TELKOM
NIKA
, Vol. 13, No. 4, Dece
mb
er 201
5, pp. 1352
~1
360
ISSN: 1693-6
930,
accredited
A
by DIKTI, De
cree No: 58/DIK
T
I/Kep/2013
DOI
:
10.12928/TELKOMNIKA.v13i4.2738
1352
Re
cei
v
ed
Jul
y
27, 201
5; Revi
sed
No
ve
m
ber 2, 2015
; Accepte
d
Novem
b
e
r
20, 2015
Transmission of Real-time Video Signal with
Interference Density an
d Human Traffic
Rizal B
r
oer
Baha
w
e
r
es*
1
, Oki Teguh Kar
y
a
2
, Mudrik Ala
y
drus
3
1,3
Post-Graduate Program of Elec
trical E
ngi
neer
ing D
e
p
a
rtement, Mercu
Buan
a Univ
ersi
t
y
,
1
F
a
cult
y
of Sci
ence a
nd T
e
chnol
og
y, State Islamic
Un
iversi
t
y
S
y
arif Hid
a
y
at
ull
ah Jak
a
rta, Indones
ia,
2
KompasT
V
*Corres
p
o
ndi
n
g
author, em
ail
:
rizalbro
er@i
e
ee.org
1
, oki.teg
uh@k
o
mpas.tv
2
, mudrikala
y
d
r
us@
y
a
hoo.co
m
3
A
b
st
r
a
ct
T
he
use
of mo
bile ph
one as a
co
mmun
icati
on
too
l
rap
i
dly
incre
a
ses, as
w
e
ll as vari
ous
types of
functions ther
e
i
n. Amo
ng of
the many a
ppl
i
c
ations o
n
the
mob
ile
p
h
o
ne,
w
h
ich are use
d
skype an
droi
d-
base
d
for telec
onfere
n
cin
g
vi
a mo
bil
e
ph
on
e. T
h
ings
can
not be sep
a
rat
ed from
thes
e app
licati
ons is
the
nee
d of qu
alifi
ed inter
net n
e
tw
ork access for these ap
p
lic
ations ca
n be fe
lt up func
tio
n
. One of the n
e
tw
or
k
access to the Internet is w
i
dely use
d
today
'
s
society
is a w
i
-fi netw
o
rk.
Access to the Internet do
es not
alw
a
ys provi
d
e the best pe
rforma
nce, thi
s
is due to
many factors, one of w
h
ich i
s
the presenc
e of
interference. In this study,
w
e
tested tra
n
s
m
i
ssion
of vi
de
o
sign
als
in r
e
a
l
-time
usi
n
g
a
n
app
licati
o
n
sky
p
e
on the
mob
ile
pho
ne. Skyp
e
run o
n
w
i
-fi ne
tw
ork,
w
h
ic
h is
influ
enc
ed by
the pres
enc
e o
f
interferenc
e
or
obj
ects such a
s
human traffic
in the netw
o
rk.
W
i
reshark w
e
re used to o
b
tain d
a
ta reli
ab
i
lity w
i
-fi netw
o
rk,
know
n as QoS
.
W
h
ile ev
alvi
d
used to
obta
i
n
the data
Qo
E re
stri
cte
d
on
l
y
ge
t th
e
da
ta
PSN
R
(Pea
k Si
gna
l
T
o
Noise R
a
tio
)
and MSE (Me
an Squ
a
re Erro
r).
Ke
y
w
ords
: Hu
ma
n T
r
affic Interferenc
e, Qo
S, QoE, Video Streamin
g Qualit
y
Copy
right
©
2015 Un
ive
r
sita
s Ah
mad
Dah
l
an
. All rig
h
t
s r
ese
rved
.
1. Introduc
tion
Video call at
this time is so very pop
ular in the community, prece
ded by ei
ther hi
s
popul
ar sm
artphone
ba
se
d
on
And
r
oid,
i
O
S, Blackbe
r
ry, and
Wi
nd
ows Ph
one
.
T
h
is en
cou
r
ag
es
the emergen
ce of a variety of video call/
chat
appli
c
atio
ns runni
ng on
the mobile p
hone pl
atform
,
despite emerging ne
w ap
plicatio
ns, bu
t Sky
pe is an application
video calls
most pop
ular.
according to
predi
ction
s
th
at have bee
n
relea
s
e
d
previously by
GigaO
M Proj
ect [1], in 20
15
con
s
um
er vid
eo call
s will reach
142.9
million users, with a total
of video calls as much as
29.6
billion. Access to the Internet by
wi-fi so
very
widely
used
t
oday, such as office buildi
n
gs,
sho
ppin
g
mal
l
s, even insid
e
the hou
se wa
s most
pe
ople, espe
cia
lly in ur
ban a
r
ea
s, the use of
wi-fi for internet access is
a commonplace. Wi
-fi standard device used is the type IEEE 802.11
a, b, g and n, depe
nding o
n
the need
s of the
netwo
rk a
nd a wid
e
ran
ge of desi
r
ed.
