T
E
L
KO
M
N
I
KA
T
e
lec
om
m
u
n
icat
ion
,
Com
p
u
t
i
n
g,
E
lec
t
r
on
ics
an
d
Cont
r
ol
Vol.
18
,
No.
3
,
J
une
2020
,
pp.
1150
~
115
7
I
S
S
N:
1693
-
6930
,
a
c
c
r
e
dit
e
d
F
ir
s
t
G
r
a
de
by
Ke
me
nr
is
tekdik
ti
,
De
c
r
e
e
No:
21/E
/KP
T
/2018
DO
I
:
10.
12928/
T
E
L
KO
M
NI
KA
.
v18i3.
14055
1150
Jou
r
n
al
h
omepage
:
ht
tp:
//
jour
nal.
uad
.
ac
.
id/
index
.
php/T
E
L
K
OM
N
I
K
A
T
h
e
i
m
p
a
c
t
of
n
oi
se
o
n
d
e
t
e
c
t
i
n
g
t
h
e
a
r
r
iv
al
an
gl
e
u
si
n
g
t
h
e
r
o
ot
-
WS
F algo
r
ith
m
B
t
is
s
am
B
ou
s
t
an
i,
Abd
e
n
n
ac
e
u
r
B
agh
d
ad
,
Aic
h
a
S
ah
e
l,
Abd
e
lh
ak
i
m
B
all
ou
k
,
Abd
e
lm
aj
id
B
ad
r
i
L
ab
o
rat
o
ry
E
l
ect
r
o
n
i
cs
,
E
n
erg
y
,
A
u
t
o
m
at
i
c
an
d
I
n
fo
rma
t
i
o
n
Pr
o
ces
s
i
n
g
,
E
l
ect
r
i
cal
E
n
g
i
n
eeri
n
g
D
e
p
ar
t
men
t
,
Facu
l
t
y
o
f
Sc
i
en
ce
an
d
T
ec
h
n
o
l
o
g
y
Mo
h
amme
d
i
a,
H
as
s
an
II
U
n
i
v
er
s
i
t
y
Cas
a
b
l
a
n
ca,
Mo
ro
cc
o
Ar
t
icle
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
ived
S
e
p
6
,
2019
R
e
vis
e
d
J
a
n
27
,
2020
Ac
c
e
pted
F
e
b
17
,
2020
T
h
i
s
ar
t
i
c
l
e
d
i
s
c
u
s
s
es
t
h
ree
s
t
an
d
ard
s
o
f
W
i
-
F
i
:
t
ra
d
i
t
i
o
n
a
l
,
cu
rren
t
a
n
d
n
ex
t
-
g
e
n
erat
i
o
n
W
i
-
Fi
.
T
h
es
e
s
t
a
n
d
ar
d
s
h
av
e
b
een
t
es
t
ed
fo
r
t
h
ei
r
a
b
i
l
i
t
y
t
o
d
et
ec
t
t
h
e
arri
v
al
an
g
l
e
o
f
a
n
o
i
s
y
s
y
s
t
em.
In
t
h
i
s
s
t
u
d
y
,
w
e
ch
o
s
e
t
o
w
o
rk
w
i
t
h
an
i
n
t
el
l
i
g
en
t
s
y
s
t
em
w
h
o
s
e
n
o
i
s
e
b
eco
me
s
mo
re
an
d
m
o
re
i
mp
o
r
t
an
t
t
o
d
e
t
ect
t
h
e
d
es
i
red
an
g
l
e
o
f
arr
i
v
a
l
.
H
o
w
e
v
er,
t
h
e
u
s
e
o
f
t
h
e
w
e
i
g
h
t
ed
s
u
b
s
p
ace
f
i
t
t
i
n
g
(W
SF)
al
g
o
r
i
t
h
m
w
a
s
ab
l
e
t
o
d
et
ec
t
al
l
a
n
g
l
es
e
v
en
f
o
r
t
h
e
5
t
h
g
en
er
at
i
on
Wi
-
F
i
w
i
t
h
o
u
t
a
n
y
p
ro
b
l
em,
a
n
d
t
h
eref
o
re
p
r
o
v
e
d
i
t
s
ro
b
u
s
t
n
es
s
a
g
ai
n
s
t
n
o
i
s
e.
K
e
y
w
o
r
d
s
:
DO
A
e
s
ti
mation
R
oot
-
W
S
F
a
lgor
it
hm
S
mar
t
a
ntenna
s
ys
tem
Wi
-
F
i
WI
-
Gig
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
ce
n
s
e
.
C
or
r
e
s
pon
din
g
A
u
th
or
:
B
ti
s
s
a
m
B
ous
t
a
ni
,
E
lec
tr
ica
l
E
ng
inee
r
ing
De
pa
r
tm
e
nt
,
F
a
c
ult
y
of
S
c
ienc
e
a
nd
T
e
c
hnology
M
oha
mm
e
dia,
Ha
s
s
a
n
I
I
Unive
r
s
it
y
C
a
s
a
blanc
a
,
B.
P
.
146
M
oha
mm
e
dia
20650
M
or
oc
c
o
.
E
mail:
bti
s
s
a
m.
bous
tani@gmail.
c
om
1.
I
NT
RODU
C
T
I
ON
M
os
t
r
e
c
e
nt
s
tudi
e
s
ha
ve
a
n
int
e
r
e
s
t
in
us
ing
the
5T
H
ge
ne
r
a
ti
o
n
in
dif
f
e
r
e
nt
f
ields
while
e
ns
ur
ing
the
c
ompatibi
li
ty
of
the
s
tanda
r
ds
us
e
d.
T
his
ne
w
ge
ne
r
a
ti
on
br
ings
a
s
igni
f
ica
nt
e
volut
ion
in
ter
ms
of
higher
da
ta
r
a
te,
r
e
duc
e
d
late
nc
y
ne
two
r
k
a
c
c
e
s
s
e
s
,
a
nd
m
or
e
e
ne
r
gy
-
e
f
f
icie
nc
y
[1
,
2]
.
W
ir
e
les
s
c
omm
unica
ti
on
r
a
dios
ope
r
a
ti
ng
a
t
f
r
e
que
nc
ies
of
a
ppr
oxim
a
tely
60
GH
z
of
f
e
r
c
ons
ider
a
ble
potential
f
or
the
s
uppor
t
of
t
he
s
e
5
G
c
omm
unica
ti
on
ne
twor
ks
[
3]
.
F
igur
e
1
gives
a
n
ov
e
r
view
of
the
global
s
pe
c
tr
um
o
f
5G
[
4]
.
T
he
ove
r
a
ll
s
pe
c
tr
um
o
f
5G
is
d
ivi
de
int
o
th
r
e
e
s
pe
c
tr
um
ba
nds
;
e
a
c
h
one
of
them
ha
s
unique
pr
ope
r
t
ies
,
they
a
r
e
a
s
f
oll
ows
;
l
ow
-
ba
nd
s
pe
c
tr
um
r
e
pr
e
s
e
nts
f
r
e
que
nc
ies
unde
r
1GH
z
,
it
is
a
c
tually
us
e
d
f
or
2G,
3G
a
nd
4G
s
e
r
vice
s
f
or
vo
ice
,
M
B
B
s
e
r
vice
s
a
nd
the
int
e
r
ne
t
o
f
thi
ngs
(
I
o
T
)
[
5
]
.
I
n
ter
media
te
ba
nd
s
pe
c
tr
um
c
or
r
e
s
ponds
to
f
r
e
que
nc
ies
be
twe
e
n
1
GH
z
a
nd
6
GH
z
,
a
ls
o
us
e
d
f
o
r
2G,
3G
a
nd
4G
s
e
r
vice
s
.
T
he
two
W
i
-
F
i
f
r
e
que
nc
ies
2.
4
GH
z
a
nd
5GH
z
that
be
long
to
thi
s
ba
nd
will
be
tr
e
a
ted
in
thi
s
a
r
t
icle
[
5]
.
High
-
ba
nd
s
pe
c
tr
um
s
ur
e
ly
of
f
e
r
s
the
e
xpe
c
ted
va
ult
in
s
pe
e
d,
c
a
pa
c
it
y,
qua
li
ty
a
nd
low
da
ta
late
nc
y
a
s
s
ur
e
d
by
5G,
thi
s
s
pe
c
tr
a
l
ba
nd
a
ll
ows
the
us
e
of
f
r
e
que
n
c
ies
f
r
om
24
GH
z
to
50
GH
z
,
with
a
djac
e
nt
ba
ndwidths
of
mor
e
than
100
M
Hz
pe
r
ne
twor
k
[
5]
.
C
ur
r
e
nt
wir
e
les
s
indoor
a
ppli
c
a
ti
ons
typi
c
a
ll
y
us
e
W
i
-
F
i
s
uit
a
ble
de
vice
s
to
s
uppor
t
the
c
onne
c
ti
vit
y
of
the
wir
e
les
s
ne
twor
k.
T
he
s
e
de
vice
s
c
onc
e
r
n
the
I
E
E
E
802.
11
s
tanda
r
ds
that
de
ploy
2.
4
GH
z
a
nd
5
GH
z
r
a
dio
ba
nds
.
How
e
ve
r
,
the
ne
xt
ge
ne
r
a
ti
on
of
w
ir
e
les
s
t
e
c
hnologi
e
s
is
f
a
c
ing
a
s
pe
c
tr
um
s
c
a
r
c
it
y
whe
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
T
he
impac
t
of
nois
e
on
de
tec
ti
ng
the
ar
r
ival
angle
us
ing
…
(
B
ti
s
s
am
B
ous
tani
)
1151
the
f
r
e
que
nc
y
ba
nd
is
be
low
10
GH
z
[
6]
.
