I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
, pp.
4281
~
4290
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
13
.i
4
.pp
4281
-
4290
4281
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
S
i
gn
al
l
i
n
g
ove
r
h
e
ad
m
i
n
i
m
i
z
at
i
on
aw
ar
e
h
an
d
ove
r
e
xe
c
u
t
i
on
u
si
n
g e
n
se
m
b
l
e
l
e
ar
n
i
n
g i
n
n
e
xt
ge
n
e
r
at
i
on
w
i
r
e
l
e
ss
n
e
t
w
o
r
k
s
B
h
avan
a S
r
in
iv
as
1
,
N
ad
ig
V
ij
aye
n
d
r
a
U
m
a R
e
d
d
y
2
1
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
oni
c
s
a
nd C
om
m
uni
c
a
t
i
on E
ngi
ne
e
r
i
ng
, N
e
w
H
or
i
z
on C
ol
l
e
ge
of
E
ngi
ne
e
r
i
ng, B
a
nga
l
or
e
, I
ndi
a
2
D
e
pa
r
t
m
e
nt
of
A
I
a
nd
M
L
, N
e
w
H
or
i
z
on C
ol
l
e
ge
of
E
ngi
ne
e
r
i
ng, B
a
nga
l
or
e
, I
ndi
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
N
ov 22, 2023
R
e
vi
s
e
d
M
a
r
7, 2024
A
c
c
e
pt
e
d
A
pr
18, 2024
Upcoming
smart
intelligent
h
eterogeneous
wireless
networks
(HWNs
)
and
their
uses
can
greatly
benefit
from
the
merging
of
long
-
term
evoluti
on
(LTE)
sub
-
6
GHz
along
with
millimeter
wave
(mmWave
)
frequenci
es
by
boosting
the
coverage,
bandwidth,
reliability,
seamless
connectivity,
and
high
quality
of
service
(QoS).
Nevertheless,
because
of
the
inabili
ty
of
directed
waves
in
terms
of
coverage,
it
is
difficult
to
locate
the
appro
priate
mmWave
remote
radio
units
(RRUs).
Therefore,
it
is
crucial
to
lessen
the
burden
of
the
handover
signaling
processes.
In
meeting
res
earch
requirements
this
paper
presents
signalin
g
overhead
minimi
zation
a
ware
handover
execution
(SOMAHE)
model.
The
SOMAHE
model
first
int
roduces
a
novel
handover
mechanism
between
LTE
and
mmWave
is
presented
in
this
research,
followed
by
a
machin
e
learning
(ML)
-
based
autonom
ous
handover
execution
techniqu
e.
To
estimat
e
the
han
dover
success
rate,
the
model
introduces
a
feature
ensemble
learning
(
FEL)
model
built
using
XGBoost
(XGB)
model
that
makes
use
of
sam
pling
windows
channel
data.
To
conclude,
combining
FEL
into
the
SOM
AHE
model
reduces
signaling
overhead
while
simultaneously
increasin
g
the
handover
success
-
rate.
Experiment
results
with
varying
mobile
termi
nals,
demonstrate
that
the
SOMAHE
model
significantly
outperforms
the
existin
g
standard
deep
q
-
networks
(DQN)
-
based
handover
-
exec
ution
method.
K
e
y
w
o
r
d
s
:
D
e
e
p l
e
a
r
ni
ng
H
e
te
r
oge
ne
ous
w
ir
e
le
s
s
ne
twor
ks
, ha
ndove
r
M
a
c
hi
ne
l
e
a
r
ni
ng
Q
ua
li
ty
-
of
-
s
e
r
vi
c
e
R
e
in
f
or
c
e
m
e
nt
l
e
a
r
ni
ng
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
B
ha
va
na
S
r
in
iv
a
s
D
e
pa
r
tm
e
nt
of
E
le
c
tr
oni
c
s
a
nd C
om
m
uni
c
a
ti
on E
ngi
ne
e
r
in
g
, N
e
w
H
or
iz
on C
ol
le
ge
of
E
ngi
ne
e
r
in
g
B
a
nga
lo
r
e
, I
ndi
a
E
m
a
il
:
bha
va
na
s
_
12@
r
e
di
f
f
m
a
il
.c
om
1.
I
N
T
R
O
D
U
C
T
I
O
N
A
s
of
th
e
ye
a
r
2020,
it
ha
s
be
e
n
obs
e
r
ve
d
th
a
t
th
e
gl
oba
l
la
nds
c
a
pe
is
in
tr
ic
a
te
ly
in
te
r
twi
ne
d
w
it
h
a
s
ta
gge
r
in
g
num
be
r
of
25
bi
ll
io
n
e
m
be
dde
d
de
vi
c
e
s
,
w
hi
c
h
a
r
e
s
e
a
m
le
s
s
ly
c
onne
c
te
d
to
a
n
e
qui
va
le
nt
num
be
r
of
in
di
vi
dua
ls
.