Figure 1. The
exponential i
n
crea
se in th
e numbe
r of consume
r
vide
o call
s
and total vide
o call
sl / chat
s [1]
3.2
10.3
27.1
59.8
97.5
142.9
0
50
100
150
200
2010
2011
2012
2013
2014
2015
In
Millions
Mobile Video
Call/Chats
Consumer
Mobile
Video
Call/Chats
…
0.6
1.6
3.2
5.6
9.9
16.6
23.2
29.6
0
10
20
30
40
2008
2009
2010
2011
2012
2013
2014
2015
In
Billions
Total Video
Calls
Total
Video
Calls
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 136
0 – 1360
1353
Whe
n
we tal
k
ab
out the
real-time vid
e
o
tr
an
smi
ssi
o
n
or vid
eo
streami
ng the
n
we
are
talking abo
ut data
tra
n
smission ru
nnin
g
contin
uou
sl
y, whe
r
e
data i
s
simultan
eo
usly tra
n
smitted
and
re
ceived
contin
uou
sly. Wh
en th
ere
i
s
inte
rfer
en
ce
on th
e net
wo
rk
used, the
i
n
formatio
n
wi
ll
b
e
pa
r
t
ly los
t
a
n
d
no
da
ta
tr
an
s
m
iss
i
on
b
a
ck
.
The
r
efore, i
n
the
video
strea
m
is ne
ede
d
to
broa
dba
nd n
e
tworks with
delay, jitter and pac
ke
t loss (Q
oS) in accord
ance with the
pre
r
eq
uisite
s.
Another
cha
llenge fa
ced
whe
n
streami
ng video
run
n
ing on
a wi
rele
ss
network
requi
re
d QoS
good qu
ality in orde
r to avoid loss of
quality on the re
ceiving side (QoE
).
T
he
other p
r
obl
e
m
is, sendi
n
g
sign
als via
mobile devi
c
e
s
with wi
reless conn
ection techn
o
lo
g
y
802.11
or
Wi
-
Fi, limited
the
coverage
area
are
be
come
an
ob
stacl
e
s to th
e qu
ality si
gna
l
transmissio
n
of video
and
audi
o
strea
m
ing, so it t
a
ke
s
relia
bility in a
han
do
ver bet
wee
n
an
acce
ss p
o
int (AP) to anoth
e
r [2].
The problem
s that arise in video stre
a
m
ing ov
er
wi
rele
ss n
e
two
r
ks i
s
the pre
s
en
ce of
interferen
ce
or di
sturban
ces th
at app
e
a
r a
r
ou
nd th
e network eit
her i
n
the fo
rm of co-cha
n
nel
(wi
r
ele
s
s si
gn
al with the
sa
me freq
uen
cy
cha
nnel
),
or
sha
d
o
w
ing
of the presen
ce
of obje
c
ts th
at
appe
ar in
the
netwo
rk, incl
uding th
e p
r
e
s
en
ce
of hu
man traffic (h
uman traffic i
n
terferen
ce).
As
has
bee
n ob
serve
d
by [3], where the
movement
of
human t
r
affic or
network
amid ind
oors ca
n
decrea
s
e lin
k through
put as mu
ch a
s
20.4%. So when a video
strea
m
run
n
i
ng on a wi
rel
e
ss
network in an office environment that i
s
congested
with traffic of employees it
will greatly af
fect
the results of the signal d
e
livery. it can be ev
aluated from two si
de
s, by measu
r
i
ng QoS sen
d
e
r
side a
nd on t
he re
ceive
r
si
de
by measuring the QoE.
Quality of service
refe
rs to the n
e
twor
k
perfo
rmance, QoS
paramete
r
s con
s
i
s
t of
Throughput, Goodput, Delay, J
i
tter, Pac
k
e
ts
Loss
.
While Quality of Experien
c
e (QoE) is d
e
fin
e
d
in many dim
e
nsio
ns
but ge
nerally Q
o
E refers to
the q
uality on the
receiving
sid
e
or p
e
rcieve
d
Quality and
well defined a
s
the use
r
satisfactio
n
of
the se
rvice o
r
service
s
p
r
ovi
ded [4].
QoS and
QoE interrela
t
ed, so if QoS disturb
ed affect QoE. Some re
sea
r
che
r
s
cond
uct re
sea
r
ch rel
a
te
d to
the correlatio
n bet
ween
Q
o
S and
QoE,
in [5] where
he d
e
velop
e
d
a m
odel
co
rrel
a
tion
QoS
/
QoE for IPTV custom
er
satisfa
c
tion.
Model devel
oped by re
searche
r
s [5]
con
c
lu
ded th
at
knowing
the QoS
paramet
e
rs
it will get a
predi
ctio
n of QoE. So did the
opposite
by getting QoE
para
m
eters
can al
so
be
u
s
ed
to d
e
termine the
co
ndition
of Qo
S.
