh
ich
m
ean
s in
prac
tice, the system
can
meet
the re
quirem
ents of SINR.
Thi
s
pa
per i
s
a
cont
i
n
uat
i
o
n f
r
om
previ
o
us
p
a
per [
2
4]
t
h
at
focu
sed
on t
h
e
schem
e
of po
w
e
r vect
o
r
i
n
a u
s
er
g
r
ou
p,
wh
ere th
ere are two
sch
e
m
e
s
for th
e sing
le
ch
ann
e
l usag
e
with
a m
u
lti-u
s
er,
n
a
m
e
l
y
th
e s
m
all
user
gr
o
up
(wi
t
h 5 u
s
ers
per
chan
nel
)
a
n
d t
h
e bi
g use
r
g
r
o
up
(wi
t
h
1
5
us
ers pe
r cha
n
nel
)
. T
h
e p
r
ev
oi
u
s
pape
r
has em
phasized on ac
hieving the differ
en
t SINR targ
et acco
rd
ing
to
th
e
si
ze
of user gr
ou
p. Wh
en
t
h
e
use
r
gr
o
up si
ze i
s
bi
g (a l
o
t
of us
ers w
ho
use t
h
e sam
e
channe
l
)
, t
h
e SI
NR
t
a
rget
s t
h
at
can
be achi
e
ve
d u
s
er i
s
g
e
ttin
g sm
aller
,
and v
i
ce v
e
rsa wh
en
th
e user
g
r
o
up size
is
sm
a
ll (litt
le u
s
er
who
u
s
e th
e sam
e
ch
ann
e
l) th
en
th
e user is ab
le to
ach
i
e
v
e gr
ea
te
r
S
I
N
R
ta
rge
t
.
Un
lik
e
p
r
ev
iou
s
st
u
d
i
es th
at
fo
cu
s
o
n
on
e ch
ann
e
l, th
is stud
y fo
cu
sed
o
n
th
e sch
e
me o
f
m
u
lti-
chan
nel
usa
g
e.
C
h
an
nel
usa
g
e schem
e
i
n
t
h
i
s
st
u
d
y
are
di
vi
ded
i
n
t
o
t
h
r
ee t
y
pes,
nam
e
l
y
:
1)
schem
e
of
si
ngl
e
-
ch
ann
e
l m
u
lti-
u
s
er
wh
ere
one ch
ann
e
l is u
s
ed
sim
u
ltan
e
o
u
sly b
y
m
u
ltip
le
u
s
ers;
2
)
sch
e
me o
f
m
u
lti-ch
an
n
e
l
m
u
lt
i
-
user
whe
r
ei
n eac
h use
r
gr
o
up
onl
y
u
s
e
s
t
h
e sam
e
channel
,
a
n
d 3
)
sc
hem
e
of m
u
l
t
i
-
chan
nel
si
n
g
l
e
-
u
ser
whe
r
e
eac
h use
r
ca
n use
m
o
re than one
c
h
annel.
Fig
u
re
4
is rep
r
esen
tatio
n of sch
e
m
e
1
.
It
sh
o
w
s th
e
u
s
ag
e im
p
act o
f
po
wer co
n
t
rol feasib
ility
m
e
t
hod t
o
t
h
e
achi
e
vem
e
nt
o
f
t
h
e t
a
r
g
et
S
I
NR
, i
f
i
t
c
o
m
p
ared
bet
w
ee
n
bef
o
re
an
d a
f
t
e
r u
s
i
n
g t
h
i
s
m
e
t
hod
.
Fig
u
re 4
exp
l
ain
s
th
at wh
en
the co
nd
itio
n
feasib
le th
en
th
e en
tire u
s
er can
g
e
t th
e SINR targ
et,
wh
ich
is eq
u
a
l
to
22
.2
d
B
wh
ile in
th
e in
itial co
nd
itio
n
t
h
ere are u
s
ers
who
d
o
no
t m
e
e
t
th
e targ
et SINR,
as sho
w
n
b
y
u
s
er 1.
Besid
e
s, th
e resu
lt also
sh
ows th
at th
e in
itial p
o
wer
h
a
s an u
n
e
v
e
n
d
i
stribu
tio
n of power o
u
t
pu
t co
m
p
ared
t
o
feasib
le so
lu
tion
p
o
wer co
n
t
rol.
Fi
gu
re
4.
