TELKOM
NIKA
, Vol.11, No
.1, March 2
0
1
3
, pp. 167~1
7
4
ISSN: 1693-6
930
accredited by D
G
HE (DIKTI
), Decree No: 51/Dikti/Kep/2010
167
Re
cei
v
ed O
c
t
ober 2
4
, 201
2; Revi
se
d Ja
nuary 7, 201
3
;
Accepte
d
Febru
a
ry 4, 20
13
Performance of Subcarrier and Power Allocation
Orthogonal Frequency-Division Multiplexing on
Millimeter Wave
N
a
e
m
ah
Mub
a
r
a
ka
h
1
, Rika Wa
h
y
uni
Arsian
ti
2
, Mul
y
ad
i
3
1
Dep
a
rtment of
Elecrical E
ngi
neer
ing,
North
Sumatra Un
ive
r
sit
y
, Meda
n
2
Departme
n
t of Electrical En
gi
neer
ing,
State
Pol
y
t
e
ch
nic of Meda
n, Meda
n
3
Departem
ent of Electrical En
gin
eeri
ng, Born
eo Un
iversit
y
, T
a
rakan
e-mail:
n
aema
h
.mubar
aka
h
@
gmai
l.com
Abs
t
rak
Pemen
u
h
an k
ebutu
h
a
n
bit r
a
te di
atas 4
0
Mbps d
a
p
a
t dil
a
kuka
n d
eng
a
n
siste
m
l
o
ca
l
mu
ltip
oin
t
distrib
u
tion s
e
r
v
ice (LMDS)
y
ang
dio
per
asik
an p
a
d
a
ge
lo
mban
g
mil
i
m
eter
. Sistem
ini
me
mp
uny
ai
mas
a
l
a
h
tersend
iri ketik
a
ditera
pka
n
p
ada
daer
ah d
e
nga
n pote
n
si c
u
rah h
u
j
an ya
n
g
cukup ti
ng
gi, seperti In
don
e
s
ia.
Untuk
itu sist
e
m
LMDS
dik
e
mb
an
gkan
d
e
n
gan
meka
ni
s
m
e cross-l
a
yer.
Pada
p
ene
litia
n i
n
i tek
n
ik
ya
n
g
dig
unak
an pad
a
cross-lay
e
r orthog
on
al
fre
que
ncy-div
i
sio
n
multip
lex
i
ng
(OF
D
M) multi
-
user a
dal
ah j
o
i
n
t
subcarrier and power
alloc
a
tion (JSPA). Diharapkan kinerja
sist
em telek
o
munik
a
si
pada k
anal gelombang
mi
lli
meter me
nin
g
kat meski
terjadi g
a
n
g
g
uan o
l
e
h
red
a
m
a
n
hu
ja
n. Pene
litia
n in
i memba
has ki
ne
rja
kapas
itas trans
mis
i
, data
rate,
utility
da
n fair
n
e
ss d
a
ri
alg
o
rit
m
a
JSPA ya
ng
ditera
pka
n
d
a
r
i
has
il
pe
nel
itia
n
pen
guk
uran
re
da
ma
n h
u
ja
n
di w
ilay
a
h
Sur
abay
a. Dar
i
h
a
sil
pe
nel
itian
dip
e
ro
leh
ad
anya
pe
ni
ngka
t
a
n
kinerj
a
sist
em mela
lui
tekn
i
k
JSPA. Pad
a
kon
d
isi r
e
d
a
m
a
n
h
u
j
an y
a
ng
me
nca
pai
40dB, tek
n
ik J
SPA
ma
mpu
men
i
n
g
katkan
kap
a
si
tas rata-rata
trans
misi
menc
a
pai
17
3,3%,
da
ta rate sist
em
me
nca
pai
18
9,
9%
dan fair
ness sistem
m
e
ncapai 9,6%. Im
plementasi te
k
n
ik J
SPA dapat m
e
ningkatkan utility system
hingga
13,61-
15,4
8
bp
s/H
z
.