More than that, the QoS
para
m
eters can
be used as
a refe
ren
c
e
level of
s
e
cu
rit
y
in t
e
lecom
m
uni
cat
i
ons
so
und
s
like
VoIP. As in t
he
study [6]
carrie
d o
u
t th
e mea
s
u
r
em
ent of Q
o
S p
a
ram
e
ters
su
ch
as thro
ug
hput,
delay an
d p
a
c
ketloss u
s
e
d
as
a pa
ram
e
ter to d
e
termi
ne the
relia
bil
i
ty of the en
cryption meth
od
use
d
.
Studies relat
ed to the tran
smissio
n
of vi
deo si
gnal
s in real
-time h
a
s be
en don
e
by other
resea
r
chers
whi
c
h we hav
e summ
ari
z
e
d
in the following matrix.
Table 1. Matri
x
Related Studies
No Researcher
Topic
Research
Research Scena
rio
Strea
ming
Video
Wifi
Inter-
ference
S
k
y
pe
PC
Mobile
OS
1
Nurul Sarkar
et.a
l [3]
“The Effect of Pe
ople Movement
on Wi-Fi Link Thr
oughput in
Indoor Pr
opagati
on
Environments,”
No Yes
Yes
No
No
2
K. K. Eudon [7]
“Video Streamin
g over 802. 11
b
in the Presence
of Fading due t
o
Human T
r
affic an
d Bluetooth
Interfer
ence,”
Ye
s
Ye
s
Ye
s
No
No
3
E. Masala et.al [8]
"Real-time tra
n
smission of h.
264 video over 8
02. 11b-b
a
sed
w
i
reless ad hoc"
Ye
s
Ad-
hoc
Yes
No
No
4
Fernan
dez et.al [
9
]
“Video Confe
r
en
ces through the
Internet:
Ho
w
to
Survive in a
Hostile Environment”
Ye
s
No
Ye
s
Ye
s
No
5
RB.Bah
w
e
res;
Oki Teguh Ka
r
y
a
;
Mudrik Alaydrus
“Tra
nsmission of
Real-Time
Video Signal w
i
th
Interfer
ence
Densit
y
and H
u
m
an Traffic”
Ye
s
Ye
s
Ye
s
Ye
s
Ye
s
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Tran
sm
issi
on
of Real-tim
e Video Signal
with Inte
rfere
n
ce
Den
s
ity
… (Ri
z
al Bro
e
r Baha
we
re
s)
1354
Starting from
the above studies, resea
r
ch
ers
trie
d to ca
rry out
a study which is the
intersectio
n
o
f
the three st
udie
s
above.
Re
sea
r
ch
e
r
s
con
d
u
c
ted a f
i
eld experi
m
e
n
t in the form
o
f
strea
m
ing
video
usi
ng S
k
ype
appli
c
at
ion a
ndroi
d-b
a
se
d
p
eer-to
-
pee
r (p2
p
). Video stre
am
ing
run
s
on wi-fi netwo
rk
that
i
n
terferen
ce b
y
traffi
c a
nd t
he d
e
n
s
ity of the e
m
ploye
e
s i
n
the
office
room. Thi
s
st
udy was conducted to
determine how the reliability of
a network
with Wi-Fi on an
office buildin
g in transmitti
ng video in real-time u
s
in
g Skype on interferen
ce b
y
the presen
ce of
human d
e
n
s
ity and traffic in the middle
of the Wi-F
i n
e
twork. Skyp
e android is
use
d
to cond
uct
video co
nferences a
r
e pe
er-to
-
pe
er (p
2p) with
othe
r Skype u
s
e
r
s on the sa
me wi-fi net
work.
Relia
bility is a measure vi
deo streamin
g quality,
while Skype takes pla
c
e ami
d
the density
of
employee
s, i
n
othe
r word
s the b
and
wi
dth is e
r
ra
tic
due to the
in
terfere
n
ce of
the den
sity a
nd
traffic of em
ployee
s. Video streami
n
g
quality in
terms
of two p
a
rts, na
mely
QoS (Q
uality of
Service) as
a wi-fi
network reliability are used
and
QoE (Quality of
Experie
nce) as a
m
easure
of
perceived q
u
a
lity on the receiving
side.
2. Rese
arch
Metho
d
In a study
co
ndu
cted, the
re
sea
r
chers t
e
sted th
e tra
n
smi
ssi
on of
video si
gnal
s in real
-
time usin
g the Skype a
p
p
licatio
n ba
sed on a
ndroi
d ope
rating
system, runn
ing on a
Wi
-Fi
netwo
rk i
n
the buildin
g of Kompa
s
TV. Skype ap
p
lication on the a
c
tual fun
c
tion
is an ap
plica
t
ion
use
d
for vide
o confe
r
e
n
ci
n
g
, but for this
study, Sk
ype
is u
s
ed o
n
ly to tran
smit video sig
nal
s onl
y.
Thus the m
e
asu
r
em
ent limit rese
arch
to be carri
e
d
out later is just video streami
ng qual
ity.
Video streami
ng quality is
closely related
to the per
ce
ption of the e
nd user o
r
kn
own by the
Q
o
E
[4], which to
measure it is
by kn
owin
g
what is pe
rceived by the
en
d u
s
er.