Ac
hi
evem
ent
of t
a
r
g
et
S
I
R
bet
w
ee
n a
f
t
e
r a
n
d
be
f
o
re
usi
n
g
feasi
b
l
e
s
o
l
u
t
i
o
n sc
hem
e
Fig
u
re
5
is th
e represen
tatio
n o
f
sch
e
m
e
2
.
It sh
ow
s t
h
e effect o
f
ch
an
n
e
l
ad
d
ition
to th
e u
s
er group
will ten
d
to
increase th
e SINR o
f
each
u
s
er. Fro
m
th
at
fig
u
r
e, we can
com
p
are th
e SINR ach
iev
e
d
b
y
five
users
w
h
e
n
usi
ng
o
n
e ch
an
nel
(re
d ba
rs
), t
w
o cha
n
nel
s
(
g
r
een ba
rs
) an
d t
h
ree c
h
a
nnel
s
(
b
l
u
e
bars
).
Wh
en we
use t
h
ree c
h
an
nel
s
f
o
r
5
use
r
s
,
t
h
e
n
cha
n
nel
1 f
o
r
use
r
1 a
n
d 2
,
c
h
an
nel
2
fo
r u
s
er
3 a
n
d
4, a
n
d cha
n
nel
3 o
n
l
y
for
u
s
er
5
.
It can
b
e
seen th
at
at th
e ti
m
e
o
n
e ch
ann
e
l
on
ly
u
s
ed
b
y
on
e user (sing
l
e-ch
an
n
e
l
sing
le-u
ser) will
obt
ai
n
t
h
e m
a
xim
u
m
SINR
, a
s
sh
ow
n i
n
t
h
e
use
r
5
.
It
s
h
o
w
s t
h
e
SI
NR
t
h
at
g
r
eat
er t
h
a
n
80
dB
,
so i
t
can
be
use
d
f
o
r vi
de
o com
m
uni
cat
i
ons.
Fig
u
re
6
is th
e represen
tatio
n o
f
sch
e
m
e
3
.
It sh
ows th
e
u
s
ag
e of m
u
lti c
h
ann
e
ls in
each
u
s
er.
Fro
m
Fi
gu
re 6 i
t
can be seen t
h
e
val
u
e of S
I
N
R
i
n
users 3
and
4, t
h
at
w
h
en
users
use
a
m
u
l
t
i
-
chan
nel
(t
wo
ch
ann
e
ls) t
h
en it will h
a
v
e
a h
i
gh
er to
tal SINR th
an
sing
l
e
ch
ann
e
l SINR. It b
ecau
s
e
o
f
each
channel h
a
s
SINR
that ca
n
be acc
um
ulate
d
as
a total SINR in each us
er.
As a
whole of sc
hem
e
s, this sc
hem
e
is the
bes
t
schem
e
that will be able to increase
the total SINR pe
r use
r
if it co
m
p
ared
to the SIR of each user in others
schem
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Fea
s
ib
ility o
f
Po
wer C
o
n
t
ro
l
fo
r Mu
lti-Chan
n
e
l
User in
Inter-Femt
o
cell Netwo
r
k (Ang
gu
n Fitria
n Isnawa
t
i
)
1
691
Figu
re
5
.
SIR
o
f
m
u
lti
-ch
a
nn
el m
u
lti-
u
s
er
Fig
u
re
6
.
Mu
lti-ch
ann
e
l u
s
ag
e of
u
s
ers
Th
e av
erage po
wer of u
s
er will d
ecrease
wh
ile in
creasi
n
g
th
e
n
u
m
b
e
r o
f
p
r
ov
id
ed
ch
ann
e
ls, as
sh
own
in
Figu
re 7
.
Th
is is d
u
e to
th
e in
creasin
g
ch
ann
e
l
m
a
k
e
s th
e u
s
er atte
m
p
ted
to
g
e
t th
e ch
an
n
e
l is no
t as
h
a
rd
a
s
th
e u
s
e
r
w
h
en
co
mp
e
t
e
w
ith
a lo
t
o
f
u
s
er
in
si
ngle channel. There are
tim
es
whe
n
t
h
e a
d
ditiona
l
ch
ann
e
ls
do
no
t affect to
t
h
e po
wer
redu
ctio
n
s
of
u
s
er
s.
