Indonesian J
ournal of Ele
c
trical Engin
eering and
Computer Sci
e
nce
Vol. 2, No. 3,
Jun
e
201
6, pp. 657 ~ 66
7
DOI: 10.115
9
1
/ijeecs.v2.i3.pp65
7-6
6
7
657
Re
cei
v
ed Ma
rch 1
7
, 2016;
Re
vised Ma
y 9, 2016; Acce
pted May 2
7
, 2016
The Impact of LTE-FDD at the LTE-TDD for the Co-
Existence under 2.6 GHz Band for Malaysia
LM Ahmed*,
MFL Abdulla
h
Univers
i
t
y
T
un Hussei
n
Onn
Mala
ysi
a
864
00 Par
i
t Ra
ja, Batu Pah
a
t, Johor, Mala
ys
i
a
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: labe
eb
330
@
g
mail.c
o
m
Ab
stra
ct
For the co-exis
t
ence scen
a
ri
o
betw
een LTE-
FDD
and LTE-
TDD systems, if
the tw
o systems
are
usin
g an a
d
j
a
c
ent frequ
ency
carrier, there
w
ill be a n
e
e
d
for spatial s
e
p
a
ratio
n
betw
e
e
n
the eN
ode
Bs
of
the tw
o system
s, otherw
i
se the tw
o
system
s w
ill interfere each other. Th
e study is im
pl
em
ented based on
realistic
p
a
ra
meters i
n
order
to he
lp
the
net
w
o
rk desi
g
n
e
r
to
make
a
d
e
c
i
sion
a
b
o
u
t the
best fr
equ
enc
y
alloc
a
tio
n
an
d netw
o
rk depl
o
y
me
nts in ord
e
r to ac
hiev
e hig
her perfor
m
ance u
n
d
e
r the low
e
st possi
b
l
e
cost. Throughout this
paper, t
he
e
ffect of the FDD syst
em at the TDD
is
evaluated
under wide range
of
ACIR and s
e
p
a
ratio
n
dista
n
c
e
s betw
een th
e tw
o system
s
eNod
eBs as
w
e
ll. The resul
t
s show
ed that, the
reco
mme
nde
d ACIR offset by the 3GPP is not enou
gh fo
r the LT
E-T
DD u
p
link thro
ug
hp
ut loss ratio to be
accepta
b
l
e
, w
hereas
115
dB,
45 dB, a
nd
3
5
dB of the A
C
IR is req
u
ire
d
for the thro
u
ghp
ut loss rati
o i
n
order to
drop
l
e
ss than
5% f
o
r the co-l
ocat
ed, Mid-
poi
nt, and E
d
g
e
-po
i
n
t
eNod
eBs de
ploy
ment scen
a
ri
o
respectiv
e
ly. Meanw
hi
le, co
mpari
ng to the u
p
link c
a
se
, the
dow
nli
n
k of the
TDD system i
s
muc
h
coh
e
re
nt;
the rec
o
mmen
ded
ACIR
offse
t
is o
n
ly
un
acc
eptab
le
fo
r th
e
co-loc
ated
de
pl
oyment c
a
se,
w
hereas
50
dB
o
f
the ACIR is req
u
ire
d
for the system to dro
p
le
ss than 5%.
Ke
y
w
ords
: LT
E-co-existe
n
ce
, interference,
LT
E-T
DD, LT
E-F
DD, ACIR
Copy
right
©
2016 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
Acco
rdi
ng to
the 3GPP stand
ardi
zati
on, t
he LTE system is
meant to su
pport hig
h
throug
hput a
nd low l
a
ten
cy, improved
coverage i
n
orde
r to kee
p
the path
with the increa
sing
global d
e
man
d
. Becau
s
e o
f
the sca
rcity of the spe
c
trum re
sou
r
ces, the solution
is to pro
pag
a
t
e
differen
c
e
sy
stem
s withi
n
the same
ge
ogra
phi
cal
area. In [1], th
e freq
uen
cy
allocation fo
r the
LTE is divide
d into t
w
o
m
a
in p
a
rt
s, un
paire
d
spe
c
trum a
nd
paire
d spe
c
trum
for T
D
D a
n
d
FDD
respe
c
tively, the unp
aire
d
spe
c
tru
m
u
s
e
s
only o
ne fre
quen
cy ba
nd
for both u
p
lin
k an
d do
wnli
nk
operation
s
,
whe
r
ea
s, th
e
pai
red
spe
c
trum u
s
e
s
two sepa
rate
d
freque
ncy
all
o
catio
n
s for
th
e
uplin
k and do
wnlin
k a
s
it is explained in
[2], eac
h one
of the paired
and the unp
aired frequ
en
cy
allocation
ha
s a
d
vantag
es and
di
sadva
n
tage
s in
the
term
of ave
r
age th
ro
ughp
ut, flexibility, and
the efficien
cy of allocating
the available
frequen
cy. The majo
rity of LTE operat
ors p
r
efe
r
u
s
ing
the paired
sp
ectru
m
(F
DD) mode. Howe
ver,
nowaday
s the LTE
-
T
D
D is evolve
d
and be
ca
me a
mature te
chn
o
logy rathe
r
than ju
st a co
mpleme
ntary techn
o
logy.