Ka
ta
k
unc
i:
al
okasi d
a
ya, al
o
k
asi subc
arrier,
kanal g
e
l
o
mba
ng mill
i
m
eter,
OF
DM, LMDS
A
b
st
r
a
ct
Loca
l
multip
oi
n
t
distributio
n sy
stem (LMDS) t
hat op
erated
in
mil
l
i
m
eter w
a
v
e
s can b
e
use
d
to fulfill
the n
e
e
d
of
bit
rate
hig
her t
h
an
40 M
bps. H
o
w
e
ver it
has
prob
le
m w
h
e
n
app
lie
d i
n
tro
p
i
c
cou
n
try such
a
s
Indon
esi
a
beca
u
se of the hi
gh
rain
fal
l
. Therefore LMDS system w
a
s dev
elo
ped by cross-
la
yer mec
h
a
n
is
m
.
In this res
earc
h
we used joint s
ubcarrier and power allocation (JSPA) te
chnique in
m
u
lti-
user cross-
layer
OFDM. This techni
que
is pr
o
pose
d
to i
n
cre
a
se the
perfor
m
a
n
ce
of telec
o
mmunic
a
tio
n
system ev
en t
here
w
e
re disturba
n
c
e of rain atten
uatio
n. T
he research
is d
i
scus
s
ing the p
e
rfor
ma
nc
e of trans
miss
ion ca
pac
i
t
y,
data rate, util
ity and fairn
e
ss
of JSPA
algorit
hm th
at appl
ie
d in rai
n
attenu
ation
me
asur
e
m
e
n
t in Sura
b
a
ya.
The result shows the increment of
perfor
m
ance system
using JSPA tec
h
niqu
e. For 40 dB ra
in attenuation,
JSPA can achieves respectiv
e
ly aver
age capacity of trans
m
i
ss
ion
up to
173,3%, 189,9
% fo
r data rate and
9,6% for fairness system
. The applic
ation
of JSPA technique im
pr
ove
t
he perfor
m
a
nc
e of utility 13,
61-
15,48 b
p
s/H
z
Ke
y
w
ords
: po
w
e
r allocati
on, subcarri
er al
loc
a
tion,
mil
l
i
m
ete
r
chann
el w
a
ve
s, OF
DM, LMDS
1. Introduc
tion
The develo
p
m
ent
of broa
dban
d se
rvices
te
ch
nol
o
g
y
increa
se
d rapidly. Service
s
tha
t
requi
re the a
v
ailability of
high sp
eed d
a
ta chan
nel
s such a
s
hig
h
spee
d internet, digital vi
deo,
audio bro
a
d
c
asting and
vi
deo confe
r
en
cing co
ntinue
s
to in
crease. The n
eed fo
r the provision
of
high spee
d d
a
ta cha
nnel
can be pe
rformed with lo
ca
l
multipoint distributio
n se
rvice (LM
D
S).
LMDS
syste
m
s a
r
e op
era
t
ed at ran
ge frequ
en
cy 20-40 G
H
z. Syst
em
s that o
p
e
r
ated at
freque
nci
e
s h
i
gher than
10
GHz a
r
e vul
nera
b
le to
i
n
terferen
ce the
re
ceiving
sig
nal cau
s
ed
b
y
rain atten
uati
on. The
r
efore the appli
c
ations
of
LM
DS syste
m
i
n
tropi
cal
co
untrie
s
, su
ch
as
Indone
sia, re
quire
spe
c
ial
handli
ng due
to
rai
n
atte
n
u
a
tion that
o
c
curs at
80
dB
with a
fre
que
ncy
of 29 GHz wit
h
a 5.7 km lo
ng link [1].
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 11, No. 1, March 2
013 : 167 – 1
7
4
168
The novel m
e
ch
ani
sm to increa
se the
quality and cap
a
city of system is cro
s
s-laye
r
mech
ani
sm a
ppro
a
ch [2] [
3
]. It is devel
oped t
he inte
grated
fram
e
w
ork
at Physical
(PHY) la
yer
and medi
a acce
ss
cont
rol (MAC) laye
r.
Cro
s
s-l
a
yer
Appro
a
ch in
multiuse
r o
r
thogo
nal fre
q
uen
cy-divisi
o
n multiplexin
g
(OF
D
M
)
can be obtai
ned by using the joint subcarri
er and power allocation (JSPA) techni
que
and
Adaptive Pa
cket Sched
uli
ng. The
s
e
are supp
ort by
the availa
bili
ty Chan
nel S
t
ate Informati
o
n
(CSI)
and A
r
rival Traffic Informatio
n. The p
r
evio
u
s
re
sea
r
ch was
study the
joint su
bcarrie
r
Allocation a
n
d
powe
r
on chann
el-a
wa
re
queue
-a
ware (CAQA
)
[4] and Re
sou
r
ce Allocation a
nd
Cro
s
s-l
a
yer S
c
he
duling o
n
multica
rrier wireles
s
network
[5] [6].