Rel
a
ted to th
e results
to be
kno
w
n
by re
sea
r
ch
ers, th
en thi
s
can
not b
e
sep
a
rate
d al
so by m
e
a
s
u
r
ing th
e QoS
or
Quality of Service of the wi-fi netwo
rk it self.
Both of these (Q
oE and QoS)
will be co
ntrolle
d b
y
the control
variabl
es traffic a
nd d
e
n
s
ity of empl
oyee
s in th
e b
u
ild
ing Komp
as
TV ba
sed
office
hours at the room natu
r
ally
.
Measu
r
em
en
t tools u
s
ed
b
y
re
sea
r
che
r
s to d
e
termi
n
e
the
quality
streami
ng vid
e
o
u
s
ing
a Lin
u
x-ba
se
d softwa
r
e th
at is EvalVid [
10], t
he
para
m
eters u
s
e
d
are
PSNR a
n
d
MSE. Th
e
most
importa
nt fact
or of PSNR i
s
MSE (Me
a
n
Squa
re
Error)
whi
c
h i
s
the maximum
possibl
e valu
e of
the luminan
ce (28
-
1 = 2
5
5
for 8 bit) [10].
∑∑
.
(
1
)
20
255
√
(
2
)
Whe
r
e
fij is t
he o
r
igin
al
si
gnal
of pixel
(i, j),
whil
e th
e Fij
a
deg
ra
ded
sig
nal, a
nd M
N
i
s
the size of the video. The
large
r
the val
ue of
MSE, the sm
aller P
S
NR valu
e, whi
c
h me
an
s the
quality of the video is ugly.
While i
n
mea
s
uri
ng Q
o
S, resea
r
chers
use
Wiresha
r
k software a
s
a tool to
d
e
termin
e
Thro
ugh
put, Packet Loss,
Delay and
Jitter. From
t
he mea
s
u
r
e
m
ent re
sults,
the rese
archers
wante
d
to kn
ow ho
w the traffic and the den
sity of
the
employee
s in the building
KompasTV a
b
le
to interfere
n
ce QoS and Q
o
E video sign
al transmissio
n via Skype androi
d.
This
study wa
s co
ndu
cted t
o
answe
r the followin
g
que
stion
s
:
1)
Ho
w much influen
ce traffic
and den
sity of em
ployees i
n
the building
KompasTV t
o
the value
QoS an
d Qo
E video si
gna
l tran
smissio
n
process
with Skype A
ndroid, whil
e traf
fic and
hi
g
h
den
sity employees?
2)
Ho
w much influen
ce traffic
and den
sity of em
ployees i
n
the building
KompasTV t
o
the value
QoS a
nd
Qo
E video
sig
n
a
l tra
n
smi
s
si
on p
r
o
c
e
s
s
with S
k
ype
Androi
d, whil
e traffic an
d
den
sity of the low employe
e
?
This
study is
given limitations a
s
follows:
1)
The stu
d
y was cond
ucte
d
in the seco
nd floor
of th
e buildin
g KompasTV use
a sepa
rate
netwo
rk f
r
om
the existing
netwo
rk i
n
th
e buildin
g,
wi
th take pl
ace
with traffic a
n
d den
sity of
employee
s (Divisio
n HR / LEGAL, GA
Divisio
n
, Division
FINANCE, and Divisio
n
MARCOMM
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 136
0 – 1360
1355
2)
The process
of transmittin
g
real
-time video si
gnal i
s
done in
a pee
r-to
-
pee
r from
the sen
d
in
g
device (and
ro
id 1) to a re
ce
iving device (android 2
)
.
3)
Measuri
ng to
ol use
d
for m
easure the Q
o
S is Wiresh
ark
with pa
ra
meter only Th
roug
hput.
4)
Measuri
ng to
ol use
d
to me
asu
r
e the Q
o
E is
EvalVid, by taking o
n
l
y
paramete
r
PSNR (Pe
a
k
Signal to Noi
s
e Ratio).
5)
Another pa
ra
meter to
me
asu
r
e
pe
rci
e
ved qu
ality, rese
arche
r
s u
s
ed
on
e
way
video
delay
method [11]. This meth
od
is used be
ca
use S
k
y
pe e
n
crypt
s
the i
n
formatio
n re
lated to the
perfo
rman
ce
of the system
. through this method
re
se
arche
r
s
can
collect data rel
a
ted to the
video delay the teleconferenc
i
ng sy
ste
m
in Skype.
Figure 2. One
Way Video Delay Method [11]
6)
To get the
da
ta, the pro
c
e
s
s of real
-time
vi
deo tra
n
sm
issi
on la
sts fo
r aroun
d 15
seco
nd
s, at
any intervals
for 30 minute
s
. The data b
egan
to be ta
ken at 09.3
0
am until 21.0
0
pm.