Th
is is
du
e to
th
e user
n
e
ed
s on
ly a fu
lfillmen
t o
f
SINR
t
a
rget
s
,
so i
f
t
h
e t
a
r
g
e
t
SINR
has
b
een reac
he
d t
h
en
ad
di
t
i
onal
cha
nnel
doe
s
not
ha
ve si
g
n
i
f
i
cant
in
flu
e
n
ce. By co
n
t
rast, wh
en
t
h
e targ
et SIR of th
e u
s
er
is rai
s
ed
, th
e user po
wer will also
in
crease as shown
in
Fig
u
re
8
.
Th
is is b
ecau
s
e the u
s
er
will try ev
en
h
a
rd
er t
o
g
e
t en
oug
h
p
o
wer tran
sm
i
t
s con
d
ition
so as t
o
ach
iev
e
a
h
i
gher targ
et SIR an
d qu
ality service b
e
tter.
Figu
r
e
7.
A
v
erag
e
p
o
w
e
r
w
h
en
th
e nu
m
b
er
of
ch
ann
e
l
Figure
8.
P
o
we
r w
h
e
n
tar
g
et S
I
R is c
h
ange
d
is
change
d
Fig
u
re
9
sh
ows th
at th
e SINR o
f
each
u
s
er will ra
ise alo
n
g
with th
e in
creasin
g
nu
m
b
er o
f
ch
an
n
e
l
s
.
Because
of the
m
o
re cha
nnels
are
prov
ide
d
,
the user ca
n
us
e the c
h
annel e
a
sily without having to s
h
are
wit
h
othe
r users. It
also has bee
n
show
n
in detail
in
Figure 5 when usi
ng
m
u
lti channel,
whe
r
e the SINR of each
u
s
er at th
e tim
e
u
s
es three ch
an
n
e
ls will b
e
greater
than th
e
u
s
e
o
f
two
ch
an
n
e
ls or
o
n
e
chan
n
e
l.
The
great
e
r
nu
m
b
er of
use
r
s i
n
a c
h
a
nnel
g
r
ou
p,
t
h
e less valu
e SI
N
R
can b
e
ob
tain
ed
by th
e user
i
n
the group as shown in Figure 10.
This
is because am
ong
users will
have
m
o
re com
p
eti
t
ors i
n
c
o
m
p
eti
tion t
o
g
e
t a p
a
rt of ch
ann
e
l fo
r m
a
i
n
tain
ing
a b
e
tter qu
ality o
f
serv
ice (QoS) that
m
ean
s h
a
v
e
m
o
re in
terferen
ce, so
th
at th
e ach
i
eved
SINR
will decrease.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 4
,
Au
gu
st 2
016
:
16
85
–
1
694
1
692
Fi
g
u
r
e
9.
SIR
w
h
en
n
u
m
b
er
of
cha
n
nel
i
s
c
h
an
ge
d
Fi
gu
re
1
0
.
SIR
i
n
t
h
e a
d
di
t
i
on
of
use
r
per
cha
nnel
Also
, t
h
e resu
lts o
f
feasi
b
ility
test can
also
sh
ow th
e influ
e
n
ce of th
e target SINR to
th
e
m
a
x
i
m
u
m
am
ount
of
use
r
s that can
be s
e
rve
d
,
by re
ference to th
e power vector
fea
s
ibility.
This
is shown in Fi
gure 11.
Based
on
t
h
e
resu
lts
o
f
t
h
e
feasib
ility test
o
f
t
h
e system can
b
e
con
c
l
u
d
e
d
ho
w m
a
n
y
u
s
ers th
e syste
m
i
s
capabl
e
o
f
bei
n
g han
d
l
e
d
ba
se
d on
t
h
e n
o
n
-
n
e
gat
i
v
e po
we
r vect
o
r
rel
a
t
e
d
t
o
S
I
NR
t
a
r
g
et
.
Fi
gu
re
1
1
. Ta
r
g
et
S
I
R
w
h
e
n
m
a
xim
u
m
num
ber
o
f
user
i
s
c
h
an
ge
d
5.