The
appli
c
ati
on of
JSPA t
e
chnique on
millimet
er wave transmi
ssi
on
i
s
i
n
terrupted by
rain atten
u
a
t
ion. Rain
attenuation
meas
urement
s have b
e
e
n
perfo
rme
d
usin
g opti
c
al
disd
rom
e
ter
gaug
es pla
c
e
d
at ITS S
u
ra
baya [7]. Thi
s
study
is aim
to kn
ow the
b
enefits
of
JSPA
techni
que
s for imp
r
oving
performan
ce
system
on
broa
dba
nd n
e
tworks in tropical co
untri
es,
esp
e
ci
ally Indone
sia to increase the c
a
p
a
city, data rat
e
, utility and fairne
ss.
2. Rese
arch
Metho
d
2.1. Rainfall Measur
e
men
t
Preci
p
itation
mea
s
ureme
n
ts pe
rforme
d in
Suraba
ya usin
g an
optical
disd
romete
r
gaug
es are pl
ace
d
o
n
the
roof of
the
Tel
e
com
uni
catio
n
La
boratory
building.
Opti
cal
Di
sdromet
e
r
work ba
se
d lase
r syste
m
s, optical sen
s
or
with 180
mm wide x 30 mm. Measurem
ents
ca
n be
perfo
rmed
in
real tim
e
, if there
is
a rain pa
rticl
e
p
a
ssing th
ro
u
gh the l
a
ser beam to
de
tect
disd
rom
e
ter rainfall (mm/h
)
and the dist
ribution of
rai
n
. Then the result
s are
sto
r
ed in
softwa
r
e
(Hyd
ra
s an
d
ASDO)
call
e
d
data
parsiv
e
l sh
ape
d ra
i
n
fall data a
n
d txt files. Da
ta obtaine
d fro
m
these me
asurements ASDO rain in
software [7].
2.2.
Genera
ti
on Attenu
ati
on Rain
The
result of
rain
attenuati
on me
asure
m
ent in Su
rab
a
y
a with the
di
stan
ce b
e
twe
en u
s
e
r
and tran
smitt
e
r 1-3 miles:
Stage 1: Generatin
g the value of m,
σ
for four users b
a
sed di
stan
ce ra
ndo
mly generate
d
by
taking a
s
refe
ren
c
e the dat
a in Table 1 [7].
Table 1. The
averag
e and
stand
ard d
e
viation of SST log rain atten
u
a
tion
Link Length(km)
Average (
)
Standard deviati
on (
)
1 -1,0539
2,0574
2 -0,4554
2,1248
3 -0,1321
2,1719
Stage 2: Ge
n
e
rating th
e
co
efficients
of rain att
enu
atio
n ra
ndomly b
y
taking the
referen
c
e
data
in
Table 2 [7].
Table 2. The
correl
ation of rain attenu
ation
usi
ng the
cal
c
ulatio
n of rain attenu
ation SST
Link 1
Link 2
SS
T
1 1
45
0
0,9065
90
0
0,8467
135
0
0,7132
180
0
0,6784
2 2
45
0
0,8484
90
0
0,7434
135
0
0,6139
180
0
0,5996
3 3
45
0
0,8103
90
0
0,6869
135
0
0,5717
180
0
0,5371
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Perfom
ance
of Subca
rrie
r
and Power Al
locati
o
n
OFDM on Millim
eter…
(
Naem
ah
Mubarakah)
169
step 3 : Prep
are 4x4 covariance matrix:
Where [8] :
b
b
a
m
n
SST
m
n
m
n
exp
1
exp
1
exp
exp
ln
1
2
2
,
(4)
2
2
2
2
m
n
m
n
a
(5)
2
2
2
2
m
n
m
n
b
(6)
μ
is the
average
(mea
n),
σ
is the
stan
dard
deviatio
n
,
ρ
SST SST is rain atten
uation
coefficient,
ρ
n, m is the attenuation co
efficient of no
rm
al rai
n
fall, n and m are the links 1 an
d 2.