I
n
a
c
c
or
d
a
n
c
e
with
the
r
e
quir
e
ments
of
5G,
it
is
mor
e
a
ppr
o
pr
iate
to
us
e
the
f
utur
e
I
E
E
E
802.
11
s
tanda
r
d
c
a
ll
e
d
W
i
-
Gig,
whic
h
ope
r
a
tes
a
t
the
f
r
e
que
nc
y
r
a
nge
of
60
GH
z
[
3]
.
I
E
E
E
802.
11
wir
e
les
s
loca
l
a
r
e
a
ne
twor
ks
known
a
s
W
i
-
F
i
ne
twor
ks
ha
ve
ga
ined
global
popular
it
y
dur
ing
the
las
t
de
c
a
de
due
to
their
low
c
os
t
a
nd
e
a
s
y
de
ploym
e
nt
[
7]
.
How
e
ve
r
,
be
c
a
us
e
of
the
b
a
ndwidth
li
mi
tation
in
tr
a
dit
ional
W
i
-
F
i
s
ys
tems
[
8]
,
the
W
i
-
F
i
indoor
pos
it
ioni
ng
s
ys
tem
c
a
n
ha
r
dly
a
c
hieve
loca
li
z
a
ti
on
a
c
c
ur
a
c
y
of
the
us
e
r
s
unde
r
ha
r
s
h
c
on
dit
ions
s
uc
h
a
s
the
non
-
li
ne
-
of
-
s
ight
(
NL
OS)
[
8]
,
higher
f
r
e
que
nc
ies
,
a
nd
nois
y
s
ys
tem,
whic
h
a
r
e
c
omm
on
f
o
r
the
indoor
e
nvir
onment.
T
he
a
na
lys
is
of
the
e
s
ti
mation
o
f
the
dir
e
c
ti
on
of
a
r
r
ival
ha
s
im
por
tant
va
lue
to
gu
ide
ne
twor
k
of
the
pos
it
ion
of
the
s
ou
r
c
e
s
to
dir
e
c
t
the
s
ignals
towa
r
d
the
pr
ope
r
dir
e
c
ti
on
,
howe
ve
r
,
none
of
thes
e
wor
k
s
ha
ve
a
ddr
e
s
s
e
d
the
a
c
tual
c
ondit
ions
a
s
will
be
d
is
c
us
s
e
d
in
thi
s
manu
s
c
r
ipt
[
9
-
11]
.
I
n
thi
s
a
r
ti
c
le,
we
d
is
c
us
s
e
d
two
is
s
ue
s
that
a
r
e
s
tr
o
ngly
r
e
late
d
to
the
W
i
-
F
i
s
tand
a
r
ds
that
c
or
r
e
s
pond
to
the
de
tec
ti
on
of
r
a
diation
s
our
c
e
s
whe
n
we
s
witch
to
higher
f
r
e
que
nc
ies
a
nd
the
pr
e
s
e
nc
e
of
a
nois
y
e
nvir
onment.
T
he
r
e
f
or
e
,
to
e
s
ti
m
a
te
the
dir
e
c
ti
on
o
f
a
r
r
ival
,
we
us
e
d
the
mos
t
pr
om
is
ing
r
oot
-
W
S
F
a
l
gor
it
hm,
whe
r
e
thes
e
thr
e
e
W
i
-
Fi
s
tanda
r
ds
will
be
e
xa
mi
ne
d
a
c
c
or
ding
to
the
f
oll
owing
pa
r
ts
:
−
W
e
pr
e
s
e
nt
a
c
ompar
a
ti
ve
s
tudy
of
the
th
r
e
e
pr
opo
s
e
d
W
i
-
F
i
s
tanda
r
ds
in
a
pe
r
f
e
c
t
c
a
s
e
withou
t
no
is
e
,
−
T
he
n
we
pe
r
f
o
r
m
a
s
ys
tem
in
a
pa
r
ti
a
ll
y
nois
y
c
a
s
e
,
−
F
inally,
we
int
r
oduc
e
a
s
ys
tem
c
ompl
e
te
ly
im
mer
s
e
d
in
the
nois
e
.
T
he
r
e
s
t
of
the
a
r
ti
c
le
de
a
ls
with
pa
r
ts
that
ha
ve
not
be
e
n
mentioned
be
f
or
e
,
they
a
r
e
dis
tr
ibut
e
d
a
s
f
oll
ows
.
I
n
s
e
c
ti
on
2
,
we
p
r
e
s
e
nt
the
s
e
a
r
c
h
methods
by
whic
h
we
us
e
the
DO
A
e
s
ti
mation
tec
hniques
,
then
the
W
i
-
F
i
ne
twor
ks
we
wor
ke
d
with.
I
n
s
e
c
ti
on
3
,
we
dis
c
us
s
the
a
s
p
e
c
t
of
the
pr
opos
e
d
s
ys
tem
s
.
I
n
s
e
c
ti
on
4
,
we
c
onc
lude
with
c
onc
lus
ions
.
F
igur
e
1.
5G
f
r
e
que
nc
ies
ba
nd
[
4]
2.
RE
S
E
AR
CH
M
E
T
HO
D
2.
1.
DOA
e
s
t
i
m
at
ion
T
he
a
im
o
f
the
DO
A
e
s
ti
mation
is
to
us
e
the
inf
o
r
mation
r
e
c
e
ived
a
t
the
a
ntenna
a
r
r
a
y
to
e
s
ti
mate
the
dir
e
c
ti
on
of
the
s
ignals
.
I
nde
e
d,
e
s
ti
mating
the
dir
e
c
ti
on
of
the
a
r
r
ival
a
ngle
p
r
e
s
e
nts
thr
e
e
major
dif
f
iculti
e
s
:
a
n
unknown
number
o
f
s
ignals
s
im
ult
a
ne
ous
ly
s
tr
iki
ng
the
a
r
r
a
y,
unknown
di
r
e
c
ti
ons
a
nd
a
mpl
it
ude
s
.
Als
o,
the
f
a
c
t
that
the
r
e
c
e
ived
s
ignals
a
r
e
c
ons
tantly
c
or
r
upted
by
nois
e
.
I
n
thi
s
c
ontext,
we
will
f
oc
us
on
the
pr
oblem
o
f
a
s
ys
tem
c
or
r
upted
by
n
ois
e
.
T
he
F
igu
r
e
2
pr
e
s
e
nts
the
ba
s
ic
model
of
D
OA
[
12]
.
T
he
r
e
a
r
e
s
e
ve
r
a
l
tec
hniques
f
or
e
s
ti
mating
the
d
ir
e
c
ti
on
of
a
r
r
ival
,
including
M
USI
C
a
lgor
i
thm
,
E
S
P
R
I
T
a
lgor
it
hm,
C
a
pon,
a
nd
other
s
.
I
n
thi
s
r
e
s
e
a
r
c
h,
we
will
us
e
the
W
S
F
a
lgor
it
hm
tha
t
pr
ove
d
it
s
e
f
f
e
c
ti
v
e
ne
s
s
in
pr
e
vious
wor
ks
[
10
,
13
,
14]
.
2.
1.
1
.
WS
F
algori
t
h
m
L
a
be
ll
e
d
we
ight
e
d
S
ubs
pa
c
e
F
it
ti
ng
a
lgo
r
it
hm
is
a
n
a
s
ympt
oti
c
a
ll
y
e
f
f
icie
nt
pa
r
a
metr
ic
method
us
e
d
to
e
s
ti
mate
the
he
ight
s
of
dif
f
e
r
e
nt
s
c
a
tt
e
r
e
r
s
in
the
s
a
me
a
z
im
uth
-
r
a
nge
r
e
s
olut
ion
c
e
ll
[
15
]
.
T
his
met
hod
c
a
n
de
tec
t
the
di
r
e
c
ti
on
o
f
a
r
r
ival
by
us
ing
the
we
ight
e
d
ve
r
s
i
on
of
a
matr
ix
whos
e
c
olum
ns
a
r
e
the
s
t
e
e
r
ing
ve
c
tor
s
a
s
s
oc
iate
d
with
thes
e
dir
e
c
ti
ons
in
c
los
e
to
a
da
t
a
-
de
pe
nding
matr
ix
[
16]
.
W
S
F
a
lgor
it
h
m
is
c
ons
i
de
r
e
d
a
s
a
unif
ied
a
ppr
oa
c
h
to
s
c
he
mes
a
s
M
USI
C
a
nd
E
S
P
R
I
T
a
lgor
it
hms
,
it
a
ls
o
r
e
qui
r
e
s
knowle
dge
of
th
e
number
of
dir
e
c
ti
ona
l
s
o
ur
c
e
s
,
a
nd
the
us
e
of
the
de
c
ompo
s
it
ion
tec
hnique
f
or
the
e
igenva
lues
.
T
his
a
pp
r
oa
c
h
uti
li
z
e
s
the
s
tr
onge
s
t
e
igenve
c
tor
s
in
a
diagona
l
matr
ix
(
̂
)
a
nd
the
matc
hing
e
igenve
c
tor
s
in
the
s
ignal
s
u
bs
pa
c
e
matr
ix
(
̂
)
.
T
he
e
xp
r
e
s
s
ion
of
W
S
F
a
lgor
it
hm
c
a
n
be
wr
it
ten
a
s
:
̂
=
(
(
П
(
)
̂
̂
)
)
(
1)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
3
,
J
une
2020:
1150
-
115
7
1152
whe
r
e
П
(
)
r
e
pr
e
s
e
nt
T
he
pr
ojec
ti
on
matr
ix
onto
the
c
o
lum
n
s
pa
c
e
of
a
(
θ)
,
a
nd
W
is
a
we
ight
ing
matr
ix
t
o
r
e
duc
e
the
im
pa
c
t
o
f
the
s
ubs
pa
c
e
s
wa
p
[
11]
.