The M
a
laysia
n Standa
rd
Radio System
Plan (S
RSP)
has sp
ecifie
d
the requi
rem
ents fo
r
the LTE
co
-e
xistence un
d
e
r the
freq
ue
ncy b
and
s b
e
twee
n 25
00
MHz a
nd
2
690 M
H
z in
[3].
Whe
r
ea
s, in
the near fu
ture, Malaysi
a
is goin
g
to coexi
s
t LT
E-TDD and
LTE-F
DD u
n
der
themention
e
d
frequ
en
cy b
and
whi
c
h i
s
not only
rev
e
rsed fo
r Ma
laysia, it is
d
i
vided am
on
g
Malaysia
an
d
its nei
ghb
or co
untrie
s
B
r
unei, an
d Si
ngap
ore. In
the
wirel
e
ss
comm
uni
cati
on
system
s, ge
n
e
rally the inte
rfere
n
ce is
no
t comple
tely
avoidable,
bu
t at least it ca
n bemitigate
d
if
it is firstly evaluated.
Before
coexi
s
ting LTE
-
T
D
D an
d LTE-FDD
sy
st
em
s, this
study
has to
be p
e
rform
ed
based o
n
th
e pre-a
g
re
ed
frequ
ency
a
llocatio
n
a
s
a preca
u
tion
ary procedu
re. Otherwi
se
,
a
mutual int
e
rf
eren
ce
can
p
r
oba
bly be
a
r
isen
bet
wee
n
the t
w
o
syst
ems,
whi
c
h
can d
a
mag
e
t
h
e
two sy
stem
s’
data an
d con
t
rol chann
els
as
well. Th
erefore, the
be
nefit of why t
he co-exi
sten
ce
has b
een d
e
signed for in th
e first pla
c
e cannot be g
a
in
ed.
As the ITU
recom
m
en
dati
ons, the
co
-exist
en
ce
sh
ould b
e
un
d
e
r adj
acent frequ
en
cy
band
s [4]. F
o
r thi
s
p
ape
r, the
coexi
s
ten
c
e
bet
ween
L
T
E-TDD and
LTE-F
D
D
sy
stems unde
r
t
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 25
02-4
752
IJEECS
Vol.
2, No. 3, Jun
e
2016 : 657
– 667
658
freque
ncy ba
nd 2500
-2
69
0 MHz i
s
goi
ng to be invest
igated; e
s
p
e
cially the impact of LTE-FDD
at LTE-TDD
will be presented for Malay
s
ia.
The S
R
SP a
l
so
provid
es
the minim
u
m
req
u
irement
s for sha
r
ing
the fre
que
n
c
y ba
nd
betwe
en M
a
laysia
and
its n
e
igh
b
ourin
g
cou
n
t
ries Si
nga
p
o
re
and
Brunei, te
chni
cal
cha
r
a
c
teri
stics of radio
systems,
freq
uen
cy ch
ann
el, coordi
nation i
n
itiatives in o
r
der to m
a
ximi
ze
the utilization,
minimize inte
rfere
n
ce
and
optimize the
usa
ge of the band
width.
Table 1 and Table 2 illustrate the worst
case
scenari
os of frequen
cy allocation for co-
existing the
TDD
and F
D
D syste
m
s fo
r Malaysi
a
when con
s
ide
r
i
ng its neig
h
b
ourin
g co
untries
Singapo
re a
n
d
Bernie.
Table 1. The
co-existen
ce
freque
ncy all
o
catio
n
for do
wnlin
k case
Cou
n
tr
y
FDD
d
o
w
n
link
MHz
TDD L
T
E
MHz
Singap
o
re
2624-263
0
2606-261
2
Brunei
2624-263
0
2612-261
8
Table 2. The
freque
ncy all
o
catio
n
for co
-existen
ce fo
r uplink
ca
se
Cou
n
tr
y
FDD
d
o
w
n
link
MHz
TDD L
T
E
MHz
Singap
o
re
2564-257
0
2606-261
2
Brunei
2558-256
4
2600-260
6
In [5], it is recom
m
en
ded
that the e
N
odeBs of F
D
D a
nd T
D
D
sho
u
ld n
o
t b
e
pla
c
ed
together if th
e two
of th
e
m
are u
s
in
g
the adj
acent freque
ncy
ca
rrie
r
s,
be
cau
s
e so
me of
the
physi
cal data and control chann
els will
experience severe
ad
jacent channel
interference.
Con
s
e
quently
, it will be unable to be dem
odulate
d
co
rrectly.