Step 4: Obtain the rain atte
nuation
valu
e
of each u
s
er
by the equati
on:
A = exp (X)
(7)
x
x
m
Y
C
X
2
/
1
, A as the rai
n
attenuation
values, y is a matrix of 4 1 Gau
ssi
an (
= 0 and
2
= 1
)
an
d m i
s
a vecto
r
of
m. Then
ente
r
the
dat
a
fo
r LMDS syste
m
param
ete
r
s that
can
be
see
n
in Table 3 [9].
Table 3. LM
DS system pa
rameters (k =
1, 38.10
-23
and T
o
= 298 K)
Parameter
Units
Form
u
la
Value
Transmit Po
w
e
r into
Antenna
dBW
Ptx: tra
n
smit pow
e
r
pe
r
c
a
rri
er
0
Transmit antenn
a gain
dBi
Gt:Gant
15
Freque
nc
y
GHz
f: Transmit freq
u
enc
y
30
Path Length
Km
d: Hub to Sub
scriber
Station
Range
2
Field Margin
dB
Lfm : Antenna
Mis-
Alignment
-1
Free
-Space Loss
dB
FSL = -92.45-
2
0
log(f)-
20log(d)
-128,013
Total Path Loss
dB
Ltot = FSL + LF
M
-129,013
Receiver Antenn
a Gain
dBi
Gr =
Gant
30
Effective Bandw
i
d
th
MHz
BRF : Receiver Noise
Band
w
i
dth
80
Receiver Noise Figure
dB
NF : Eff
e
ctive Noise
Figure
5
T
her
mal Noise
dBW/MHz 10log(kTo)
-143,85
S
y
stem Loss
dB
Lsy
s
=
G
t+Ltot+
G
r
-84,013
Received Signal Level
dBw
RSL=Ptx+Ls
y
s
-84,013
Thermal Noise Power
Spectral densit
y
dBW/MHz N0=10log(kTo
)
+NF
-138,859
Carrier t
o
Noise r
a
tio
dB
C/N =
R
S
L-No-
10log(BRF
)
35,8151
2
4
3
4
43
2
4
42
1
4
41
4
3
24
2
3
2
3
23
1
3
13
4
2
24
3
2
23
2
2
1
2
12
4
1
14
3
1
13
2
1
12
2
1
C
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 11, No. 1, March 2
013 : 167 – 1
7
4
170
2.3. Joint Subcarrier a
n
d Po
w
e
r Allo
cation (JSPA)
OFDM
-b
ase
d
LMDS sy
stems a
r
e a
pplied in mi
llimeter wave ch
annel,
cha
nnel
conditions (p
i
(f))
whi
c
h va
ries de
pen
din
g
on
the
w
e
a
t
her, ma
kin
g
i
t
necessa
ry f
o
r
perfo
rma
n
c
e
optimizatio
n sub
c
a
rri
er all
o
catio
n
and
power a
llo
cat
i
on. Dynami
c
Subca
rri
er
Allocation
(DSA)
applied for Subcarrier allocation
and A
daptive Power Allocation
(APA) for P
o
wer all
o
cation.
JSPA simultaneously to optimize cross-l
a
yer to re
ach the optimum
achi
evable value of fairness.
The combined DSA and APA optimization can be formulated with
[5] [6]:
df
f
f
p
U
M
r
U
M
i
D
M
i
i
M
i
i
i
i
)
(
)
(
1
log
1
)
(
1
*
2
1
1
(8)
p * (f) = po
we
r emitted opti
m
um.
2.3.1. D
y
namic Subcarrier
Allocation (DSA)
The pu
rpo
s
e
of the DSA is to improve t
he pe
rform
a
n
c
e of OF
DM-based net
work wh
en
the tran
smission po
wer i
s
uniformly di
stributed th
rou
ghout the fre
quen
cy ban
d. To cal
c
ulate
the
cap
a
cit
y
c
i
(f
) is tran
smitted
can b
e
formul
ated by [5]:
)
(
)
(
1
log
)
(
2
2
f
N
f
H
p
f
c
i
i
i
(9)
Hz
bps
f
f
p
f
c
i
i
))
(
)
(
1
(
log
)
(
2
(10
)
is a con
s
tant
value to give the desired
BER, can be
formulate
d
by:
)
5
ln(
5
,
1
BER
(11
)
and
)
(
f
i
is
a
s
u
bs
cr
ib
er
ch
a
nne
l c
o
nd
itio
ns
o
n
the frequ
e
n
cy f of the user i, where:
)
(
)
(
)
(
2
f
N
f
H
f
i
i
i
(12
)
)
(
f
H
i
is the
ch
ann
el gain
at u
s
er i at
sub
c
a
r
rier
f
r
eq
uen
cy f and Ni
(f) is the
noi
se
power
at
use
r
i at sub
c
arri
er fre
que
n
c
y f.