F
or
a
be
tt
e
r
unde
r
s
tanding
of
thi
s
e
xp
r
e
s
s
ion,
we
ne
e
d
to
know
thes
e
f
or
mul
a
s
:
П
(
)
=
(
)
(
)
†
(
2)
(
)
†
=
(
(
)
(
)
)
−
1
(
)
(
3)
=
(
̂
−
2
̂
2
+
̂
2
̂
−
1
)
(
4)
̂
2
=
1
−
∑
̂
,
∗
−
=
1
(
5)
he
r
e
(
)
†
I
s
the
ps
e
udo
-
inver
s
e
of
a
(
θ)
,
̂
2
is
the
nois
e
va
r
ianc
e
,
̂
is
e
igenve
c
tor
s
in
a
diagona
l
nois
e
matr
ix,
the
M
is
the
number
of
tar
ge
ts
,
N
is
the
numbe
r
of
s
e
ns
or
s
a
nd
K
is
the
number
o
f
s
na
ps
hots
.
F
igur
e
2.
T
he
ba
s
ic
model
of
DO
A
e
s
ti
mation
[
12]
2.
1.
2.
Root
-
WS
F
algorit
h
m
R
oot
-
W
S
F
is
the
r
ooti
ng
ve
r
s
ion
of
we
ight
e
d
s
ubs
pa
c
e
f
it
ti
ng
.
I
n
th
is
s
tudy,
we
c
hos
e
to
us
e
thi
s
a
lgor
it
hm
f
o
r
be
tt
e
r
a
c
c
ur
a
c
y.
T
he
pur
pos
e
o
f
thi
s
t
e
c
hnique
is
to
mi
nim
ize
the
c
os
t
f
unc
ti
on
with
[
17]
:
(
)
=
(
(
)
⊥
̂
̂
)
(
6)
whe
r
e
:
(
)
⊥
=
−
(
)
(
(
)
(
)
)
−
1
(
)
(
7)
=
(
̂
−
̂
2
)
̂
−
1
(
8)
̂
2
=
1
−
(
̂
)
(
9)
he
r
e
(
)
⊥
indi
c
a
te
the
o
r
thogonal
pr
ojec
ti
on
ma
tr
ix
o
f
the
a
r
r
a
y
s
tee
r
ing
matr
ix,
is
the
a
s
ympt
oti
c
-
opti
mum
we
ight
matr
ix
a
nd
s
a
me
a
s
a
bove
,
̂
2
r
e
pr
e
s
e
nt
the
nois
e
va
r
ianc
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
T
he
impac
t
of
nois
e
on
de
tec
ti
ng
the
ar
r
ival
angle
us
ing
…
(
B
ti
s
s
am
B
ous
tani
)
1153
2.
2.
I
E
E
E
802.
11
s
t
an
d
ar
d
s
:
Wi
-
F
i
f
am
il
y
T
he
I
ns
ti
tut
e
of
e
lec
tr
ica
l
a
nd
e
lec
tr
onics
e
nginee
r
s
(
I
E
E
E
)
ha
s
de
ve
loped
a
f
a
mi
ly
of
802.
11
c
ompl
iant
s
pe
c
if
ica
ti
ons
f
or
wi
r
e
les
s
loca
l
a
r
e
a
ne
twor
k
(
W
L
AN
)
tec
hnology,
a
ls
o
known
a
s
W
i
-
F
i.
T
he
s
e
f
a
mi
li
e
s
ha
ve
many
s
pe
c
if
ica
ti
ons
.
A
letter
is
a
dde
d
to
de
s
c
r
ibe
thei
r
c
ha
r
a
c
ter
is
ti
c
s
s
uc
h
a
s
da
ta
r
a
tes
,
f
r
e
que
nc
y
ba
nd,
e
tc.
[
7,
18]
.
T
his
s
tanda
r
d
is
ba
s
e
d
on
tw
o
ba
s
ic
pr
oto
c
ols
,
media
a
c
c
e
s
s
c
ontr
ol
(
M
AC
)
a
nd
(
P
HY
)
[
19
]
.
T
he
I
E
E
E
802
.
11
ne
twor
k
in
c
ludes
s
e
ve
r
a
l
ba
s
ic
s
e
r
vice
s
e
ts
,
in
whic
h
a
number
of
wir
e
les
s
s
tations
tr
a
ns
mi
t
or
r
e
c
e
ive
f
r
om
a
s
ingl
e
a
c
c
e
s
s
po
int
s
hown
in
F
igur
e
3.
T
he
f
o
ll
owing
T
a
ble
1
out
li
ne
s
s
ome
of
thes
e
s
tanda
r
ds
in
ter
ms
o
f
the
ope
r
a
ti
ng
f
r
e
que
nc
y
a
nd
da
ta
r
a
te
[
20,
21]
.
F
igur
e
3.
W
i
-
F
i
pr
e
s
e
ntation
[
18]
T
a
ble
1.
I
E
E
E
802.
11
s
tanda
r
ds
[
20]
P
r
ot
oc
ol
O
pe
r
a
ti
ng f
r
e
que
nc
y
D
a
ta
r
a
te
(
ma
x)
802.11
2.4 G
H
z
2 M
bi
t/
s
802.11a
5G
H
z
54 M
bi
t/
s
802.11n
2.4 G
H
z
-
5
G
H
z
72 M
bi
t/
s
802.11a
d
60 G
H
z
6.75 G
bi
t/
s
2.
2.
1
.
802.
11a
T
he
I
E
E
E
802
.
11n
s
tanda
r
d
is
the
f
ir
s
t
e
xpa
ns
ion
s
c
he
me;
it
ope
r
a
tes
a
t
a
5
GH
z
r
a
diof
r
e
que
nc
y
a
nd
a
20
M
Hz
ba
ndwidth
a
nd
c
or
r
e
s
ponds
to
the
u
s
e
of
s
ingl
e
-
input
a
ntenna
tec
hnologi
e
s
(
S
I
S
O)
[
21
]
.
2.
2.
2
.
802.
11
b
T
he
I
E
E
E
802.
11b
s
tanda
r
d
is
c
ons
ider
e
d
the
f
ir
s
t
W
i
-
F
i
ne
two
r
k
that
ope
r
a
tes
a
r
ound
the
2.
4
GH
z
r
a
dio
f
r
e
que
nc
y.
T
his
ba
nd
wa
s
li
mi
ted
to
the
u
s
e
of
indus
tr
ial,
s
c
ientif
ic
a
nd
medic
a
l
(
I
S
M
)
e
q
uipm
e
nt.
F
or
tunate
ly,
the
F
C
C
(
F
e
de
r
a
l
C
omm
unica
ti
ons
C
omm
unica
ti
on)
ha
s
de
r
e
gulate
d
th
is
ba
nd
to
take
a
dva
ntage
of
wide
r
us
e
.
T
he
maximu
m
theor
e
ti
c
a
l
da
ta
r
a
te
th
a
t
thi
s
s
tanda
r
d
c
a
n
pr
ovide
c
a
n
be
up
to
11
M
bps
.
How
e
ve
r
,
in
pr
a
c
ti
c
e
,
th
is
s
pe
e
d
is
not
a
c
hieva
ble,
whic
h
is
w
hy
other
s
tanda
r
ds
ha
ve
be
e
n
pr
opos
e
d
to
s
olve
thi
s
pr
oblem
a
nd
of
f
e
r
be
tt
e
r
pe
r
f
or
manc
e
[
22]
.
2.
2.
3
.
802.
11
n
T
he
I
E
E
E
802
.
11n
s
tanda
r
d
r
e
f
e
r
s
to
W
i
-
F
i
4
o
r
dua
l
-
ba
nd
W
i
-
F
i,
or
W
i
-
F
i
Alli
a
nc
e
us
e
s
two
f
r
e
que
nc
ies
ba
nd
2.
4
GH
z
a
nd
5
GH
z
.
T
his
is
a
n
i
mpr
ove
ment
of
both
s
tanda
r
ds
802.
11
a
,
b.
i
t
is
c
ons
ider
e
d
the
f
ir
s
t
s
tanda
r
d
a
c
knowle
dging
M
I
M
O
tec
h
nology
[
11,
23]
.
2.
2.
4
.
802.
11a
d
T
he
I
E
E
E
802
.
11a
d
s
tanda
r
d,
a
ls
o
known
a
s
W
i
-
Gig
f
or
W
ir
e
les
s
Giga
bit
Alli
a
nc
e
,
c
e
r
ti
f
ied
by
Wi
-
F
i,
ope
r
a
tes
in
the
60
GH
z
f
r
e
que
nc
y
r
a
nge
,
whic
h
is
s
uit
a
ble
f
or
5G
a
ppli
c
a
ti
ons
.
T
his
tec
hnology
us
e
s
much
lar
ge
r
ult
r
a
-
wid
e
ba
nd
c
ha
nne
ls
,
much
highe
r
s
pe
c
tr
um
ba
nd,
f
a
s
t
da
ta
tr
a
ns
mi
s
s
ion
r
a
te,
a
ntenna
a
r
r
a
y,
be
a
mf
or
mi
ng,
a
nd
s
o
on.
How
e
ve
r
,
it
is
li
mi
ted
by
it
s
s
hor
t
dis
tanc
e
[
24,
25]
.
2.
3
.
Num
b
e
r
of
u
s
e
r
s
in
Wi
-
F
i
n
e
t
wor
k
s
T
he
numbe
r
o
f
us
e
r
s
of
W
i
-
F
i
a
c
c
e
s
s
point
e
f
f
e
c
ts
on
s
ignal
powe
r
a
nd
th
r
oughput.