The stu
d
y in [6] also co
ncl
ude
s that, for the co-exist
ence betwee
n
the LTE-T
DD a
nd
LTE-F
DD in
adja
c
ent freq
uen
cy band the interfe
r
en
ce
should b
e
taken into a real con
s
id
era
t
ion
to insu
re th
e
quality of the
data tra
n
smi
s
sion
and
to a
c
hieve th
e g
o
a
l of why the
coexi
s
ten
c
e h
a
s
been ma
de fo
r in the first pl
ace.
In gene
ral, the interferen
ce issue h
a
s
been
inve
stig
ated many times b
e
fore,
and the
r
e
are al
so p
r
op
ose
d
sol
u
tion
s su
ch a
s
in [7]. Howeve
r, throug
hout thi
s
wo
rk, mo
re
spe
c
ifically, the
study evaluat
ed the FDD sy
stem inte
rferein
g
sig
nal
s whi
c
h a
ffect the TDD sy
stem. Therefore,
the u
s
er op
e
r
ator can ev
aluate
the
sy
stem d
a
ma
g
e
ratio.
Co
nseque
ntly they will b
e
a
b
le
to
make
a de
ci
sion
abo
ut the eNo
deB di
stributio
n sc
e
nario, the b
e
st frequ
en
cy allocation, an
d a
prop
er inte
rfe
r
en
ce mitigati
on mechani
sm.
The ACI
R
is
Adjace
nt Ch
a
nnel to Interf
eren
ce
Ratio,
from [8] rep
o
rt the ACI
R
can
be
defined a
s
th
e ratio of the
total powe
r
t
r
an
smitt
ed from a sou
r
ce
(eNodeB o
r
UE) to the to
tal
interferen
ce power affecti
ng a victim receive
r
(e
No
deB or UE), resulting fro
m
both transmitter
and re
ceive
r
i
m
perfe
ction
s
the ACIR is
calcul
ated u
s
in
g the followin
g
equatio
n:
1
1
1
(1)
whe
r
ea
s the
ACLR i
s
the
Adjacent Ch
annel lea
k
a
g
e
power ratio, it is a ratio of th
e
transmitted p
o
we
r to
the
p
o
we
r m
e
a
s
u
r
ed afte
r a
re
ceiver filter in
the a
d
ja
cent
RF
ch
ann
el,
and
ACS stan
ds f
o
r Adja
ce
nt Cha
nnel Sel
e
ctivity, it
is th
e ratio of the
receiver filte
r
attenuatio
n
on
the assig
ned
cha
nnel f
r
e
quen
cy to th
e re
ceiv
e
r
filter attenu
atio
n on the
adj
ace
n
t ch
ann
el
freque
ncy.
Acro
ss this
pape
r, se
ctio
n 2 is g
o
ing
to cover th
e SINR the
o
r
y, and the
system
modelling. Section
3 will
be focused into
simulation result a
nd the analyses,
whil
e section 4
will
discu
ss the
simulation
result
s. Eventually
se
ction 5, include
s the con
c
lu
sio
n
s and
recomme
ndat
ions.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJEECS
ISSN:
2502-4
752
The Im
pact of LTE-FDD at the LTE-T
D
D for the Co-E
xiste
n
ce und
er 2.6 G
H
z …
(LM Ahm
ed)
659
2. The Sy
ste
m
Modeling
The Figu
re
1 sho
w
s the
sce
na
rio of
the FDD u
p
link a
nd F
D
D do
wnlin
k
system i
s
interferi
ng th
e TDD
uplin
k system,
as
well the
Fig
u
r
e 2
sho
w
s the
scena
rio
of the F
D
D u
p
link
and FDD do
wnlink
system i
s
interf
e
r
ing t
he TDD do
wn
link sy
stem.
Figure 1. The
interfere
n
ce of the TDD u
p
link a
nd do
wnlink at the F
DD u
p
lin
k
Figure 2. The
interfere
n
ce of the TDD u
p
link a
nd do
wnlink at the F
DD d
o
wnlink
2.1. SINR
Acco
rdi
ng to
[9] the p
e
rfo
r
mance
of the
wi
re
le
ss cell
ular
can be e
v
aluated ba
sed
o
n
th
e
received
sign
al com
pared
to the interference
and
noi
se (SI
N
R), which
can
be calcul
ated u
s
i
ng
Eq. 2. The t
h
rou
ghp
ut of
the sy
stem
signifi
cantly
can
be d
e
g
r
aded
be
cau
s
e of two typ
e
o
f
interferen
ce. Firstly,
the Co-Chan
nel
In
terfere
n
ce (CCI),
which m
ean
s the
co
mbination
of
the
interferen
ce
signal
s fro
m
th
e UE
s
or
eNo
deBs th
o
s
e
b
e
long
to the
same
system.