Figure 3. De
sign study con
ducte
d to answer the
re
sea
r
ch q
u
e
s
tion
3. Results a
nd Analy
s
is
3.1. Data T
r
a
ffic and
Den
s
it
y
of Emp
l
o
y
ees
Data traffic a
nd de
nsity of
employee
s,
obt
aine
d thro
ugh mo
nitori
ng of existin
g
CCTV
came
ra
s
on t
he
se
cond
flo
o
r of th
e b
u
ilding Ko
m
p
a
s
TV. From th
e
data
colle
cte
d
, it wa
s n
o
ted
that the employees' p
e
a
k
den
sity at interval ti
me 14
:30 pm to 15:30 pm, with a rang
e between
51 to 53 peo
p
l
e, as sh
own in Figure 4 an
d 5.
QoS measu
remen
t
Android Dev
i
ce
- 1
Transmit
real-
t
im
e
video
si
gnal
Android
dev
ice -
2
Receive
rea
l
-t
ime
vi
deo s
i
gnal
Tr
affi
c
and dens
i
t
y
interfe
r
ence
of
Kom
p
asTV
e
m
p
lo
y
ee
Ev
alVid Soft
w
a
re
QoE M
easure
men
t
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Tran
sm
issi
on
of Real-tim
e Video Signal
with Inte
rfere
n
ce
Den
s
ity
… (Ri
z
al Bro
e
r Baha
we
re
s)
1356
Figure 4. The
Amount Of Employee Ba
sed On Interva
l
Time
3.2. Data o
f
Net
w
o
r
k T
h
r
oughpu
t
To obtai
n th
e data
of ne
twork th
rou
g
hput in
th
e f
o
rm
of interf
eren
ce
by traffic an
d
den
sity of e
m
ployee
s, re
sea
r
che
r
s
used the
Wi
re
shark
softwa
r
e while the
video d
e
livery
via
skyp
e
pro
c
e
s
s und
erway, with re
sults of
data colle
ctio
n as sho
w
n in
Figure 6
Figure 5. Ca
mera Mo
nitori
ng at 15.30 p
Figure 6. Net
w
ork Th
rou
g
h
put Based o
n
Interval Time
3.2. Data Vid
e
o Dela
y
To facilitate i
n
tera
ction b
e
twee
n user tel
e
co
nfere
n
ci
n
g
, it’s need
s
a stan
dard d
e
lay time
for video (vid
eo delay) not
too long. The use
r
experi
ence will be decrea
s
e
d
if
the video del
ay
excee
d
s
350
ms [12]. In this stu
d
y we u
s
ed
a video d
e
lay mea
s
ure
m
ent metho
d
s
u
s
ed by [11
],
in this
way
we re
co
rd it i
n
a video file
fo
r 15
second
s,
then o
n
eve
r
y frame
we
calcul
ate the ti
me
differen
c
e bet
wee
n
the video se
nt to the incomin
g
video.
28
42
37
45
33
34
31
35
28
38
51
48
53
41
38
21
34
18
17
14
15
13
4
1
0
20
40
60
09.30
10.00
10.30
11.00
11.30
12.00
12.30
13.00
13.30
14.00
14.30
15.00
15.30
16.00
16.30
17.00
17.30
18.00
18.30
19.00
19.30
20.00
20.30
21.00
Inter
v
a
l
Time
Amoun
t
of
Employ
ee
Amount
of
employee
156
201
101
134
138
164
96
30
80
135
139
81
4
56
136
51
35
52
149
164
158
142
69
75
0
10
20
30
40
50
60
0
50
100
150
200
250
09.30
10.00
10.30
11.00
11.30
12.00
12.30
13.00
13.30
14.00
14.30
15.00
15.30
16.00
16.30
17.00
17.30
18.00
18.30
19.00
19.30
20.00
20.30
21.00
Amount
of
Emp
l
oyee
Throughpu
t
in
kpbs
Inter
v
a
l
Time
Thr
o
ughput
vs
Amoun
t
of
Employ
ee
A
m
o
unt
of
Employee
Thro
ughput
(kbps)
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 136
0 – 1360
1357
Figure 7. Screen capture video del
ay on
data retrieval
14.30 pm
On Fig
u
re
7, we
can
se
e
the stop
wat
c
h o
n
the lef
t
side of the
video imag
e
sent by
android devi
c
e 1, and stop
watch on
the right sid
e
is receive
d
video image
s on android devi
c
es
2. It ca
n b
e
see
n
that
the
differe
nce in
time b
e
twe
e
n
the
left ima
ge a
nd
a
rig
h
t imag
e, whi
c
h
defined a
s
a video delay. In the picture above cle
a
rly
visible whe
n
the picture stopwat
ch se
n
t
alrea
d
y on 7
.
71 se
con
d
s,
at the recei
v
ing androi
d
device
s
still
in 0.00 se
cond
s. The d
a
ta
colle
cting vid
eo delay can
be se
en in Fi
gure 8
Figure 8. Video Del
a
y based on Work T
i
me
3.3. Data Pe
ak Signal to Noise
Ratio (PSNR)
Peak Sig
nal
to Noi
s
e
Rat
i
o (PSNR) is us
ed a
s
a
measure of
satisfa
c
tion
the u
s
e
r
experie
nce.