De
vice
s
s
uc
h
a
s
c
omput
e
r
s
a
nd
s
mar
tphones
ne
e
d
to
s
ha
r
e
li
mi
ted
r
e
s
our
c
e
c
a
pa
c
it
y
on
a
ne
twor
k,
e
a
c
h
de
vice
c
onn
e
c
ted
to
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
3
,
J
une
2020:
1150
-
115
7
1154
the
wir
e
les
s
ne
twor
k
us
e
s
a
li
tt
le
mor
e
ba
ndwidth
a
nd
mus
t
be
ge
ne
r
a
ted
s
omew
he
r
e
onc
e
the
m
a
xim
um
ba
ndwidth
is
r
e
a
c
he
d
[
26
]
.
I
n
theo
r
y,
a
wir
e
les
s
r
ou
ter
c
a
n
s
uppor
t
250
de
vice
s
c
onne
c
ted
to
the
W
i
-
F
i
ne
twor
k.
I
n
p
r
a
c
ti
c
e
,
s
ome
mobi
le
p
r
ovider
s
c
ons
ider
that
t
he
maximum
number
o
f
us
e
r
s
c
a
n
r
e
a
c
h
up
to
12
5
us
e
r
s
.
How
e
ve
r
,
ther
e
is
a
f
or
mul
a
f
or
c
a
lcula
ti
ng
the
nu
mber
of
us
e
r
s
ba
s
e
d
on
da
ta
r
a
te
a
nd
W
i
-
F
i
thr
ou
ghput
a
s
f
oll
ows
[
26]
:
=
(
10)
the
number
of
us
e
r
s
in
a
h
ome
who
us
e
the
I
nter
ne
t
with
mul
ti
ple
de
vice
s
a
t
the
s
a
me
ti
me
is
a
c
r
it
ica
l
f
a
c
tor
in
de
ter
mi
ning
the
I
n
ter
ne
t
s
pe
e
ds
ne
e
de
d
a
t
the
p
oint
of
maximum
us
e
.
B
a
s
e
d
on
thi
s
f
o
r
mul
a
,
we
c
a
n
a
ls
o
de
ter
mi
ne
the
li
mi
ts
of
the
da
ta
t
r
a
ns
mi
s
s
ion
r
a
te
t
ha
t
a
us
e
r
c
a
n
be
ne
f
it
f
r
om
s
hown
in
T
a
ble
2.
F
or
ba
s
ic
us
e
of
W
i
-
F
i
2.
4
GH
z
,
s
ix
us
e
r
s
s
ha
r
e
the
c
onne
c
ti
on
with
a
low
b
it
r
a
te
of
one
M
bit
/s
.
T
a
ble
2
.
M
a
xim
um
number
of
us
e
r
s
ba
s
e
d
on
the
d
a
ta
r
a
te
Wi
-
F
i
a
Wi
-
Fi
b
Wi
-
Fi
n
Wi
-
Fi
ad
M
in
im
um
(
1M
bps
)
27 us
e
r
s
6 us
e
r
s
72 us
e
r
s
3375 us
e
r
s
B
a
s
ic
(
3M
bp
s
)
9 us
e
r
s
2 us
e
r
s
24 us
e
r
s
1125 us
e
r
s
M
ode
r
a
te
(
10 M
bps
)
3 us
e
r
s
0 us
e
r
s
7 us
e
r
s
337 us
e
r
s
I
de
a
l
(
35 M
bps
)
0 us
e
r
s
0 us
e
r
s
2 us
e
r
s
96 us
e
r
s
H
e
a
vy
(
75 M
bps
)
0 us
e
r
s
0 us
e
r
s
1
us
e
r
45 us
e
r
s
3.
RE
S
UL
T
S
A
ND
AN
AL
YSI
S
T
he
pu
r
pos
e
of
thi
s
s
tudy
is
to
e
va
luate
the
pe
r
f
or
manc
e
of
the
W
S
F
a
lgor
it
hm
to
de
tec
t
the
a
r
r
ival
a
ngles
in
a
nois
y
s
ys
tem,
whe
r
e
it
is
dif
f
icult
to
dis
ti
nguis
h
the
r
e
c
e
ived
s
ignals
.
T
o
do
s
o,
we
c
ons
ider
a
unif
or
m
li
ne
a
r
a
r
r
a
y
of
10
e
leme
nts
with
int
e
r
s
pa
c
ing
o
f
λ
/2,
s
ix
r
e
c
e
ived
s
ignals
with
the
r
e
s
pe
c
ti
ve
a
ngles
o
f
a
r
r
ival
(
AO
A)
ϴ
1
=
-
60°;
ϴ
2
=
-
50°;
ϴ
3
=
-
30°;
ϴ
4
=
5°;
ϴ
5
=
20°;
ϴ
6
=
30°;
ϴ
7
=
50°;
ϴ
8
=
60°.
S
ince
we
a
r
e
in
a
c
r
it
ica
l
si
tuation
whe
r
e
the
s
ys
tem
e
mer
ge
s
with
white
Ga
us
s
ian
nois
e
,
s
ome
a
ngle
s
of
a
r
r
ival
a
r
e
too
c
los
e
,
whic
h
c
r
e
a
tes
a
ne
w
c
ons
tr
a
int
f
o
r
the
de
tec
ti
on
o
f
the
a
ngles
of
a
r
r
ival
,
s
uc
h
a
s
the
10°
s
e
pa
r
a
ti
on
be
t
we
e
n
ϴ
1
a
nd
ϴ
2
,
s
a
me
with
ϴ
5
a
nd
ϴ
6
, ϴ
7
a
nd
ϴ
8
.
T
his
r
e
s
e
a
r
c
h
is
ba
s
e
d
on
W
i
-
F
i
a
ppli
c
a
ti
ons
us
i
ng
dif
f
e
r
e
nt
f
r
e
que
nc
ies
ba
nds
f
r
om
2.
4
GHz
to
60
GH
z
a
nd
a
ba
s
ic
model
of
da
ta
r
a
te
with
e
ight
u
s
e
r
s
.
I
n
the
f
oll
owing
s
e
c
ti
on,
we
will
inves
ti
ga
te
th
e
im
pa
c
t
of
a
ll
thes
e
c
r
it
e
r
ia.
F
i
r
s
t,
r
e
ga
r
ding
a
pe
r
f
e
c
t
c
a
s
e
without
nois
e
,
then
incr
e
a
s
ing
s
igni
f
ica
ntl
y
the
nois
e
va
lue,
a
nd
f
inally
,
with
a
s
ys
tem
in
whic
h
nois
e
is
domi
na
nt.
T
he
s
im
ulations
pr
e
s
e
nted
in
thi
s
a
r
ti
c
le
we
r
e
m
a
de
with
M
AT
L
AB
a
nd
S
I
M
UL
I
NK
R
2018a
.
3.
1.
A
p
e
r
f
e
c
t
s
ys
t
e
m
wit
h
ou
t
n
ois
e
B
a
s
e
d
on
the
s
ys
tem
de
s
c
r
ibed
a
bove
,
we
wil
l
e
va
luate
the
p
r
opos
e
d
DO
A
a
lgor
it
hm
(
r
oot
-
W
S
F
)
in
a
p
e
r
f
e
c
t
s
ys
tem
without
nois
e
.
T
he
r
e
s
ult
of
thi
s
wor
k
is
given
in
the
f
oll
owing
table
.
Ac
c
or
ding
to
T
a
ble
3,
we
c
a
n
c
lea
r
ly
noti
c
e
that
a
ll
a
ngles
of
a
r
r
ival
a
r
e
we
ll
de
tec
ted
in
the
thr
e
e
W
i
-
F
i
a
ppli
c
a
ti
ons
,
whic
h
is
quit
e
logi
c
a
l
in
the
a
bs
e
nc
e
of
nois
e
.
T
his
s
ys
tem
will
be
c
ons
ider
e
d
a
s
a
r
e
f
e
r
e
nc
e
f
or
the
s
tudi
e
s
e
s
tab
li
s
he
d
in
the
f
oll
owing
s
e
c
ti
ons
.
T
a
ble
3.
S
ys
tem
without
nois
e
A
ngl
e
s
Wi
-
F
i
2.4 G
H
z
Wi
-
F
i
5G
H
z
Wi
-
G
ig
60
G
H
z
-
60°
-
60.21°
-
60.31°
-
60.18°
-
50°
-
55.65°
-
53.75°
-
48.93°
-
30°
-
29.99°
-
30
.1°
-
30.01°
5°
5.21°
5.06°
5.19°
20°
19.49°
14.89°
15.19°
30°
27.41°
27.42°
28.7°
50°
49.78°
49.68°
49.77°
60°
59.72°
59.82°
59.93°
3.
2
.
S
ys
t
e
m
wit
h
p
ar
t
ial
n
ois
e
T
he
s
a
me
s
ys
tem
is
us
e
d
a
s
be
f
or
e
,
a
dding
a
n
a
ddi
ti
ve
Ga
us
s
ian
nois
e
to
e
a
c
h
of
the
r
e
c
e
ived
s
ignals
to
pr
ovide
a
ne
a
r
-
r
e
a
l
wor
ld
s
ys
tem.
I
n
thi
s
c
ontext,
we
will
e
va
luate
the
s
ys
tem’
s
r
e
s
pons
e
in
ter
ms
of
de
tec
ti
ng
the
a
ngles
of
a
r
r
ival
us
ing
the
r
oot
-
W
S
F
a
lgor
it
hm
,
while
a
dding
the
AW
GN
nois
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
T
he
impac
t
of
nois
e
on
de
tec
ti
ng
the
ar
r
ival
angle
us
ing
…
(
B
ti
s
s
am
B
ous
tani
)
1155
3.