Secon
d
ly, the
Inter Cha
nnel
Interferen
ce
(ICI), whi
c
h
mean
s t
he combinatio
n of the interfere
n
ce si
gnal
s from
the UEs o
r
e
N
od
eBs tho
s
e belon
g to the other
syste
m
within the same propa
ga
tion area.
(2)
Whe
r
ea
s S is the received sign
al, and N
is the noi
se.
The stu
d
y consi
dered two UE
s d
epl
oyment scen
ario
s, whe
r
e
a
s the T
D
D UEs will
rand
omly b
e
distrib
u
ted
overall th
e T
DD eNodeB
s,
which
is called
the n
o
rm
al d
i
stributio
n. Th
e
se
con
d
di
stri
bution i
s
call
ed the
ed
ge
distrib
u
tion,
whe
r
ea
s,
the
TDD UEs
i
s
only
lo
cated
at
the
edge of the T
DD e
N
o
deB.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 25
02-4
752
IJEECS
Vol.
2, No. 3, Jun
e
2016 : 657
– 667
660
2.2. The Mathematical M
odelling
From
the
sim
u
lation
pro
c
e
s
s in
Figu
re
3, at the
first
of all, the
T
D
D e
N
o
deB
s a
nd F
D
D
eNo
deBs a
r
e
con
s
ide
r
ed i
n
stalle
d at the same
pl
ace whi
c
h mea
n
s the sepa
ration dista
n
ce
betwe
en th
e
m
is zero
me
ters. Se
co
ndl
y, each
time t
he T
D
D e
N
o
deBs is goin
g
to be
move
d
by
20 m
e
ters, t
he
shifting
wi
ll co
ntinue
till the T
D
D e
N
odeBs b
e
co
mes at the
e
dge
of the
F
D
D
eNo
deBs whi
c
h
mea
n
the
distan
ce
bet
wee
n
them
i
s
300
0 m
e
ters (the
ra
diu
s
o
f
the e
N
od
eB
s)
.
For ea
ch shi
fting process a range of -50 dB up to 150 dB of ACIR offset will be appl
ied
con
s
id
erin
g a
n
increm
ent
of 5 dB p
e
r
each ste
p
. T
he p
r
eviou
s
pro
c
e
s
ses wil
l
be repe
ated
for
each po
we
r
control p
a
ra
meters sets
and b
o
th UE
s di
strib
u
tion
scena
rio
s
, th
e othe
r si
mul
a
tion
para
m
eters is in the Table 4 from the 3G
PP in [10].
Figure 3. The
general sim
u
lation algo
rith
m
For th
e
wireless
syste
m
the p
r
o
p
agated
si
gn
al is affecte
d
by the
e
n
vironm
ent
para
m
eters [8], the effect
of these pa
ra
meters
are di
fferently mod
e
lled in equ
ations de
pen
di
ng
on the type of
the nature of
the tran
smitter, t
he receiver, and th
e p
r
opag
ation pa
rameters, the
s
e
equatio
ns a
r
e
called p
a
thlo
ss m
odel
s eq
uation.
The Eq. 3 is pathlo
ss mo
d
e
l for the free
spa
c
e lo
ss from [11]:
Evaluation Warning : The document was created with Spire.PDF for Python.
IJEECS
ISSN:
2502-4
752
The Im
pact of LTE-FDD at the LTE-T
D
D for the Co-E
xiste
n
ce und
er 2.6 G
H
z …
(LM Ahm
ed)
661
1
0
4
1
0
(3)
Whe
r
ea
s, n i
s
num
ber
of the sp
ecifi
c
e
N
od
eB, on is numbe
r of the othe
r eNo
deB,
λ
is
the wavele
ng
th, R is the
distan
ce
bet
wee
n
the e
N
odeBs
numb
e
r (n) a
nd th
e other
eNod
eB
numbe
r (o
n),
d is the avera
ge se
paratio
n
betwee
n
the row of the bui
lding
s
.
The pathlo
s
s betwe
en the
UEs an
d eNo
deBs can be
cal
c
ulate
d
usi
ng the Eq. 4 in [10].
161
.40
7.1
7
.
5
24.
37
3
.7
43
.
422
3
.1
3
20
3
.
2
11.75
4
.97
(4)
Whe
r
ea
s,
W is
stre
et wi
dth, h
i
s
ave
r
age
hei
ght
of buildin
gs,
f
c
is the t
r
an
smissio
n
freque
ncy, h
e
N
odeB
the highest of th
e
eNo
deB, d
eNodeB-U
E
the distan
ce b
e
tween the
UE
and
eNodeB in meters
, h
UE
the
highe
st of the UE.
The pathlo
s
s betwe
en the
UE-UE can b
e
cal
c
ulate
d
usin
g Eq.5 in [12].