CO
NCL
USI
O
N
In t
h
i
s
pape
r
we
have
p
r
o
v
e
d
t
h
at
by
usi
n
g t
h
e
m
e
t
hod
f
easi
b
l
e
sol
u
t
i
o
n t
h
e
st
u
d
i
e
d
p
o
we
r c
ont
rol
sch
e
m
e
can
b
e
i
m
p
l
e
m
en
ted
as it sh
o
w
s th
at th
e resu
lt is a
l
ways p
o
s
itiv
e
an
d
ach
i
ev
e the targ
et SINR
for all
u
s
ers. Fu
rth
e
rm
o
r
e, we h
a
v
e
lso
d
e
m
o
n
s
trated
th
at
th
e SINR u
s
ers will in
crease
wh
en th
ere are ad
d
i
tio
n
a
l
ch
ann
e
ls fo
r user g
r
ou
ps, and
th
e av
erag
e
p
o
wer o
f
each u
s
er will d
ecrease wh
ile in
creasing
th
e amo
u
n
t
o
f
provide
d
cha
n
nels. T
h
e
great
er num
b
er
o
f
users in a user
group
, th
e less
v
a
lu
e
o
f
SINR
ob
tain
ed b
y
the u
s
er
in
th
e
gr
oup
.
In
th
e
fu
tu
re
works, th
e st
ud
y on
th
e feasib
ility o
f
p
o
wer con
t
ro
l sche
m
e
in
a d
i
strib
u
t
ed
system
u
s
ing
au
tono
mo
u
s
d
i
st
ribu
ted po
wer con
t
ro
l
and
g
a
m
e
th
eo
ry m
e
th
o
d
wi
ll b
e
carried ou
t as su
stain
a
bilit
y o
f
th
is research.
REFERE
NC
ES
[1]
D. T. Ngo,
et al.
, “Distributed I
n
terfer
e
nce Man
a
gement
in
Two
-
Tier
CDMA Femtocell Networ
ks,”
IEEE Trans.
Wirel. Commun.
, vol/issue: 11(3)
, pp
. 979–989
, 2
012.
[2]
S. Ha
y
k
in,
“
C
ognitive
radio
:
br
ai
n-em
powered wi
reless com
m
unic
a
tions,
”
IE
EE J.
Sel.
Areas
Commun.
, vol/issue:
23(2), pp
. 201–2
20, 2005
.
[3]
J. Zand
er, “
P
erf
o
rm
ance of op
ti
m
u
m
transm
itter power cont
rol
in ce
llul
a
r r
a
dio
s
y
stem
s,”
Veh
.
Technol. IEEE
Trans. On
, vol/issue: 41(1), pp
. 5
7–62, 1992
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Fea
s
ib
ility o
f
Po
wer C
o
n
t
ro
l
fo
r Mu
lti-Chan
n
e
l
User in
Inter-Femt
o
cell Netwo
r
k (Ang
gu
n Fitria
n Isnawa
t
i
)
1
693
[4]
H. Koivo and M. Elmusrati, “Sy
s
te
ms engineerin
g in wireless co
mmunica
tion,” Hoboken,
NJ,
John Wiley
& Sons,
2009.
[5]
V. Chandras
ekh
a
r,
et a
l
.
, “Power contro
l in
two-
tier
femtocell networks,”
IEEE Trans. Wirel. Co
mmun.
, vol/issue:
8(8), pp
. 4316–4
328, 2009
.
[6]
R.
D.
Yates,
“A framework
for uplink power co
ntrol in
cel
lula
r radio
s
y
s
t
em
s
,
”
Sel.
Areas Commun. IEEE J. On
,
vol/issue: 13(7), pp.
1341–1347
, 1995.
[7]
G. J. Foschini and Z. Miljanic,
“A
si
mple distributed autonomo
u
s power c
ontrol algorithm and its converg
ence,”
IEEE Trans. Veh
.
Techno
l.
, vo
l/is
sue: 42(4), pp
. 6
41–646, 1993
.
[8]
N.
Amalia,
et a
l
.
, “
S
tud
y
on Fea
s
ible Solution of
Power Control
in Cognitiv
e Ra
dio Networks,”
presented
at
the
ICSGTEIS 2014
, Bali, pp. 58–61, 2014.
[9]
N.
Nie,
et al.
,
“A game theor
e
tic appro
ach
to interferen
c
e
management in
cogni
tive netw
orks,” in
Wirele
ss
Communications
, Springer
,
pp
. 1
99–219, 2007
.
[10]
W.
L
i
,
et al.
, “Energ
y
-
eff
i
cient power allocatio
n with dual-util
i
t
y
in two-t
i
er O
F
DMA fem
t
ocel
l networks,” in
Globecom Workshops (GC Wkshps)
,
2012 IEEE
,
pp. 535–540
, 20
12.