2.
1
.
S
ys
t
e
m
wit
h
S
NR
=
20
I
n
thi
s
s
e
c
ti
o
n,
we
c
ons
ider
that
ou
r
s
ys
tem
int
r
o
duc
e
s
nois
e
with
a
s
ignal
-
to
-
nois
e
r
a
ti
o
of
20
dB
while
r
e
s
pe
c
ti
ng
the
s
a
me
s
pe
c
if
ica
ti
ons
pr
e
vious
ly
us
e
d.
W
e
mus
t
a
ls
o
take
int
o
a
c
c
ount
that
one
of
the
W
i
-
F
i
a
ppli
c
a
ti
ons
us
e
d
(
W
i
-
Gig)
ope
r
a
tes
i
n
a
high
-
f
r
e
que
nc
y
ba
nd
up
to
60
GH
z
.
T
he
f
oll
ow
ing
table
pr
ovides
the
r
e
s
ult
s
of
thi
s
s
im
ulation.
T
a
ble
4
s
hows
the
e
f
f
e
c
t
o
f
us
ing
20
dB
nois
e
on
the
thr
e
e
W
i
-
F
i
a
ppli
c
a
ti
ons
.
W
e
c
a
n
obvious
ly
noti
c
e
that
the
r
oot
-
W
S
F
a
lgo
r
it
hm
a
ll
ows
pe
r
f
e
c
t
de
tec
ti
on
e
ve
n
in
the
pr
e
s
e
nc
e
of
nois
e
f
or
a
ll
f
r
e
que
nc
y
ba
nds
.
I
nde
e
d,
the
s
ignal
is
much
mor
e
im
por
tant
than
the
nois
e
,
whic
h
r
e
s
ult
s
in
the
pos
it
ive
va
lue
o
f
the
S
NR
.
T
a
ble
4.
S
ys
tem
with
pa
r
ti
a
l
no
is
e
o
f
S
NR
=
20
A
ngl
e
s
Wi
-
F
i
2.4 G
H
z
Wi
-
F
i
5G
H
z
Wi
-
G
ig
60
G
H
z
-
60°
-
59.89°
-
60.93°
-
60.06°
-
50°
-
43.5°
-
54.19°
-
48.64°
-
30°
-
30.02°
-
29.97°
-
30°
5°
5.18°
5.60°
5.30°
20°
21.32°
20.97°
12.85°
30°
34.12°
32.17°
28.33°
50°
50.41°
55.72°
49.82°
60°
60.36°
59.07°
59.83°
3.
2.
2
.
S
ys
t
e
m
wit
h
S
NR
=
-
50
I
n
or
de
r
to
e
va
luate
the
im
pa
c
t
of
the
no
is
e
in
the
s
ys
tem
we
us
e
d,
we
opted
f
or
a
s
ignal
whos
e
input
nois
e
is
much
im
por
tant
than
the
incoming
s
ignal
with
a
va
lue
o
f
-
50dB
.
T
a
ble
5
s
hows
the
r
e
s
ult
s
obtaine
d
f
or
the
th
r
e
e
p
r
opos
e
d
W
i
-
F
i
a
ppli
c
a
ti
ons
.
Ac
c
or
di
ng
to
T
a
ble
5,
the
W
S
F
a
lgor
it
h
m
ha
s
onc
e
a
ga
in
p
r
ove
d
it
s
e
f
f
e
c
ti
ve
ne
s
s
in
de
tec
ti
ng
a
r
r
ival
a
ngles
e
ve
n
in
t
he
pr
e
s
e
nc
e
of
nois
e
.
How
e
ve
r
,
ther
e
is
a
s
mall
d
if
f
e
r
e
nc
e
be
twe
e
n
the
thr
e
e
W
i
-
F
i
a
ppli
c
a
ti
ons
in
ter
ms
of
a
c
c
ur
a
c
y.
T
his
dif
f
e
r
e
nc
e
will
be
dis
c
us
s
e
d
f
ur
ther
in
the
pr
e
c
is
ion
a
nd
pr
e
c
is
ion
pa
r
t
.
T
a
ble
5.
S
ys
tem
pa
r
ti
a
l
nois
e
of
S
NR
=
-
50
A
ngl
e
s
Wi
-
F
i
2.4 G
H
z
Wi
-
F
i
5G
H
z
Wi
-
G
ig
60
G
H
z
-
60°
-
60.1°
-
59.84°
-
60.08°
-
50°
-
47.4°
-
45.44°
-
47.1°
-
30°
-
29.99°
-
30. 44°
-
29.99°
5°
5.49°
4.80°
5.48°
20°
11.73°
21.24°
12.37°
30°
28.44
°
33.78°
28.61°
50°
49.81°
41.07°
49.82°
60°
59.83°
58.79°
59.86°
3.
3.
S
ys
t
e
m
wit
h
m
as
s
ive
n
ois
e
I
n
thi
s
pa
r
t,
we
e
va
luate
our
s
ys
tem
in
c
r
it
ica
l
c
a
s
e
s
,
whe
r
e
it
e
mer
ge
s
c
ompl
e
tely
in
no
is
e
.
T
his
s
tudy
will
be
divi
de
d
int
o
two
pa
r
ts
;
the
f
ir
s
t
will
t
r
e
a
t
a
n
S
NR
of
20dB
a
nd
the
s
e
c
ond
one
o
f
-
50dB
.
3.
3.
1.
S
ys
t
e
m
wit
h
S
NR
=
20
T
he
s
ignal
-
to
-
nois
e
r
a
ti
o
(
S
NR
)
va
r
ies
f
r
om
a
pos
it
ive
va
lue
to
a
n
e
ga
ti
ve
one
,
in
th
is
a
na
lys
e
s
,
we
c
hos
e
to
e
va
luate
the
pe
r
f
or
manc
e
of
the
nois
e
w
he
n
the
s
ignal
is
mor
e
powe
r
f
ul
than
the
pr
opos
e
d
nois
e
.
T
he
r
e
s
ult
s
of
thi
s
s
tudy
a
r
e
given
in
the
f
ol
lowing
T
a
ble
6
.
I
t
is
obvious
that
the
im
pa
c
t
o
f
nois
e
c
a
nn
ot
a
f
f
e
c
t
the
pe
r
f
or
manc
e
of
the
s
ugge
s
ted
s
ys
tem.
How
e
ve
r
,
it
is
a
ls
o
tr
ue
that
the
ope
r
a
ti
ng
f
r
e
que
nc
y
plays
a
n
im
por
tant
r
ole
in
de
ter
mi
ning
the
a
r
r
ival
a
ngle
de
tec
ti
on.
T
he
higher
the
f
r
e
que
nc
y
the
mor
e
the
s
ys
tem
be
c
omes
mor
e
s
e
ns
it
ive
to
nois
e
but
in
a
n
ins
igni
f
ica
nt
wa
y.
W
hich
s
hows
the
e
f
f
icie
nc
y
of
our
r
oot
-
W
S
F
a
lgor
it
hm
.
T
a
ble
6.
S
ys
tem
with
tot
a
l
nois
e
of
S
NR
=
20
A
ngl
e
s
Wi
-
F
i
2.4 G
H
z
Wi
-
F
i
5G
H
z
Wi
-
G
ig
60
G
H
z
-
60°
-
60.01°
-
59.7°
-
60.5°
-
50°
-
48.93°
-
49.36°
-
53.22°
-
30°
-
29.99°
-
30.04°
-
29.99°
5°
5.15°
5.42°
5.02°
20°
16.26°
19.54°
12.09°
30°
28.94°
25.31°
27.27°
50°
49.92°
43.26°
49.56°
60°
59.84°
58.86
°
59.7°
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
3
,
J
une
2020:
1150
-
115
7
1156
3.
3.
2.
S
ys
t
e
m
wit
h
S
NR
=
-
50
I
n
or
de
r
to
c
ompl
e
te
ou
r
s
tudy,
it
is
ne
c
e
s
s
a
r
y
to
t
a
ke
int
o
a
c
c
ount
the
c
a
s
e
whe
r
e
the
nois
e
is
mor
e
powe
r
f
ul
than
the
input
s
ignal.
T
he
r
e
s
ult
of
thi
s
s
im
ulation
is
int
e
r
pr
e
ted
in
T
a
ble
7
.
T
he
r
e
s
ult
a
c
h
ieve
d
in
T
a
bl
e
7
s
hows
the
im
pa
c
t
of
nois
e
on
de
tec
ti
ng
the
a
r
r
ival
s
ignals
in
thr
e
e
dif
f
e
r
e
nt
f
r
e
que
nc
y
ba
nds
of
the
W
i
-
F
i,
the
R
OO
T
-
W
S
F
a
lgor
it
hm
gives
a
lm
os
t
identica
l
r
e
s
ult
s
in
a
lm
os
t
a
ll
c
a
s
e
s
,
with
a
mi
nor
mar
gin
of
e
r
r
o
r
.
T
a
ble
7.
S
ys
tem
with
tot
a
l
nois
e
o
f
S
NR
=
-
50
A
ngl
e
s
Wi
-
F
i
2.4 G
H
z
Wi
-
F
i
5G
H
z
Wi
-
G
ig
60
G
H
z
-
60°
-
60.1°
-
60.2°
-
60.72°
-
50°
-
47.42°
-
52.82°
-
55.6°
-
30°
-
29.99°
-
29.99°
-
30.01°
5°
5.48°
5.91°
5.17°
20°
11.73°
12.14°
13.73°
30°
28.43°
23.84°
26.84°
50°
49.81°
45.92°
49.56°
60°
59.83°
5
9.56°
59.76°
3.
4.