The PS
NR value it’
s
h
a
d
from t
he
re
sult of a
com
pari
s
on
of th
e received vi
deo
image
s with v
i
deo imag
es
sent via skyp
e. In th
is stud
y, the video image receive
d
throug
h on
the
android
2 re
corded
u
s
ing
Camta
s
ia
Reco
rde
r
u
s
in
g a frame
ra
te of 15 fp
s.
While
the vi
deo
image
that i
s
sent i
n
via
an
droid
1
recorded
usi
ng S
c
reen
Re
co
rde
r
app
al
so
with
a frame
rate
of
15 fps. The e
x
ample co
mp
arison of the android 1
(se
nder) an
d the androi
d 2 (re
c
eiver) can b
e
see
n
on figure 9, and All the PSNR dat
a
colle
cted
can
be se
en in Fi
gure 1
0
.
419
403
303
610
1208
1707
1030
3057
6267
353
314
2766
325
2335
890
727
0
10
20
30
40
50
60
0
1000
2000
3000
4000
5000
6000
7000
09.30
10.00
10.30
11.00
11.30
12.00
12.30
13.00
13.30
14.00
14.30
15.00
15.30
16.00
16.30
17.00
17.30
18.00
18.30
19.00
19.30
20.00
20.30
21.00
Amount
of
Emp
l
oyee
Video
Delay
in
millisecond
Inter
v
a
l
Time
Video
Dela
y
vs
Amoun
t
of
Employ
ee
A
m
o
unt
of
Employee
Vid
e
o
Delay
(mS)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Tran
sm
issi
on
of Real-tim
e Video Signal
with Inte
rfere
n
ce
Den
s
ity
… (Ri
z
al Bro
e
r Baha
we
re
s)
1358
(a)
(b)
Figure 9. Co
mpari
s
o
n
video image
sen
t
fr
om Androi
d 1 (figure 9a
) and video i
m
age
received on A
ndroi
d 2 (figu
r
e 9b
), on dat
a retriveal at
15.00 pm.
3.4. Data
An
aly
s
is
In the p
r
evio
us
se
ction
we ha
d
see
n
t
he d
a
ta of th
ree
area, n
e
twork th
ro
ugh
put a
s
a
para
m
eter fo
r quality of service (QoS
), while
video delay time and PSNR a
s
a paramete
r
for
quality of experien
c
e (QoE
). In every paramete
r
(Q
oS and QoE) we comp
are by
the condition
of
amount
of e
m
ployee. In
my assumpti
on ea
rlie
r
be
fore thi
s
stu
d
y bega
n, if the amo
unt
of
employee in
crea
sed
can b
e
linear o
n
the traffic
and
den
sity of employee, and it can give so
me
interferen
ce to the QoS param
eter. Th
en if the
QoS paramete
r
cha
nge
d so i
t
can cha
nge
the
QoE pa
ramet
e
r. Based on
the data that
we h
a
ve,
in figure
4 we ca
n se
e the pe
a
k
value a
m
ou
nt
of employe
e
at interval tim
e
14.3
0
to
15
.30 pm.
An
d i
f
we
ma
ke
co
mparation to
the next d
a
ta
in
figure
6,
we
can
seen
tha
t
at the
pea
k value
of em
ployee, the
n
e
twork thro
u
ghput
ch
ang
ed t
o
the lo
wer po
sition at rang
e
4 to
81
kb
ps. If comp
are
again
with
an
other
data i
n
figure
8, at th
e
same valu
e (pea
k) of the employee, th
e Delay Ti
me
of video at the high po
siti
on, that it me
ant
the video del
ay time is very lag.
Figure 10. PSNR b
a
sed on
Wo
rk Tim
e
Let’s we jum
p
to another
data in figure
10,
we ca
n see
n
the value of PSNR
comp
are
with the (p
ea
k) a
m
ount of
employee. In this dat
a, t
he PSNR
sh
ow to u
s
that
in this condi
tion
PSNR go
to the lowe
st value (1
9,3 dB) from
the other. Temp
o
r
ary we
can
con
c
lu
de that in
interval time
at 14.30
to 1
5
.00 pm,
all
para
m
eters
b
o
th QoS
and
QoE affe
cte
d
by traffic a
nd
den
sity of employee.
24.29
21.38
25.96
26.41
26.78
19.79
24.01
19.3
24.7
21.35
26.08
19.79
24.09
21.87
25.72
24.03
24.22
23.41
0
10
20
30
40
50
60
0
5
10
15
20
25
30
09.30
10.00
10.30
11.00
11.30
12.00
12.30
13.00
13.30
14.00
14.30
15.00
15.30
16.00
16.30
17.00
17.30
18.00
18.30
19.00
19.30
20.00
20.30
21.00
Amount
of
Emp
l
oyee
PSNR
in
dB
Inter
v
a
l
Time
PSNR
vs
Amoun
t
of
Employ
ee
A
m
o
unt
of
Employee
PSNR
(dB
)
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 136
0 – 1360
1359
On the co
nditions in an
oth
e
r time interv
al, t
he data in Figure 6, at intervals time
at 10:00
am, numbe
r
of the employ
ees
sho
w
e
d
42 peo
ple, wi
resha
r
k m
e
a
s
ure the n
e
two
r
k th
roug
hput
at
positio
n 14
2
kbp
s
. If we
comp
are the
con
d
ition
s
i
n
whi
c
h th
e
numbe
r of e
m
ployee
s in
top
positio
ns i
n
53 empl
oyee
s, wiresha
r
k
sho
w
s extre
m
e value
s
in
positio
n 4
kbps.