[11]
T. H. Chuang,
et al.
, “Alleviating Interf
eren
ce through Cogniti
ve Rad
i
o for
LTE-Advanced N
e
twork,”
Int. J.
Electr. Comput.
Eng. I
J
EC
E
, vol/issue: 5(3)
, pp
.
539, 2015
.
[12]
T. Zahir
,
et a
l
.
, “Interferen
ce
M
a
nageme
nt in
Fe
m
t
ocells,
”
IEEE Commun. Surv. Tutor.
, vol/issue: 15(1), pp
. 293
–
311, 2013
.
[13]
S. Yong and Q.
Jiansheng, “Cog
nitive
Radio
Channel Sel
e
ction St
rateg
y
B
a
sed on
Experi
ence-Wei
ghted Attr
action
Learn
i
ng,
”
TEL
K
OMNIKA Indo
nes. J.
Electr. Eng.
, vol/issue: 1
2
(1), 2014
.
[14]
Y.
Xu,
et a
l
.
, “
G
am
e-theore
ti
c ch
annel se
le
ction f
o
r interf
eren
ce
m
itigation
in co
gnitive
radio n
e
t
w
orks with bloc
k-
fading chann
e
ls,”
Wireless Com
m
unications and
Netw
orking Co
nference (
W
CNC)
, 2013 IEEE
,
pp. 303–308
, 20
13.
[15]
H.
N.
Pham,
et
al
., “QoS-aware chann
e
l selection
in cognitive radio networks
: a ga
me-th
e
oretic appro
ach
,” in
Global Telecom
m
unications Con
f
eren
ce, 2008. I
EEE GLOBECOM 2008. IEEE
, p
p
. 1–7
, 2008
.
[16]
Z. L
i
ang
,
et
al
.
,
“
D
ecentr
ali
zed
bit, s
ubc
arri
er
and power a
llo
cat
ion with in
te
rferenc
e avo
i
da
nce in m
u
lt
ic
el
l
OFDMA sy
stem
s using game
the
o
reti
c appro
ach,
”
in
Mi
litary Co
mmunications Conference, 2008
. MILCOM 2008.
IEEE
, pp
. 1–7
, 2
008.
[17]
S.
A.
Sa
a
d
,
et a
l
.
, “A Survey
o
n
Power Control Te
chniques in Femtocell
Networks,”
J. Commun.
, pp
. 845–85
4,
2013.
[18]
J. Mounzer
, “Distributed
Power
Control
Algor
ith
ms in Wireless
Networking,” 20
09.
[19]
G. J. Foschini
and Z. Miljanic,
“Distributed
autonomous wireless cha
nnel as
signment algorithm with power
control,”
Veh
.
T
echnol. I
EEE Trans. On
, vo
l/issu
e: 44(3)
, pp
. 420
–429, 1995
.
[20]
C. W. Sung and K. K. Leung,
“A Generalized Framework for Distributed
Power
Control in Wireless Networks,”
IEEE
T
r
ans
. Inf.
T
h
eor
y
, vol/issue: 51(7)
, pp
. 262
5–2635, 2005
.
[21]
F.
L
i
,
et a
l
.
, “
A
New Gam
e
Algorithm
for P
o
wer Control
in
Cognitiv
e R
a
di
o Networks,”
I
EEE Trans. Veh.
Technol.
, vo
l/iss
u
e: 60(9)
, pp
. 43
84–4391, 2011
.
[22]
P. X. - and
Z. Q. -, “Power Control
Bas
e
d On Ga
m
e
Theor
y
In F
e
m
t
ocell
,
”
In
t.
J.
Digit. Con
t
ent T
echnol. Its App
l
.
,
vol/issue: 6(18), pp.
357–366
,
20
12.
[23]
X.
Deng,
et al.
, “A novel distr
i
buted power
co
ntrol bas
e
d on
game
theor
y
in
cognitive
wireless network,” in
Communications in Ch
ina (
I
CCC)
, 2014 IEEE/CI
C
International
Conference on
,
pp. 59–63
, 2014
.
[24]
A.
F.
Isnawati,
et al.
, “
F
easib
le solution of
c
e
ntra
liz
ed powe
r
control for m
u
lti ch
anne
l co
gnitive f
e
m
t
oce
l
l
network,” in
2015 7th International Conference on Information
Technology and
Electrica
l
Engin
eering (
I
CITEE)
,
pp. 539–542
, 20
15.