Ac
c
u
r
ac
y
an
d
p
r
e
c
is
ion
T
he
a
c
c
ur
a
c
y
o
f
our
r
e
s
ult
s
is
a
n
im
po
r
tant
c
r
it
e
r
ion
in
the
de
tec
ti
on
o
f
a
r
r
ival
a
ngles
.
E
a
c
h
s
tudy
tr
e
a
ted
pr
e
vious
ly
ga
ve
a
ne
gli
gibl
e
ma
r
gin
o
f
e
r
r
or
in
ter
ms
o
f
pr
e
c
is
ion.
How
e
ve
r
,
to
e
va
luate
the
pe
r
f
or
man
ce
of
our
s
ys
tem
we
pr
opos
e
d
to
c
a
lcula
te
the
pe
r
c
e
nt
e
r
r
or
of
e
a
c
h
c
a
s
e
.
T
a
ble
8
a
nnounc
e
the
wor
k
do
ne
in
thi
s
f
ield.
T
a
ble
8
pr
e
s
e
nts
f
ive
d
if
f
e
r
e
nt
c
a
s
e
s
f
or
e
a
c
h
W
i
-
F
i
a
ppli
c
a
ti
on,
gr
oupe
d
a
s
f
oll
ows
:
no
is
e
-
f
r
e
e
s
ys
tem,
pa
r
ti
a
l
s
ignal
-
to
-
nois
e
r
a
ti
o
s
ys
tem
o
f
20
dB
a
nd
-
50
dB
,
a
nd
f
inally
,
a
s
ys
tem
with
a
to
tal
s
ig
na
l
-
to
-
nois
e
r
a
ti
o
of
20dB
a
nd
-
50
dB
.
W
e
c
a
n
s
e
e
that
f
o
r
W
i
-
F
i
of
2.
4
GH
z
,
the
a
c
c
ur
a
c
y
is
96
.
51%
with
a
pe
r
c
e
nt
e
r
r
or
of
3
.
49%
.
T
his
va
lue
incr
e
a
s
e
s
with
the
pr
e
s
e
nc
e
of
nois
e
s
;
it
goe
s
f
r
om
4
.
82%
to
7
.
80
%
f
or
pa
r
ti
a
l
nois
e
.
S
a
me
f
or
the
c
a
s
e
of
to
tal
nois
e
whe
r
e
the
pe
r
c
e
nt
e
r
r
or
goe
s
f
r
om
5
.
18%
to
7
.
77%
.
S
im
il
a
r
to
the
5
GH
z
W
i
-
F
i,
the
a
c
c
ur
a
c
y
o
f
a
s
ys
tem
without
nois
e
is
94
.
42%
with
a
pe
r
c
e
nt
e
r
r
or
of
5
.
58%
.
T
his
va
lue
incr
e
a
s
e
s
with
the
pr
e
s
e
nc
e
of
nois
e
s
;
a
nd
goe
s
f
r
om
5
.
89%
to
6
.
67%
f
or
pa
r
ti
a
l
nois
e
,
a
nd
f
r
om
s
a
me
5
.
46%
to
11.
63
%
in
a
s
ys
tem
with
tot
a
l
nois
e
.
T
he
pr
oc
e
s
s
is
the
s
a
me
f
or
W
i
-
Gig.
A
pe
r
f
e
c
t
s
ys
tem
ha
s
a
n
a
c
c
ur
a
c
y
of
95.
6%
with
a
n
e
r
r
or
pe
r
c
e
ntage
of
4.
40%
.
T
his
va
lu
e
incr
e
a
s
e
s
ve
r
y
s
li
g
htl
y
with
the
pr
e
s
e
nc
e
of
nois
e
s
;
a
nd
goe
s
f
r
om
6
.
34%
to
7
.
3
6%
f
or
pa
r
ti
a
l
nois
e
,
a
nd
f
r
om
7
.
21%
to
7
.
37%
in
a
s
ys
tem
with
tot
a
l
nois
e
,
whic
h
pr
ove
s
the
r
obus
tnes
s
of
ou
r
s
ys
tem
in
a
nois
y
e
nvir
onment.
T
a
ble
8
.
Ac
c
ur
a
c
y
of
our
s
ys
te
m
ba
s
e
d
on
the
pe
r
c
e
nt
e
r
r
or
Wi
-
F
i
2,4
Wi
-
F
i
5
Wi
-
F
i
60
S
ys
te
m wit
hout
noi
s
e
3
.
49%
5
.
58%
4
.
40%
P
a
r
ti
a
ll
y nois
y 20
4
.
82%
5
.
89%
6
.
34%
P
a
r
ti
a
ll
y nois
y
-
50
7
.
80%
6
.
67%
7
.
36%
to
ta
ll
y nois
y 20
5
.
18%
5
.
46%
7
.
21%
to
ta
ll
y nois
y
-
50
7
.
77%
11
.
63%
7
.
37%
4.
CONC
L
USI
ON
T
his
s
tudy
inves
ti
ga
tes
the
i
mpac
t
o
f
nois
e
on
de
tec
ti
ng
the
a
r
r
ival
a
ngle
us
ing
the
r
oo
t
-
W
S
F
a
lgor
it
hm.
I
n
o
r
de
r
to
a
c
hieve
thi
s
objec
ti
ve
,
we
r
e
li
e
d
on
the
a
ll
ianc
e
be
twe
e
n
the
numbe
r
of
us
e
r
s
s
ha
r
ing
the
s
a
me
W
i
-
F
i
Ac
c
e
s
s
point
a
t
the
s
a
me
ti
me,
s
ignal
s
tr
e
ngth
ve
r
s
us
the
nois
e
a
nd
thr
oughput
of
a
ba
s
ic
mode
of
us
e
a
t
3M
bps
.
T
o
c
a
r
r
y
out
thi
s
s
tudy,
we
de
ve
loped
a
s
ys
tem
c
ons
i
s
ti
ng
of
a
unif
or
m
li
ne
a
r
a
r
r
a
y
(
UL
A)
of
10
a
ntenna
e
leme
nts
with
a
s
pa
c
ing
of
λ
/2,
a
nd
whe
r
e
a
ll
the
s
our
c
e
s
a
r
e
a
s
s
umed
unc
or
r
e
late
d.
S
e
ve
r
a
l
mea
s
ur
e
ments
we
r
e
pe
r
f
or
med
to
e
ns
ur
e
th
e
pr
ope
r
f
unc
t
ioni
ng
of
our
s
ys
tem,
a
s
a
r
e
f
e
r
e
nc
e
,
we
f
ir
s
t
e
va
luate
d
the
pe
r
f
o
r
manc
e
of
a
pe
r
f
e
c
t
s
ys
tem
wi
thout
the
pr
e
s
e
nc
e
of
nois
e
.
I
n
thi
s
c
a
s
e
,
the
p
r
opos
e
d
r
oot
-
W
S
F
,
DO
A
a
lgor
it
hm
ga
ve
the
b
e
s
t
r
e
s
ult
s
in
ter
ms
of
de
tec
ti
ng
the
a
r
r
ival
a
ngles
in
the
th
r
e
e
a
ppli
c
a
ti
on
of
W
i
-
F
i:
2.
4
GH
z
,
5
GH
z
,
a
nd
60GH
z
.
T
he
n,
w
e
s
tar
ted
to
inves
ti
ga
te
two
other
pr
omi
s
ing
c
a
s
e
s
,
c
los
e
to
r
e
a
li
ty,
in
whic
h
the
nois
e
a
ppe
a
r
s
in
the
pa
r
t
iall
y
nois
y
s
ys
tem
a
nd
in
a
ve
r
y
nois
y
s
ys
tem.
C
om
pa
r
e
d
to
the
r
e
f
e
r
e
nc
e
,
the
r
oot
-
W
S
F
a
lgor
it
hm
s
tood
out
b
y
pr
oving
the
be
s
t
r
e
s
ult
in
a
lm
os
t
e
ve
r
y
s
it
ua
ti
ons
a
nd
f
or
the
dif
f
e
r
e
nt
W
i
-
F
i
a
p
pli
c
a
ti
ons
.
R
e
ga
r
ding
a
c
c
ur
a
c
y,
f
or
e
a
c
h
of
the
pr
opos
e
d
W
i
-
F
i
a
ppli
c
a
ti
ons
,
we
c
a
lcula
te
the
pe
r
c
e
nt
e
r
r
o
r
f
r
om
f
ive
dif
f
e
r
e
nt
pe
r
s
pe
c
ti
ve
s
to
de
ter
mi
ne
the
r
obus
t
ne
s
s
of
our
s
ys
tem.
T
he
s
e
c
ondit
ions
a
r
e
a
na
lyze
d
ba
s
e
d
on
nois
e
-
f
r
e
e
s
ys
tem,
pa
r
ti
a
l
s
igna
l
-
to
-
nois
e
r
a
ti
o
s
ys
tem
of
20
dB
a
nd
-
50
dB
,
a
nd
s
ys
tem
with
a
to
tal
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
T
he
impac
t
of
nois
e
on
de
tec
ti
ng
the
ar
r
ival
angle
us
ing
…
(
B
ti
s
s
am
B
ous
tani
)
1157
s
ignal
-
to
-
nois
e
r
a
ti
o
of
20
dB
a
nd
-
50
dB
.
As
a
c
on
c
lus
ion,
the
va
lue
of
the
pe
r
c
e
nt
e
r
r
o
r
incr
e
a
s
e
s
s
li
ghtl
y
with
the
pr
e
s
e
nc
e
of
nois
e
s
,
C
ha
nging
the
ope
r
a
ti
ng
f
r
e
que
nc
y
doe
s
not
a
f
f
e
c
t
the
r
obus
tnes
s
of
the
s
ys
tem,
a
nd
thus
pr
ovides
be
tt
e
r
pe
r
f
o
r
manc
e
in
the
de
tec
ti
on
of
th
e
a
ngles
of
a
r
r
ival
us
ing
the
r
oot
-
W
S
F
a
lgor
it
hm.