_
20
2
√
2
1
0
2
1
1
2
10
2
∆
2
1
1
2
(5)
Whe
r
ea
s,
∆
h
b
is the heig
h
t difference
betwee
n
the eNo
deB a
n
tenna a
nd the mean
building
rooft
op heig
h
t, b is the average
sepa
rati
on between
rows of buil
d
ings,
∆
h
m
is
the
differen
c
e b
e
twee
n the me
an buildi
ng h
e
ight and the
mobile ante
n
na heig
h
t, x is the ho
rizont
al
distan
ce
bet
ween th
e
UE a
nd the
diffra
c
ting e
dge
s, h i
s
the
average
heig
h
t of buil
d
ing, h
m
is
the
height of UE, h
b
height of eNod
eB.
The tran
smitted po
wer of
eNo
deB num
ber (n
) pe
r UE numb
e
r (u) is fixed it can b
e
cal
c
ulate
d
usi
ng the Eq.6.
∗
∗
(6)
Whe
r
ea
s,
R i
s
the
num
ber of RB
per UE, U
mea
n
s
numbe
r of th
e a
c
tive UE
s, M is the
numbe
r of th
e all availabl
e RBs i
n
ea
ch cell,
the m
a
ximum tran
smitted power
from the e
N
o
deB
numbe
r (n
).
The
UE tran
smissio
n
p
o
w
er to th
e e
N
od
eB
can
be
cal
c
ulate
d
u
s
ing
the
followin
g
equatio
n:
∗m
i
n
1
,
m
a
x
,
γ
(7)
Whe
r
ea
s, P
t
i
s
the tran
smi
tted power i
n
dB, P
ma
x
is the maximu
m
allowed tran
smitted
power of the
UE, R
min
is the minimum
redu
ction value
which p
r
events the
UE in the good
cha
nnel
co
n
d
ition not to
tran
smit at
very low
po
wer,
CL i
s
t
he coupli
ng
loss that
ca
n be
cal
c
ulate
d
by the equatio
n bello
w.
CL =
max {path los
s
- G_Tx - G_Rx, MCL}.
MCL i
s
the
m
i
nimum
cou
p
l
i
ng lo
ss,
CL
x
-
ile
is pe
rcenta
ge of
UEs
wh
ich h
a
ve the
highe
st
cou
p
ling lo
ss, and con
s
eq
uently they will tran
smit
at P
ma
x
,
and finally
γ
i
s
a
balan
cing f
a
ctor
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 25
02-4
752
IJEECS
Vol.
2, No. 3, Jun
e
2016 : 657
– 667
662
betwe
en UE
s with bad
ch
annel
con
d
itions to the
UEs with the
good
cha
nne
l conditio
n
s it
is
rang
ed a
s
0
<
γ
<=1. Th
e TA
BLE II includ
es the p
o
wer
control pa
ra
meters
γ
an
d C
L
x
-
ile
which
are
recomme
nde
d by 3GPP techni
cal repo
rt
in [9].
Table 3. Power co
ntrol al
g
o
rithm pa
ram
e
ter set
s
Parameter s
e
t
Gam
m
a
(
γ
)
CL
x-ile
5 MHz
ban
d
w
i
dth
Set 1
1
112
Set 2
0,8 129
Thereby, the
received
sign
al (S) from th
e UE
s
at the
eNo
deB can
be calculated
usin
g
the equatio
n bello
w:
(8)
Due to
the fu
ll ortho
gonall
y
of the LTE
system, the
r
e will n
o
t be
interferen
ce
with the
UEs th
ose b
e
long to
the
same
cell, the interfe
r
en
ce only
come
s from the
oth
e
r a
d
ja
cent
cells
that are
u
s
in
g the
sam
e
RBs
of
the specifi
c
UE,
the
u
p
lin
k co-cha
nnel
interferen
ce ca
n be
cal
c
ulate
d
usi
ng the Eq.9, and the do
wn
link inte
rfe
r
e
n
c
e can be
cal
c
ulate
d
usi
n
g
the Eq.10.
,
,
∗
,
,
(9)
,
∗
,
,
(10
)
The ICI is divi
ded into four t
y
pes:
1-
The re
ceive
d
sign
als at the
eNod
eB fro
m
the UEs which b
e
long t
o
the other
system:
,
,
∗
,
∗
(11
)
2-
The re
ceive
d
sign
al at the UE from
the
UEs which be
long to the other sy
stem:
,
,
∗
,
,
,
∗
(12
)
3-
The re
ceive
d
sign
al at the eNo
deB from
the eNod
eBs which belo
n
g
to the other system:
,
∗
,
∗
(13
)
4-
The re
ceive
d
sign
al at the UE from
the
eNo
deBs
whi
c
h bel
ong to the other
syst
em:
,
∗
,
∗
(14
)
Whe
r
ea
s, ACIR is the attenuation fa
ctor wh
ich deg
ra
de the effect
of the interference.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJEECS
ISSN:
2502-4
752
The Im
pact of LTE-FDD at the LTE-T
D
D for the Co-E
xiste
n
ce und
er 2.6 G
H
z …
(LM Ahm
ed)
663
3. Simulation and Re
sult
The inte
rference evalua
tion mechan
ism is
perf
o
rme
d
acco
rding to the
system
throug
hput lo
ss; the
re
sult
s are plotted
as 3
-
D
figu
res. Wherea
s, the
z-axis repre
s
e
n
ts th
e
percenta
ge
o
f
the throug
h
put lo
ss, th
e
x-axis re
pre
s
ent
s the
co
nsid
ere
d
sep
a
ration
di
sta
n
ce
betwe
en the
TDD
eNodeB
s and th
e
FDD eNodeB
s, and eventu
a
ll
y t
he y-axis for the ap
plie
d
ACIR value for ea
ch po
we
r cont
rol sets
and UE
s dist
ribution sce
n
a
r
io.