BIOGRAP
HI
ES OF
AUTH
ORS
Anggun Fitrian Isnawati received Master de
gree in Teleco
mmunication Engineer
ing from
Gadjah Mad
a
U
n
iversity
, Yog
y
akarta, Indon
esia
in
2011. She joined as
a Lectur
er in th
e Telkom
School of Telematics (ST3 Telko
m
) since 2002.
Cu
rrently
she
is a student on
the
Ph.D program
at Depart
em
ent of Elec
tric
al Eng
i
neer
ing and Inform
ation Techn
o
log
y
, Gadjah
M
a
da Univers
i
t
y
,
Yog
y
ak
arta
, Ind
ones
i
a. S
h
e
is
current
l
y
m
e
m
b
er of Digit
a
l S
y
s
t
em
and
Ele
c
t
ronic S
y
s
t
em
laborator
y
at Universitas Gadj
a
h
M
a
da. Her res
earch in
ter
e
sts include telecommunication,
cellular
network, game
theor
y
an
d cognitive rad
i
o
.
Risanuri Hiday
a
t, Dr. Eng. He r
eceived Mast
er in Information
and
Communication Technolog
y
from Agder University
College, Norway
,
in 2002 and Doctor in Telecommunication
Engineering, K
i
ng Mongkut’s Institute of
Techno
log
y
Ladkrabang (KMITL), B
a
ngkok
,
Thailand, in 200
9. He
is now an
Associate Profes
sor at Dep
a
rtment of
Electrical Engineer
ing and
Information Technolog
y
,
Gadjah
Ma
da University
. His inter
e
st is
in Signal Processing, Digital
and Analog
Cir
c
uit Design
, Telecommunication
S
y
ste
m
s,
Da
ta
Communi
cation and Networking,
Java Program
m
i
ng and
its Appl
i
cat
ions,
and a
l
so Speech
and
Im
a
g
e Re
cognit
i
on.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 4
,
Au
gu
st 2
016
:
16
85
–
1
694
1
694
Selo Sulisty
o
is an Associate P
r
ofessor in
Infor
m
ation and
Co
mmunication Technolog
y
at
the
Departm
e
nt of
Ele
c
tri
cal
Eng
i
n
eering
and
Info
rm
ation T
echno
log
y
and
als
o
Head of S
i
s
t
em
Electronis labor
ator
y
at Gadjah
Ma
da
Unive
r
sity
.
His re
sea
r
c
h
inter
e
sts inc
l
u
d
ing Software
Certifi
c
a
tion, So
ftware Modeling
,
Dom
a
in Speci
fic Langu
ages
, M
obile a
pplication
development
and Software Service Engin
eer
in
g
which r
e
l
a
tes
t
o
the
Int
e
rnet
of
Things
and
conn
ect
ed obj
ec
ts
.
I Way
a
n
Must
i
k
a
.
,
Ph.
D
.
He
re
ce
i
v
e
d
t
h
e
B.E
n
g.
de
gree
i
n
el
ec
tri
c
al
e
ngi
ne
e
r
i
n
g from Ga
dj
a
h
Mada University, Indonesia, in 2
005, and th
e M.
Eng degree in computer
engineering from King
Mongkut’s Institute of
T
echnol
og
y
Ladkr
abang
(KMITL),
Thailand
, in
2008,
and th
e Ph.D
.
degree in infor
m
atics from Ky
oto University
,
Japa
n, in 2011. He is currently
a Lectur
er at
Gadjah Mada Universit
y
. He w
a
s a Student Ac
tivit
ies Advisor of IEEE Indone
sia Section in
2014 and he
is
currently
a Secr
etar
y
of
IEEE I
ndonesia Sectio
n. His rese
arch interests in
clud
e
smart sy
stems, machine-to-m
achine communications,
and reso
urce management in cognitive
radio
and heter
ogeneous networks with a par
ticular
emphasis
on spect
rum sh
aring
and game
theor
y
. He re
cei
ved the Young Research
er’s Encouragem
ent A
w
ard from
IEEE VTS Japan in
2010 and Stud
ent Paper
Award
from
IEEE Kansai Section
in
2011.
He is no
w a Secr
etar
y
o
f
IEEE Indonesia
Section
.
Evaluation Warning : The document was created with Spire.PDF for Python.