RE
F
E
RE
NC
E
S
[1
]
M.
J
ab
er,
M.
A
.
Imran
,
R.
T
afazo
l
l
i
,
A
.
T
u
k
man
o
v
,
“
5
G
Back
h
au
l
Ch
a
l
l
e
n
g
e
s
an
d
E
mer
g
i
n
g
Res
earc
h
D
i
rec
t
i
o
n
s
:
A
Su
rv
e
y
,
”
IE
E
E
A
cce
s
s
,
v
o
l
.
4
,
p
p
.
1
7
4
3
‑
1
7
6
6
,
2
0
1
6
.
[2
]
Y
.
H
u
o
,
X
.
D
o
n
g
,
W
.
X
u
,
et
M.
Y
u
en
,
“
Ce
l
l
u
l
ar
an
d
W
i
Fi
C
o
-
d
e
s
i
g
n
f
o
r
5
G
U
s
er
E
q
u
i
p
me
n
t
,
”
i
n
2
0
1
8
IE
E
E
5
G
W
o
r
l
d
F
o
r
u
m
(5
G
W
F
)
, p
p
.
2
5
6
‑
2
6
1
,
2
0
1
8
.
[3
]
E
.
M.
Mo
h
amed
,
M.
A
.
A
b
d
e
l
g
h
an
y
,
et
M.
Z
areei
,
“
A
n
E
ffi
ci
e
n
t
Para
d
i
g
m
fo
r
M
u
l
t
i
b
an
d
W
i
G
i
g
D
2
D
N
et
w
o
r
k
s
,
”
IE
E
E
A
cces
s
,
v
o
l
.
99
, p
p
.
7
0
0
3
2
‑7
0
0
4
5
,
2
0
1
9
.
[4
]
Y
.
Fel
d
man
,
A
.
Pu
zen
k
o
,
P.
Ben
Is
h
ai
,
A
.
Cad
u
ff,
et
A
.
J
.
A
g
ran
at
,
“
H
u
ma
n
Sk
i
n
as
A
rray
s
o
f
H
e
l
i
ca
l
A
n
t
en
n
a
s
i
n
t
h
e
Mi
l
l
i
met
er
a
n
d
Su
b
mi
l
l
i
met
er
W
av
e
Ran
g
e
”
,
P
h
ys
.
R
ev.
Let
t
.
,
v
o
l
.
1
0
0
,
n
o.
1
2
,
p
p
. 1
2
8
1
0
2
,
2
0
0
8
.
[5
]
5
G
s
p
ect
r
u
m
,
“
T
h
e
v
a
l
u
e
o
f
an
al
l
-
b
an
d
s
t
ra
t
eg
y
,”
2
0
1
8
.
[O
n
l
i
n
e].
A
v
ai
l
ab
l
e:
h
t
t
p
s
:
/
/
w
w
w
.
eri
cs
s
o
n
.
co
m
/
en
/
n
e
t
w
o
r
k
s
/
t
ren
d
i
n
g
/
h
o
t
-
t
o
p
i
c
s
/
5
g
-
s
p
ec
t
ru
m
-
s
t
ra
t
eg
i
es
-
to
-
ma
x
i
m
i
ze
-
al
l
-
b
an
d
s
.
[6
]
Ch
i
p
D
es
i
g
n
,
“
60
G
H
z
t
ech
n
o
l
o
g
y
o
p
e
n
s
d
o
o
r
t
o
W
i
G
i
g
®
an
d
5
G
ap
p
l
i
cat
i
o
n
s
,”
2
0
1
7
.
[O
n
l
i
n
e].
A
v
ai
l
ab
l
e
:
h
t
t
p
:
/
/
eecat
a
l
o
g
.
co
m/
c
h
i
p
d
e
s
i
g
n
/
2
0
1
7
/
1
1
/
1
6
/
6
0
g
h
z
-
t
ec
h
n
o
l
o
g
y
-
o
p
e
n
s
-
d
o
o
r
-
to
-
w
i
g
i
g
-
a
n
d
-
5g
-
a
p
p
l
i
ca
t
i
o
n
s
/
.
[7
]
Y
.
G
ao
,
L
.
D
ai
,
et
X
.
H
ei
,
“
T
h
r
o
u
g
h
p
u
t
O
p
t
i
m
i
zat
i
o
n
o
f
Mu
l
t
i
-
BSS
IE
E
E
8
0
2
.
1
1
N
e
t
w
o
rk
s
w
i
t
h
U
n
i
v
er
s
al
Freq
u
en
cy
Reu
se
,”
IE
E
E
Tr
a
n
s
.
Co
m
m
u
n
.
,
v
o
l
.
6
5
,
n
o.
8
,
p
p
.
3
3
9
9
‑
3
4
1
4
,
2
0
1
7
.
[8
]
C.
Ch
en
,
Y
.
C
h
en
,
Y
.
H
a
n
,
H
.
L
ai
,
&
K
.
J
.
R.
L
i
u
,
“
A
c
h
i
e
v
i
n
g
Ce
n
t
i
met
er
-
A
ccu
rac
y
I
n
d
o
o
r
L
o
ca
l
i
za
t
i
o
n
o
n
W
i
F
i
Pl
at
f
o
rms
:
A
Freq
u
en
c
y
H
o
p
p
i
n
g
A
p
p
r
o
ach
,
”
IE
E
E
In
t
e
r
n
e
t
Th
i
n
g
s
J.
,
v
o
l
.
4
,
n
o
1
,
p
p
.
1
1
1
‑1
2
1
,
2
0
1
7
.
[9
]
Y
.
Z
h
ao
,
L
.
Z
h
an
g
,
Y
.
G
u
,
Y
.
G
u
o
,
&
J
.
Z
h
an
g
,
“
E
ffi
ci
en
t
s
p
ars
e
rep
re
s
en
t
at
i
o
n
met
h
o
d
fo
r
w
i
d
e
b
an
d
D
O
A
es
t
i
mat
i
o
n
u
s
i
n
g
f
o
cu
s
i
n
g
o
p
era
t
i
o
n
,”
S
o
n
a
r
Na
v
i
g
.
I
E
T
R
a
d
a
r
,
v
o
l
.
1
1
,
n
o.
1
1
,
p
p
.
1
6
7
3
‑1
6
7
8
,
2
0
1
7
.
[1
0
]
B.
Bo
u
s
t
an
i
,
A
.
Bag
h
d
a
d
,
A
.
Sa
h
el
,
A
.
Ba
l
l
o
u
k
,
&
A
.
Ba
d
ri
,
“
Perfo
rma
n
ce
a
n
al
y
s
i
s
o
f
d
i
rect
i
o
n
o
f
arr
i
v
a
l
e
s
t
i
mat
i
o
n
u
n
d
er
h
ard
co
n
d
i
t
i
o
n
,
”
in
2
0
1
8
4
t
h
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
e
r
e
n
ce
o
n
O
p
t
i
m
i
z
a
t
i
o
n
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
(ICO
A
)
,
p
p
.
1
‑5
,
2018
.
[1
1
]
S.
O
u
el
h
a,
A
.
A
ï
s
s
a
-
El
-
Bey
,
&
B.
Bo
as
h
a
s
h
,
“
Imp
r
o
v
i
n
g
D
O
A
E
s
t
i
ma
t
i
o
n
A
l
g
o
r
i
t
h
ms
U
s
i
n
g
H
i
g
h
-
Res
o
l
u
t
i
o
n
Q
u
a
d
rat
i
c
T
i
me
-
Fre
q
u
e
n
cy
D
i
s
t
r
i
b
u
t
i
o
n
s
,
”
IE
E
E
Tr
a
n
s
.
S
i
g
n
a
l
P
r
o
ce
s
s
.
v
o
l
.
6
5
,
n
o.
1
9
,
p
p
.
5
1
7
9
‑5
1
9
0
,
o
ct
.
2
0
1
7
.
[1
2
]
S.
S.
Bal
ab
ad
ra
p
at
r
u
n
i
,
“
Perfo
rman
ce
E
v
al
u
at
i
o
n
o
f
D
i
r
ect
i
o
n
o
f
A
rr
i
v
a
l
E
s
t
i
ma
t
i
o
n
U
s
i
n
g
Mat
l
ab
,
”
S
i
g
n
a
l
Im
a
g
e
P
r
o
ces
s
.
In
t
.
J.
,
v
o
l
.
3
,
n
o.
5
,
p
.
5
7
‑
7
2
,
o
c
t
.
2
0
1
2
.
[1
3
]
Mu
h
amma
d
U
.
M.
et
al
.
,
“
Co
m
p
arat
i
v
e
a
n
al
y
s
i
s
b
et
w
een
d
i
rec
t
i
o
n
o
f
arri
v
al
al
g
o
r
i
t
h
ms
,”
I
E
E
E
Co
n
f
e
r
e
n
ce
P
u
b
l
i
c
a
t
i
o
n
,
2
0
1
7
.
[1
4
]
Y
.
K
h
m
o
u
,
S.
Safi,
E
T
M.
Fri
k
e
l
,
“
Co
m
p
arat
i
v
e
S
t
u
d
y
b
et
w
een
Se
v
eral
D
i
rec
t
i
o
n
o
f
A
rr
i
v
a
l
E
s
t
i
mat
i
o
n
Met
h
o
d
s
,
”
Jo
u
r
n
a
l
o
f
Tel
ec
o
m
u
n
i
ca
t
i
o
n
s
a
n
d
i
n
f
o
r
m
a
t
i
o
n
t
ec
h
n
o
l
o
g
y,
p
p
.