If the transmit
t
ed po
wer
no
rmali
z
ed,
wh
ere p
(f)
= 1,
then to a
c
hie
v
e the ca
pa
ci
ty of the
freque
ncy f, ci (f), can be e
x
presse
d in Equation 1
3
.
))
(
1
(
log
)
(
2
f
f
c
i
i
(13
)
Each u
s
e
r
or each
se
ssi
o
n
has a
weig
ht
expresse
d
as CSI and
related q
ueui
ng time.
The weight indicates the
utility
function used
for optimization
cro
s
slayer
and the bal
a
n
ce
betwe
en
efficiency
and
fai
r
ne
ss.
Fo
r b
e
st effo
rt tr
af
fic (non
-real
time traffic) a
dopted
a
utility
function
with r = x kbp
s
an
d is given by:
)
3
,
0
ln(
8
,
0
16
,
0
)
(
r
r
U
(14
)
U(r) is value of utility.
The ultimate
goal i
s
to m
a
ximize the
nu
mber
of u
s
ers
whe
r
e the
overall
utility DSA to
maximize ju
st
ice services t
o
use
r
s
(faine
ss) form
ulate
d
:
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Perfom
ance
of Subca
rrie
r
and Power Al
locati
o
n
OFDM on Millim
eter…
(
Naem
ah
Mubarakah)
171
i
D
i
M
i
i
M
i
i
i
df
f
c
U
M
r
U
M
)
(
1
)
(
1
1
1
(15
)
M
is numb
e
r of
user
. Th
e t
o
tal num
ber
of M u
s
er su
bca
rri
er i
s
the
total availabl
e ban
dwi
d
th t
hat
can b
e
formul
ated as follo
ws:
M
i
B
Di
1
,
0
(16
)
M
j
i
j
i
D
D
j
i
,...,
2
,
1
,
an
d
,
(17
)
D
i
is sub
c
arri
er for u
s
e
r
i and ri is the da
ta rate of use
r
i.
2.3.2. Adap
tiv
e
Po
w
e
r
Allocation
(AP
A
)
In order to obtain the BER quality of
s
i
gnal
information we
optimiz
ed the APA performanc
e
,
Water-filling
algorithm is used as a function to
obtain the opt
imal power
allocation. In this
theore
m
, fixed sub
c
a
rri
er a
llocatio
n
is kn
own; the opti
m
um po
wer a
llocatio
n
is eq
uation [5]:
f
r
U
f
p
i
i
i
1
*
'
*
(18
)
is con
s
tanta
of normali
ze
d
optimal power den
sity.
0
,
0
0
,
x
x
x
x
f
p
*
Is optimal po
wer d
e
n
s
ity and
*
i
r
is value of
optimal bit-
ra
te
[5].
Optimal reso
urce allo
catio
n
ca
n not be
di
re
ctly cal
c
ulated by
the
equation
ab
ove that
requi
re
d a
n
it
erative al
go
rithm. In the
water-fillin
g
al
g
o
rithm
ea
ch
use
r
has a
p
a
rticul
ar valu
e of
margi
nal utility, power
obt
ained i
s
com
pare
d
to
the
total transmit power p
e
r u
s
er
po
we
r. If the
achi
evement
of through
put
as a functio
n
of power all
o
cation, then:
df
f
p
f
f
c
i
D
i
i
)
(
)
(
1
(
log
)
(
*
2
(19
)
Utility is the data transmission ca
pacity that is formul
ated:
)
3
.
0
ln(
8
.
0
16
.
0
)
(
r
r
U
(20
)
f
n
ci
r
).
(
and
k
B
f
U
(r) i
s
expressed
as a
utility value, B is
the ba
ndwidt
h and
k i
s
the num
ber
of
subcarriers for all
use
r
s. Th
e ultimate purpo
se of the APA
is to ma
ximize the value of fairness that is the avera
g
e
of all utility users, namely:
M
i
i
r
i
U
M
1
)
(
1
(21
)
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 11, No. 1, March 2
013 : 167 – 1
7
4
172
3. Results a
nd Analy
s
is
The data
gen
erated
und
er
3 km fro
m
the BTS to the use
r
ba
se
d o
n
table 1. Simulation
results fo
r an
area
with le
ss than
3 km f
r
om the BTS
can
not be
rai
s
ed m
o
re tha
n
4 u
s
ers d
u
e
to
a very large
correlation d
a
m
ping. Use
r
limited to four with 90
0
for eac
h
.