Table 4. The
prop
agatio
n e
n
vironm
ent a
s
sumed p
a
ra
meter
Parameter
A
s
s
u
mpti
on
(co
m
m
on)
Environment
Macro cell, Urba
n area, U
n
coordi
nated
deplo
y
ment
Carrier f
r
eque
ncy
2500-269
0 MHz.
Cellular
lay
out
Hexag
onal grid,
19 cell sites, 57 sectors w
i
th
eNodeB in the co
rner of t
he cell.
eNodeBs centre
to centre distance (R)
6000m
Building to building distance (d)
80m
The Height of
the
eNodeB
30m
The
w
i
dth of t
he
streets
20m
The height of t
h
e
UE
1.5m
eNodeB ante
nna
gain
(include feeder lo
ss)
15 dBi
eNodeB ante
nna
height
30 m
log-normal fade
shado
w
10 dB
MCL (including antenna gain)
70 dB
w
h
ite n
o
ise po
w
e
r densit
y
-174 dBm/Hz
eNodeB noisefigure
5 dB
UE noisefigure
9 dB
sy
stem b
and
w
i
dt
h
5 MHz
eNodeB ma
x T
x
po
w
e
r
61 dBm
U
E
m
a
x
T
x
pow
er
2
3
d
B
m
UE min Tx p
o
w
er
-40 dBm
R
mi
n
-64dB
number of active
UEs
3 UEs per
site for
dow
n
link case
1 UE per site for
uplink
Among th
e
whole
ran
g
e
of
the
sep
a
ration di
stan
ce
b
e
twee
n the
T
D
D e
N
od
eBs and
the
FDD eNodeB
s, thre
e e
N
od
eBDepl
oyme
nts Scena
rio
s
(e
DS
s) are i
n
vestigate
d
i
n
detail
s
in te
rm
of the sep
a
ra
tion distan
ce
betwee
n
the
FDD
e
N
o
d
e
B
s and the T
DD e
N
o
deB
s, firstly the Co-
locate
d e
D
S
(Ce
D
S),
which me
an
s the
se
pa
ration
distan
ce
is zero
mete
r. S
e
co
ndly, at t
he
sep
a
ratio
n
di
stan
ce
of 15
00 mete
rs, this rep
r
esents the Mid
-
poi
n
t
eDS (Me
D
S). Finally, at t
h
e
sep
a
ratio
n
di
stan
ce of 3
0
00 which me
ans th
e e
N
o
deBs of the
two
system
s
are lo
cate
d a
t
the
edge
of e
a
ch
othe
r
whi
c
h i
t
is
calle
d Ed
ge e
D
S
(Ee
D
S) such a
s
in
the Fi
g.4 fo
r
the sub
-
figures
(a), (b), and
(c) respe
c
tively.
Figure 4. The
eNod
eBs de
ployment sce
nario
s
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 25
02-4
752
IJEECS
Vol.
2, No. 3, Jun
e
2016 : 657
– 667
664
The stu
d
y has also incl
ud
ed the User
Equipme
n
t distributio
n sce
nario a
s
on
e of main
para
m
eters,
whe
r
e t
hey
are, th
e Edg
e
Use
r
Eq
ui
pment
distri
b
u
tion
scena
ri
o (E
UDS
) a
nd
Normal
User
Equipme
n
t di
stributio
n
sce
nario
(NUDS), such a
s
recommen
ded i
n
[13], whe
r
ea
s
the cell-edge
use
r
equip
m
e
n
t thro
ugh
put
is con
s
ide
r
e
d
on
e of
the
main
ch
allen
g
ing i
ndi
cator of
LTE-A to fulfil Internati
onal Tel
e
co
mmuni
cation
Unio
n –
Radi
o com
m
unication
Sector
(ITU
-R
).