Why di
d it
happ
en?, Ou
r analy
s
is
co
nclu
ded in th
e cu
rrent co
n
d
itions of the
numbe
r of e
m
ployee
s in t
o
p
positio
ns co
u
p
led with
the movement
of
the
tra
ffic co
ndition
s
of
th
e
empl
oyee
s are quite
b
u
sy,
this can h
a
ve
a sig
n
ifica
n
t
influen
ce o
n
the wi
fi n
e
two
r
k
co
ndition
s,
com
pared to
only state th
e
numbe
r of e
m
ployee
s cro
w
de
d yet traffic movement
is slig
ht. Thus we con
c
lu
de that the time
intervals at 1
0
.00 am, the
movement
of traffi
c i
s
n
o
t as
bu
sy e
m
ployee
s at
15.30 p
m
. T
h
is
analysi
s
i
s
i
n
line
with
wh
at ha
s b
een
con
c
lu
ded
in
the stu
d
y [3], that the
mov
e
ment
of pe
o
p
le
throug
h the n
e
twork eithe
r
moving forwa
r
ds o
r
rando
m moving ca
n redu
ce th
ro
uhput lin
k.
Figure 11. PSNR
comp
are
d
with video d
e
lay time based on Work T
i
me
In figure 11,
we can
see t
he influen
ce
of video dela
y
time value again
s
t the value of
PSNR a
s
a q
uality use
r
ex
perie
nce, re
searche
r
s gav
e a sig
n
in th
e form of a d
o
tted line at three
intersectio
n
.
At the first int
e
rsectio
n
of v
i
deo d
e
lay ti
me po
sition
s
in 120
8 m
s
a
nd PSNR val
ue
indicates th
e
positio
n of 1
9
.79 dB, the
n
we
loo
k
at
the se
co
nd li
ne of inte
rse
c
tion vide
o d
e
lay
positio
ns do
wn at 1030
ms and PSNR value ro
se
to
24.01 dB, at the intersection of the line
s
third, the
po
si
tion of the vid
eo del
ay is in
the
extrem
e
value at 6
267
ms
and PS
NR value
fell t
o
19.3 dB
po
sit
i
on.
From
the
pictu
r
e
abov
e is very
obv
ious that the
video d
e
lay ti
me affe
cts th
e
quality of u
s
e
r
expe
rien
ce
i
s
rep
r
esented
by the
valu
e
of PSNR. if th
e value
rises
then the vid
e
o
delay PSNR
value fall
s, th
e op
po
site
co
ndition if
th
e
value of vid
e
o
del
ay do
wn
the valu
e
of
the
PSNR ri
se
s, whi
c
h mea
n
s
the video ima
ges a
r
e recei
v
ed in good q
uality.
4. Result a
n
d Discus
s
io
n
With the
wi
d
e
sp
rea
d
u
s
e
of video
stre
aming,
e
s
p
e
cially on vide
o
co
nfere
n
ci
ng
usi
ng
a
wirel
e
ss network,
th
ro
ugh
t
h
is re
sea
r
ch we can
kno
w
that traffic a
n
d
den
sity of e
m
ployee
s in t
he
room
able to
interferen
ce t
o
quality of th
e wirele
ss
ne
twork (wifi)
which i
s
u
s
e
d
, also
red
u
ce t
he
quality of u
s
e
r
expe
rien
ce
sho
w
n
on
Qo
E value. In
th
is stu
d
y note
that the de
cli
ne in th
e valu
e
of PSNR is
affected
by t
he time
del
a
y
video. the
longe
r the
tim
e
video
del
a
y
, the wo
rse
the
value of PSNR.
In future
stu
d
ies we
can
add
a
scen
ario
if th
e se
nder an
d recipient an
droi
d device
moves in
doo
rs to clo
s
e
r
o
r
move away
to acce
s poi
nt. so that we can
kn
ow t
he interfe
r
en
ce
para
m
eters that most influ
ence on the va
lue of QoS a
nd QoE video
streami
ng.
Referen
ces
[1]
S Jana, A Pande, A Cha
n
, and P Moh
a
p
a
tra.
“Mobil
e
vide
o chat: Issues an
d cha
l
l
eng
es”.
IEEE
Co
mmun. Ma
g
.
2013; 51(
6) 1
44–
15
1.
19.79
24.01
19.3
1208
1030
6267
0
1000
2000
3000
4000
5000
6000
7000
0
5
10
15
20
25
30
09.