Parameters influenced
JSPA technique on m
illimet
er wave are capacity, data rate,
utility and fairness. Table 4
shows a co
mpari
s
o
n
of
the cap
a
city o
f
clear sky co
ndition
s and the
condition affected rai
n
attenuation
wi
th JSPA and without JSPA techni
ques.
Table 4. Co
m
pari
s
on of the
capa
city of clear sk
y condi
tions an
d the effect of rain
attenuation
with JSPA and without JSP
A
techniques.
User
Num
b
er
Distan
ce
(km)
Capaci
t
y
(
bps/
H
z
)
Clear Sk
y
Rain A
t
te
nua
tio
n
W
i
tho
u
t JSP
A
JSP
A
1 2,9137
7,79379
0,2347
1,6582
2 2,7603
7,9492
1,0636
3,6939
3 2,5896
8,1326
0,0494
0,4180
4 2,0452
8,8117
2,5721
4,9424
From
table
4 shows the average capacity
wi
thout JSPA technique
is 0.9799
bps /
Hz.
Meanwhile
with JSPA techniques i
s
2.
6781 bps
/ Hz. This m
eans that
the capacity increase
1.6982 bps / Hz. This i
s
because the JSPA te
chni
que is applied adaptiv
e power allocati
on.
Simulation p
e
r
forme
d
10,
0
00 time
s iterations fo
r mo
re a
c
curate result
s, it sho
w
in figu
re
4
and
5. With
JSPA
tech
niqu
es the
cap
a
city o
f
the av
erage
value i
n
crea
se
s fro
m
1.0
0
2
to 1
0
.49
bp
s /
Hz.
Table 5. Co
m
pari
s
on of dat
a rate
User
Num
b
er
Distan
ce
(km)
Data
Ra
te
(Mb
p
s
)
Clear Sky
Rain A
t
te
nua
tio
n
W
i
tho
u
t JSP
A
JSP
A
1 2,9137
155,8757
4,5765
32,9226
2 2,7603
158,9835
20,7394
76,6932
3 2,5896
162,6521
0,9638
7,6458
4 2,0452
176,2344
50,1554
104,2467
Figure 1. CDF Capa
city
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Perfom
ance
of Subca
rrie
r
and Power Al
locati
o
n
OFDM on Millim
eter…
(
Naem
ah
Mubarakah)
173
Figure 2. CDF Utility
Table 6 sho
w
s
that the average d
a
ta rat
e
without
JS
PA technique
is 19.1088 M
bps a
nd
with JSPA increase to 36.
2683 M
b
ps. Implementati
on of Power Allocation in JSPA technique
cau
s
e in
crea
sing of data rate.
Data
rate val
ue e
qual
to t
he
system
uti
lity.
The g
r
ea
ter the
value
of its
utility it sho
w
s
the better performance of
the system. This is because the
utility associ
ated
with the value of
efficiency and fair
ness.
JSPA techni
ques in
creased the aver
age utility value to
be
13.61
to
15.49 bp
s / Hz.
Table 6. Utility Comparison
User
Num
b
er
Utility
W
i
tho
u
t
D
S
A,
AP
A
and JSP
A
D
S
A AP
A
J
S
P
A
1
12,4292
12,4197
14,0136
14,0077
2 13,6380
13,7029
14,6543
14,6843
3 11,1829
11,0890
12,9111
12,9397
4 14,3445
14,4498
14,8873
14,9298
Figure 3. Gra
ph of Averag
e Cap
a
city
Figure 4. Gra
ph of data rat
e
averag
e
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 11, No. 1, March 2
013 : 167 – 1
7
4
174
Fairn
e
ss valu
es
obtaine
d f
r
om th
e ave
r
age ut
ility for
all users. Th
e
appli
c
ation
o
f
JSPA
techni
que
in
crea
se
s the
fai
r
ne
ss
system
. In cl
e
a
r sky
con
d
ition fai
r
ness i
s
99.99
% above
15,
25.
While the
condition affect
ed rain attenuati
on without JSPA technique fairness above 15 only
86.99% and f
a
irness increase 2.4%
by
using JSPA technique. Tabl
e 7 shows the com
p
arison of
fairne
ss
syste
m
affected rai
n
attenuation.