Figure 5. The
through
put lo
ss of the T
D
D uplink
syste
m
The T
DD
upli
n
k throughput loss
ratio
(T
LR) i
s
illustrated in the Figu
re 5 for the
Normal
UEs Di
stri
but
ion Scen
ario
(NUDS) a
n
d
Edge UEs
Distri
bution S
c
en
ario (E
UDS) re
sp
ecti
vely,
whe
n
the
sy
stem i
s
inte
rf
ered
by the
uplin
k
a
nd d
o
wnli
nk of th
e FDD
sy
ste
m
. The
re
sul
t
s
sho
w
e
d
that,
the TL
R i
s
al
most the
sam
e
for t
w
o
po
wer
cont
rol p
a
rameters
(set1 an
d set 2
)
a
nd
for the t
w
o
case
s
of UE
s Di
stributio
n
Scena
rio
s
(UDSs) b
e
ca
use of the
narrow tran
smission
cha
nnel (only
5 MHz). It is
also
sho
w
ed t
hat the
FDD system
inte
rferen
ce
s the u
p
link of the T
DD
system
pretty badly, spe
c
ifically
at th
e CeDS
,
wh
erea
s 115
d
B
of ACI
R
i
s
re
quired fo
r the
system th
rou
ghput lo
ss to
dro
p
le
ss th
an 5%
co
n
s
i
derin
g the t
w
o UE
s di
strib
u
tion. Mea
n
while,
the req
u
ire
d
ACIR is o
n
ly 45 and 3
5
dB
for the ca
se
s MeDS and E
e
DS. Le
ss th
an all, the ed
ge-
point eNo
d
e
B
s deployme
nt achieved t
he lowe
st AC
IR offset to
achieve le
ss than 5% of
the
throug
hput lo
ss a
s
offset o
f
50 dB for the two ca
se of
UDSs.
Figure 6. The
through
put lo
ss of the T
D
D downli
n
k
system
115
45
35
115
45
35
115
45
35
115
45
35
CEDS
MEDS
EEDS
ACIR of
fset (dB)
PC1
NUDS
PC2
NUDS
PC1
EUDS
PC2
EUDS
0
15
25
0
15
25
50
20
10
50
20
10
CEDS
MEDS
EEDS
ACIR of
fset (dB)
PC1
NUDS
PC2
NUDS
PC1
EUDS
PC2
EUDS
Evaluation Warning : The document was created with Spire.PDF for Python.
IJEECS
ISSN:
2502-4
752
The Im
pact of LTE-FDD at the LTE-T
D
D for the Co-E
xiste
n
ce und
er 2.6 G
H
z …
(LM Ahm
ed)
665
FROM the Fi
gure
6 it cle
a
rly appe
ars
that, t
he TDD do
wnlin
k i
s
less affe
cte
d
by the
FDD
system
comp
ared to the uplin
k
reg
i
on of the system. Wherea
s,
the worst case
scena
rio
is
recorded
at the CeDS,
wh
erea
s
50 dB i
s
requi
red to
achi
eve thro
u
ghput lo
ss ra
tio of less tha
n
5% co
nsi
d
e
r
i
ng the
for th
e EUDS
whil
st a
0 dB
of
the A
C
IR i
s
re
quired fo
r the
ca
se
of
th
e
NUDS. For the MeDS, the ACIR ratio
minimized
to 20 dB for the EUDS, mea
n
whil
e the value
increa
sed
for the
NUDS
b
y
15 dB. F
o
ll
owin
g th
e
sa
me ma
nne
r f
o
r th
e EUDS,
the A
C
IR
ke
p
t
minimizi
ng fo
r the EUDS
whe
r
ea
s it reatch
ed onl
y
10 dB. As well, the the
ACIR ratio
kept
increasing till it became
25 dB for the NUDS.
4. Discussio
n
The 3GPP in [14] recom
m
ende
d the minimum AC
IR offset in Table 5, wh
erea
s, it
sho
u
ld
be ta
ken into
a
re
al
co
nsi
d
e
r
atio
n. Acco
rding
to
the simul
a
tion re
sults
th
e
table
conte
n
t
con
c
lu
ded th
e results in the Figure 7 and Figu
re
8
for the uplink and do
wnli
nk of the TDD
system throu
ghput lo
ss at
the of
the TDD syste
m
re
spectively.
Table 5. The
minimum recommen
ded A
C
IR offset.
The Link
A
C
I
R
Off
set
eNode
B -> UE
32.7
eNode
B -> e
N
o
d
eB
41.2
UE -> eNo
d
eB
29.8
UE -> UE
28.2
Figure 7. The
expected through
put
loss
for the TDD u
p
link
ca
se
The Fi
gure 9
explain
the
severity of
th
e F
D
D uplin
k and
do
wnlin
k inte
rferen
ce at the
uplink of the
TDD sy
stem, it can be conclude t
hat, 100% of the upl
ink th
roughput will be lost for
the Ce
DS a
n
d
Me
DS, except for the E
e
DS, wh
ere
a
s
the thro
ughp
ut loss ratio i
n
only 50%
which
it stills quite b
e
yond the a
c
cepta
b
le ra
ng
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 25
02-4
752
IJEECS
Vol.