4
1
-
4
8
,
2
0
1
4
.
[1
5
]
M.
V
i
b
erg
,
B.
O
t
t
ers
t
en
,
et
T
.
K
a
i
l
a
t
h
,
“
D
e
t
ect
i
o
n
an
d
es
t
i
m
at
i
o
n
i
n
s
e
n
s
o
r
array
s
u
s
i
n
g
w
ei
g
h
t
ed
s
u
b
s
p
ace
fi
t
t
i
n
g
,
”
IE
E
E
Tr
a
n
s
.
S
i
g
n
a
l
P
r
o
ce
s
s
.
,
v
o
l
.
3
9
,
n
o.
1
1
,
p
.
2
4
3
6
‑2
4
4
9
,
1
9
9
1
.
[1
6
]
W
.
W
an
g
,
“
O
v
erv
i
ew
o
f
freq
u
en
c
y
d
i
v
ers
e
array
i
n
ra
d
ar
an
d
n
a
v
i
g
at
i
o
n
ap
p
l
i
ca
t
i
o
n
s
,
”
S
o
n
a
r
Na
v
i
g
.
IE
T
R
a
d
a
r
,
v
o
l
.
1
0
,
n
o.
6
,
p
p
.
1
0
0
1
‑1
0
1
2
,
2
0
1
6
.
[1
7
]
B.
A
.
J
o
h
n
s
o
n
,
Y
.
I.
A
b
ram
o
v
i
ch
,
e
t
X
.
Me
s
t
re,
“
T
h
e
ro
l
e
o
f
s
u
b
s
p
ace
s
w
a
p
i
n
max
i
mu
m
l
i
k
el
i
h
o
o
d
es
t
i
ma
t
i
o
n
p
erfo
rma
n
ce
b
rea
k
d
o
w
n
,
”
i
n
2
0
0
8
IE
E
E
In
t
er
n
a
t
i
o
n
a
l
Co
n
f
er
e
n
ce
o
n
A
c
o
u
s
t
i
cs
,
S
p
eec
h
a
n
d
S
i
g
n
a
l
P
r
o
ce
s
s
i
n
g
,
p
p
.
2
4
6
9
‑2
4
7
2
,
2
0
0
8
.
[1
8
]
Ma
rg
aret
R
o
u
s
e,
“
W
h
at
i
s
8
0
2
.
1
1
?
-
D
ef
i
n
i
t
i
o
n
fro
m
W
h
a
t
Is
.
co
m
,
”
Searc
h
Mo
b
i
l
eCo
m
p
u
t
i
n
g
,
2
0
1
5
.
[
O
n
l
i
n
e].
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
s
earch
m
o
b
i
l
ec
o
mp
u
t
i
n
g
.
t
ec
h
t
ar
g
et
.
c
o
m/
d
ef
i
n
i
t
i
o
n
/
8
0
2
1
1
.
[1
9
]
IE
E
E
8
0
2
.
1
5
.
8
-
2
0
1
7
,
“
IE
E
E
St
an
d
ard
fo
r
W
i
rel
e
s
s
M
ed
i
u
m
A
cces
s
Co
n
t
ro
l
(MA
C)
an
d
Ph
y
s
i
cal
L
ay
er
(P
H
Y
)
Sp
eci
f
i
cat
i
o
n
s
fo
r
Peer
A
w
are
Co
mmu
n
i
ca
t
i
o
n
s
(PA
C
)
,
”
[O
n
l
i
n
e].
A
v
ai
l
a
b
l
e
:
h
t
t
p
s
:
/
/
s
t
an
d
ard
s
.
i
eee.
o
rg
/
s
t
an
d
ard
/
8
0
2
_
1
5
_
8
-
2
0
1
7
.
h
t
ml
.
[2
0
]
R.
B.
M.
A
b
d
el
rah
ma
n
,
A
.
B.
A
.
Mu
s
t
afa,
et
A
.
A
.
O
s
man
,
“
A
Co
mp
ari
s
o
n
b
et
w
een
IE
E
E
8
0
2
.
1
1
a,
b
,
g
,
n
an
d
ac
St
an
d
ard
s
,”
l
i
f
ewi
r
e
i
n
t
er
n
et
,
n
e
t
wo
r
ki
n
g
&
s
ec
u
r
i
t
y
,
2
0
1
9
.
[2
1
]
R.
K
h
an
d
u
ri
&
S.
S.
Rat
t
an
,
“
Perfo
rman
ce
C
o
m
p
ari
s
o
n
A
n
al
y
s
i
s
b
et
w
een
IE
E
E
8
0
2
.
1
1
a/
b
/
g
/
n
St
an
d
ard
s
,
”
In
t
.
J.
Co
m
p
u
t
.
A
p
p
l
.
,
v
o
l
.
7
8
,
n
o.
1
,
p
.
1
3
‑
2
0
,
s
e
p
t
.
2
0
1
3
.
[2
2
]
B.
M.
A
.
M
.
g
rad
u
z
at
e
w
h
o
b
ri
n
g
s
y
ear
s
o
f
t
ech
n
i
cal
ex
p
eri
e
n
ce
t
o
art
i
c
l
es
o
n
SE
O
,
co
mp
u
t
ers
,
&
W
.
N
et
w
o
r
k
i
n
g
,
“
T
h
e
R
o
l
e
o
f
8
0
2
.
1
1
b
i
n
E
s
t
a
b
l
i
s
h
i
n
g
W
i
-
Fi
a
s
a
H
o
me
N
et
w
o
r
k
T
ec
h
n
o
l
o
g
y
”
Li
f
ewi
r
e
,
2
0
1
8
.
[O
n
l
i
n
e].
A
v
ai
l
a
b
l
e
:
h
t
t
p
s
:
/
/
w
w
w
.
l
i
fe
w
i
re.
c
o
m/
h
i
s
t
o
ry
-
of
-
w
i
rel
e
s
s
-
s
t
a
n
d
ar
d
-
8
0
2
-
1
1
b
-
8
1
6
5
5
5
.
[2
3
]
K
.
Sh
aw
,
“
8
0
2
.
1
1
:
W
i
-
F
i
s
t
an
d
ard
s
a
n
d
s
p
ee
d
s
e
x
p
l
ai
n
ed
,
”
Net
w
o
r
k
W
o
r
l
d
,
2
0
1
8
.
[O
n
l
i
n
e
].
A
v
a
i
l
a
b
l
e
:
ht
t
p
s
:
/
/
w
w
w
.
n
et
w
o
r
k
w
o
rl
d
.
co
m
/
art
i
cl
e
/
3
2
3
8
6
6
4
/
8
0
2
1
1
-
wi
-
fi
-
s
t
an
d
ard
s
-
an
d
-
s
p
eed
s
-
ex
p
l
a
i
n
e
d
.
h
t
ml
.
[2
4
]
A
BI
res
earch
f
o
r
v
i
s
i
o
n
ar
i
es
,
“
Para
l
l
e
l
s
Bet
w
een
W
i
G
i
g
an
d
5
G
,
”
2
0
1
9
.
h
t
t
p
s
:
/
/
w
w
w
.
a
b
i
re
s
earch
.
c
o
m/
b
l
o
g
s
/
2
0
1
4
/
0
1
/
2
2
/
p
ara
l
l
e
l
s
-
b
et
w
een
-
w
i
g
i
g
-
an
d
-
5
g
/
.
[2
5
]
K
.
N
g
u
y
en
,
M.
G
o
l
am
K
i
b
ri
a,
K
.
Is
h
i
z
u
,
et
F.
K
o
j
i
ma
,
“
Perfo
rman
ce
E
v
al
u
at
i
o
n
o
f
IE
E
E
8
0
2
.
1
1
ad
i
n
E
v
o
l
v
i
n
g
Wi
-
F
i
N
et
w
o
r
k
s
,”
A
d
va
n
ced
W
i
r
e
l
es
s
Tech
n
o
l
o
g
y
f
o
r
U
l
t
r
a
h
i
g
h
D
a
t
a
R
a
t
e
Co
m
m
u
n
i
ca
t
i
o
n
,
v
o
l
.
2
0
1
9
,
p
p
.
1
-
1
1
,
2
0
1
9
.
h
t
t
p
s
:
/
/
w
w
w
.
h
i
n
d
aw
i
.
co
m
/
j
o
u
r
n
al
s
/
w
cmc/
2
0
1
9
/
4
0
8
9
3
6
5
/
.
[2
6
]
B.
M.
A
.
M.
g
rad
u
at
e
w
h
o
b
ri
n
g
s
y
ear
s
o
f
t
ec
h
n
i
cal
ex
p
eri
e
n
ce
t
o
ar
t
i
c
l
es
o
n
S
E
O
,
c
o
mp
u
t
er
s
,
&
W
.
N
e
t
w
o
rk
i
n
g
,
“
Is
T
h
ere
a
L
i
m
i
t
t
o
H
o
w
Man
y
D
e
v
i
ce
s
Can
Co
n
n
e
ct
t
o
a
W
i
-
Fi
N
et
w
o
r
k
?
,
”
Li
f
ew
i
r
e
i
n
t
e
r
n
et
,
n
e
t
wo
r
k
i
n
g
&
s
ecu
r
i
t
y,
2
0
1
9
.
[
O
n
l
i
n
e].
A
v
a
i
l
a
b
l
e:
ht
t
p
s
:
/
/
w
w
w
.
l
i
fe
w
i
re.
co
m
/
h
o
w
-
ma
n
y
-
d
ev
i
ces
-
can
-
s
h
are
-
a
-
w
i
f
i
-
n
e
t
w
o
rk
-
8
1
8
2
9
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.