Table 7. Co
m
pari
s
on of 4 u
s
er fai
r
ne
ss system
C
o
nd
i
t
io
n F
a
i
r
n
e
s
s
Clear sk
y
15,2887
Attenuation Rain
w
i
thout JSPA
12,8987
Attenuation Rain
using JSPA
14,1404
The compari
s
on of capacity, data rate, and
fairness utility of J
SPA technique shows
better pe
rformance impro
v
ement
then without JSPA
techniq
ue.
4. Conclusio
n
JSPA techni
que can increase t
he
average of capacity to
1,002 –
10,49 bp
s/Hz because
it has adapti
v
e power all
o
cation. For
average
of capacity above 7,7 bps/
H
z JSPA techni
que
increa
se
s up
to 9,24%.
The appli
c
ati
on of
JSPA
techni
que im
prove
the performance of
data
rate
and utilit
y
respe
c
tively 36,283 Mbp
s
a
nd 13,61
-15,
48 bp
s/Hz.
For th
e fairne
ss value, it ju
st 86,9
9
% re
ach
ed 15
in
affected rain
attenuation
condition
without JSPA
and fairness incr
ease 2,4% ug JSPA technique.
The impl
ementation of
JS
PA technique in OF
DM
downlink system
at
millimet
er wave
i
s
recomme
nde
d for are
a
wit
h
in high inte
n
s
ity of rain.
Referen
ces
[1]
Jun Y
Z
ang,
Ben
Letai
ef K
.
Cross-lay
e
r
Adaptive
Res
ource
Management for Wireless Pack
et
Net
w
ork w
i
th
OF
DM
Signa
li
ng.
IEEE Transactions
on Wi
reless Comm
unications.
20
04; 5
(
1
1
)
: 32
44
-
325
4.
[2]
Jun Y Z
hang,
Ben Leta
i
ef K.
Adaptive R
e
s
ource All
o
cati
o
n
and Sch
e
d
u
l
i
ng for Multius
e
r Packet-
base
d
OF
DM Net
w
orks.
IEEE International
Confer
ence on
Communic
a
tions
. 2006. 3(
15)
:1565-
15
75.
[3]
Moha
nram C,
Hash
ya
m S. Joint
Su
bcarri
er
and Po
w
e
r Al
locati
on i
n
Ch
ann
el-A
w
a
re Q
ueu
e-A
w
ar
e
Sched
uli
ng for
Multius
e
r OF
DM.
IEEE
Transaction on Wireless Comm
unic
ations.
20
07; 6
(
9).
[4]
Song G, Ye
Li.
Cross-la
ye
r Optimizati
on for
OF
DM
W
i
reles
s
Net
w
orks-p
a
r
t
I:
T
heoretical
F
r
ame
w
ork.
IEEE Transaction on Wireless
Communic
a
tions
. 2005; 4(2):
614-6
24.
[5]
Song G, Ye Li. Cross-la
yer Optimiz
a
ti
on fo
r OF
DM W
i
reless Net
w
orks-P
a
rt II: Algorithm
Devel
opm
ent.
IEEE Transaction on Wireless
Communic
a
tions
. 2005; 4(
2): 625-
634.
[6]
Chu CY, Che
n
KS. Effects of Rain F
adi
ng
on the
Efficien
c
y
of the Ka-B
and LMD
S
S
y
stem in the
T
a
i
w
a
n
Area.
IEEE Transacti
on on Ve
hic
u
la
r Technol
ogy
. 200
5; 54(1): 9-
19.
[7]
Sale
hud
in M,
Han
antas
ena
B, W
ijdema
n
L
.
Ka Ban
d
Li
ne
-of-Sight R
adi
o
Propa
gati
on E
x
peri
m
e
n
t
i
n
Su
ra
ba
ya
In
don
e
s
ia
. 5
th
Ka-Band Uti
l
. Confe
r
ence.T
aormin
a
.199
9:16
1-16
5.
[8]
Mahmu
dah
H. Prediksi R
e
d
a
man H
u
j
an
Meng
gun
aka
n
S
y
nt
hetic
Stor
m
T
e
chniq
ue.
Ph.D. T
hesis.
Surab
a
y
a: Pos
t
gradu
ate IT
S;
200
8.
Evaluation Warning : The document was created with Spire.PDF for Python.