2, No. 3, Jun
e
2016 : 657
– 667
666
Figure 8. The
expected through
put lo
ss
for the TDD d
o
wnli
nk
ca
se
From th
e Fig
u
re
8, it is
o
b
vious th
at the
do
wnli
nk
of the T
D
D
system i
s
no
t quietly
affected by t
he FDD
syst
em co
mpa
r
e
d
to the up
lin
k case, whe
r
eas th
e thro
u
ghput lo
ss
ra
tio is
only una
ccep
table for the Ce
DS ca
se, mean
while fo
r the other two eDSs the th
roug
hput lo
ss will
remai
n
less than 5%.
5. Conclusio
n
The co
-exi
ste
n
ce b
e
twe
en
LTE-F
DD a
n
d
LTE-T
DD
systems
cann
ot be done, so long a
s
system
s a
r
e
usin
g an a
d
j
a
ce
nt frequ
e
n
cy carrie
r, a
nd propa
gati
ng in the
sa
me geo
gra
p
h
i
cal
area, the im
possibility because of the uplink of t
he
T
DD system will
suffer
a sever
interference
whe
n
co
nsi
d
ering the lo
west re
com
m
e
nded ACI
R
o
ffset. The result also sho
w
ed that, a little
different b
e
twee
n the t
w
o po
we
r
con
t
rol pa
ram
e
ters, thi
s
different
be
cau
s
e of the
na
rrow
con
s
id
ere
d
transmi
ssion
chann
el. Fo
r t
he
syst
em
to work
pro
p
e
rly, the
stu
d
y su
gge
sts
an
enou
gh
spati
a
l sepa
ration
and
ACIR
offset
shoul
d
be co
nsid
ered,
on
the other han
d an
interferen
ce
mitigation me
cha
n
ism
sho
u
ld be con
s
id
ered a
s
well.
Ackn
o
w
l
e
dg
ment
The autho
rs woul
d like to than
k the univ
e
rs
ity Tun
Hu
ssein On
n Malaysia (UT
H
M) for
giving an op
p
o
rtunity to do this re
sea
r
ch.
Referen
ces
[1]
H Z
a
rrink
ou
b.
“Und
erstand
in
g LT
E
w
i
th
MAT
L
AB:
F
r
om Mathematic
al
Mode
lin
g to
Si
mulati
on
an
d
Protot
ypin
g”. 2014: 15
–1
6.
[2]
AZ
Yonis, MF
L Abdul
lah
and
MF
Ghanim. “LT
E
-F
DD
and LT
E-T
DD for cellu
lar comm
u
n
icati
ons”. in
Proc. Progress
In Electromag
netics Res
earc
h
Symposi
u
m
. 201
2.
[3]
“Req
uirem
ents
for Internationa
l Mobi
le
T
e
leco
mmunic
a
tions (IMT
)
S
y
stems Oper
ating i
n
th
e
F
r
eque
nc
y
B
a
n
d
250
0 MHz to 269
0 MHz”. 20
12.
[4]
Y Liu,
X Z
h
o
n
g
, J W
ang, Y
Lan
an
d A Har
ada. “C
ontrol
Cha
nne
ls Perf
ormanc
e Eval
u
a
tion
w
i
th th
e
Coe
x
ist
ence
of T
D
-L
T
E
and LT
E-F
DD”.
Ve
hicul
a
r T
e
ch
no
logy C
onfer
en
ce (VT
C
F
a
ll),
201
3 IEE
E
78th
. 20
13: 1–
5.
[5]
J W
e
i, X Z
hon
g, L Liu a
nd
X
F
u
.
“Interference eval
uati
on
of control ch
an
nels u
n
d
e
r the
co-e
xiste
n
c
e
of LT
E-F
DD an
d T
D
-L
T
E
”.
Int.
Conf. Inf. Net
w
.
2014: 192–
196.
[6]
B Hu
ang,
H T
an, W
W
e
i, J F
a
ng
and
N
Z
hen
g. “Co
e
x
is
tenc
e stud
ies f
o
r L
T
E-F
DD
w
i
th T
D
-LT
E
in
t
h
e
ban
d 25
00-2
6
9
0
MHz”.
IET
Int. Conf. Commu
n. T
e
chnol. Ap
pl. (ICCT
A 201
1)
. 2011: 4
11–
416.
[7]
T
Chuang, G Che
n
, M T
s
ai
and C
Li
n. “All
eviati
ng
Interfe
r
ence thr
oug
h
Cog
n
itive
Rad
i
o for LT
E-
Advanc
ed N
e
tw
o
r
k”. 201
5; 5(3): 539–
54
7.
[8]
E Dahlm
an, S
Parkvall, J Sko
l
d an
d P Bemi
ng.
3G evolution:
HSPA and
LTE for
m
o
bile broadband
.
Academ
ic pres
s. 2010.
Evaluation Warning : The document was created with Spire.PDF for Python.