30
10.
00
10.
30
11.
00
11.
30
12.
00
12.
30
13.
00
13.
30
14.
00
14.
30
15.
00
15.
30
16.
00
16.
30
17.
00
17.
30
18.
00
18.
30
19.
00
19.
30
20.
00
20.
30
21.
00
Video
Delay
in
Millisecond
PSNR
in
dB
Time
Inter
v
a
l
Video
Dela
y
vs
PSNR
PSNR
(dB
)
Vid
e
o
Delay
(mS)
1
2
3
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Tran
sm
issi
on
of Real-tim
e Video Signal
with Inte
rfere
n
ce
Den
s
ity
… (Ri
z
al Bro
e
r Baha
we
re
s)
1360
[2]
I Praset
y
o
, M Anif and AS N
ugro
ho. “Han
d
o
ver An
a
l
ysis
of Data an
d VoIP Services i
n
802.1
1b/g/
n
Wireless LAN”.
T
E
LKOMNIKA Indones. J. Electr. Eng.
2014
; 12(11): 78
32–
784
4.
[3]
Sarkar, Nurul I and O Mussa. “
T
he Effect of
Peo
p
le
Move
me
nt o
n
W
i
-F
i
Link T
h
rou
g
h
p
u
t in I
ndo
o
r
Propa
gati
on E
n
viro
nments
”. Procee
din
g
20
13 IEEE T
E
NC
ON. 2013: 59
8
–60
2.
[4]
JMG Stensen. “Evaluati
ng
QoS and QoE
Dimens
ions
i
n
Ada
p
tive Vi
deo Stre
amin
g
”
. Nor
w
e
g
ia
n
Univers
i
t
y
of Scienc
e an
d T
e
chno
log
y
. 2
012.
[5]
HJ Kim, KS C
ho, HS K
i
m a
nd SG C
hoi. “
A
Study
on a QoS / QoE
Co
rrelation Mo
del for QoE
Evalu
a
tion
on
IPT
V
Service”. in
Advanc
e
d
Co
mmu
n
icat
ion T
e
ch
no
log
y
(ICACT
), 2010 T
he 1
2
t
h
Internatio
na
l C
onfere
n
ce
. 20
1
0
: 1377, 1
382.
[6]
A W
ahab, RB
Baha
w
e
res,
M Ala
y
dr
us a
nd R S
a
r
no. “
P
erformanc
e a
nal
ysis
of VoI
P
clie
nt
w
i
t
h
integr
ated e
n
cr
ypti
on mo
dul
e”
.
2013 1st Int. Conf. Co
mmu
n
.
Signal Proc
es
s. their Appl.
2
013: 1–
6.
[7]
KK Eudo
n an
d
BR Petersen.
“
Vide
o strea
m
i
ng ov
er 80
2.11
b in t
he
prese
n
c
e of fadin
g
d
u
e
to hu
ma
n
traffic and
blu
e
t
ooth int
e
rfere
n
ce
”. In Proc
e
edi
ngs
of the
7th Ann
u
a
l
Co
mmunicati
on
N
e
t
w
o
r
ks a
n
d
Services R
e
se
arch Co
nferen
ce, CNSR 20
0
9
. 2009: 3
3–4
0
.
[8]
E Masa
la, CF
Chi
a
sser
i
ni,
M Meo
an
d J
C
De
Martin.
“REAL-T
IME
TRANSMISSION OF H. 264
VIDEO OVER
802. 11B-B
AS
ED WIRELESS AD HOC”.
DSP In-Vehicl
e Mob. Syst.
2005: 193–
20
7.
[9]
C F
e
rná
n
d
e
z,
J Sald
an
a, J F
e
rná
ndez-
N
av
ajas
, L
Se
que
i
r
a, an
d L
Cas
ades
us. “Vid
eo
confer
ence
s
throug
h the int
e
rnet: ho
w
to s
u
rvive i
n
a host
ile e
n
viro
nment
”.
Scientific W
o
rld Jour
nal.
20
14; 201
4.
[10]
VR Jo
nna
la
ga
dda. “Ev
a
lu
ati
on of
Vid
e
o
Qualit
y of
E
x
p
e
rie
n
ce usin
g EvalVi
d”.
BT
H-BELEKINGE
T
EKNISKA HOGSKOLA. 2012.
[11]
Y Xu, C Yu, J
Li, and Y Liu.
“Vide
o
telep
h
o
n
y
for en
d-co
n
s
umers: Meas
ureme
n
t stud
y
of Google+
,
iChat, an
d Sk
ype”.
IEEE/ACM Trans. Netw.
2
014; 22(
3): 826
–83
9.
[12]
J Jansen, P Cesar, DCA Bu
lt
erman, T
Stevens, I Kegel a
nd J
Issing. “Enab
lin
g Comp
o
s
ition-B
a
se
d
Vide
o-Co
nfere
n
cin
g
for the Home”.
Multim
edia, IEEE Trans.
2011; 13(
5): 869
–8
81.
Evaluation Warning : The document was created with Spire.PDF for Python.