T
hi
s
in
te
r
c
onne
c
te
dne
s
s
ha
s
r
e
s
ul
te
d
in
th
e
ge
ne
r
a
ti
on
of
a
n
i
m
m
e
ns
e
vol
um
e
of
da
ta
,
e
s
ti
m
a
te
d
to
be
a
ppr
oxi
m
a
te
ly
40
tr
il
li
on
gi
ga
byt
e
s
th
e
s
e
s
t
a
ti
s
ti
c
s
,
obt
a
in
e
d
f
r
om
r
e
put
a
bl
e
s
our
c
e
s
[
1]
,
hi
ghl
ig
ht
th
e
s
ig
ni
f
ic
a
nt
r
ol
e
pl
a
ye
d
by
bot
h
e
m
be
dde
d
de
vi
c
e
s
a
nd
in
te
r
c
onne
c
te
d
in
di
vi
dua
ls
in
to
da
y'
s
w
or
ld
.
T
he
w
ir
e
le
s
s
in
te
r
ne
t
of
th
in
gs
(
I
oT
)
is
im
po
r
ta
nt
in
th
e
c
ont
e
xt
of
f
ut
ur
e
in
te
ll
ig
e
nt
a
ppl
ic
a
ti
ons
w
it
hi
n
th
is
e
nvi
r
onm
e
nt
.
H
ow
e
ve
r
,
th
e
in
te
gr
a
ti
on
of
a
la
r
ge
-
s
c
a
le
de
pl
oym
e
nt
of
m
obi
le
te
r
m
in
a
ls
pr
e
s
e
nt
s
num
e
r
ous
r
e
s
e
a
r
c
h c
om
pl
e
xi
ti
e
s
a
nd obs
ta
c
le
s
t
ha
t
r
e
qui
r
e
c
a
r
e
f
ul
c
ons
id
e
r
a
ti
on [
2]
.
T
he
s
e
c
h
a
ll
e
nge
s
e
nc
om
p
a
s
s
va
r
io
us
a
s
pe
c
ts
in
c
lu
di
ng
c
om
m
uni
c
a
ti
on,
s
to
r
a
ge
c
a
pa
bi
li
ti
e
s
,
a
nd
ne
twor
ki
ng
in
f
r
a
s
tr
uc
tu
r
e
.
I
n
a
ddi
ti
on
to
th
e
c
ons
id
e
r
a
ti
ons
,
s
e
ve
r
a
l
a
ddi
ti
ona
l
is
s
ue
s
de
m
a
nd
a
tt
e
nt
io
n
in
th
e
f
ie
ld
.
T
he
s
e
c
h
a
ll
e
nge
s
in
c
lu
de
but
a
r
e
not
li
m
it
e
d
to
in
te
r
-
ope
r
a
bi
li
ty
,
di
ve
r
s
it
y,
a
s
w
e
ll
a
s
in
f
or
m
a
ti
on
a
nd
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4281
-
4290
4282
m
a
na
ge
m
e
nt
of
de
vi
c
e
s
[
3]
.
I
n
th
e
c
ont
e
xt
of
a
he
te
r
oge
nous
w
ir
e
le
s
s
ne
twor
k
e
nvi
r
onm
e
nt
,
it
is
im
por
ta
nt
to
not
e
th
a
t
th
e
a
va
il
a
bl
e
r
a
di
o
r
e
s
our
c
e
s
a
r
e
in
he
r
e
nt
ly
c
r
ow
de
d.
C
ons
e
que
nt
ly
,
m
a
na
gi
ng
th
e
u
s
e
of
th
e
s
e
r
e
s
our
c
e
s
pos
e
s
a
s
ig
ni
f
ic
a
nt
c
ha
ll
e
nge
,
pa
r
ti
c
ul
a
r
ly
in
s
c
e
na
r
i
os
in
vol
vi
ng
de
ns
e
ne
twor
ks
w
he
r
e
num
e
r
ous
s
m
a
r
t
de
vi
c
e
s
c
ont
e
nd f
or
t
he
s
e
r
e
s
our
c
e
s
. I
n t
he
c
ont
e
xt
of
t
h
is
ne
twor
k, i
ts
e
f
f
ic
ie
nc
y i
s
c
ont
in
ge
nt
upon
t
he
e
f
f
ic
ie
nt
a
ll
oc
a
ti
on
of
r
e
s
our
c
e
s
,
s
pe
c
if
ic
a
ll
y
or
th
ogona
l
c
od
e
s
,
f
r
e
que
nc
ie
s
,
a
s
w
e
ll
a
s
ti
m
e
s
lo
ts
,
in
a
n
e
vol
vi
ng
m
a
nne
r
.
I
n
a
ddi
ti
on,
it
is
im
po
r
ta
nt
to
in
ve
s
ti
ga
te
how
va
r
ia
ti
ons
r
e
la
te
d
to
s
ta
bi
li
ty
a
nd
th
e
m
a
na
ge
m
e
nt
of
pe
a
k
tr
a
f
f
ic
a
r
e
a
ddr
e
s
s
e
d
to
f
ul
f
il
l
th
e
qu
a
li
ty
of
s
e
r
vi
c
e
(
Q
oS
)
r
e
qui
r
e
m
e
nt
s
of
u
s
e
r
s
.
T
he
pr
ol
if
e
r
a
ti
on
of
ne
w
I
oT
-
ba
s
e
d
a
ppl
ic
a
ti
ons
ne
c
e
s
s
it
a
te
s
t
he
ne
e
d
f
or
lo
w
-
la
te
nc
y,
hi
gh
ba
ndw
id
th
,
th
r
oughput, and s
pe
c
tr
a
l
e
f
f
ic
ie
nc
y, a
s
w
e
ll
a
s
c
on
s
id
e
r
a
ti
on of
us
e
r
’
s
c
ont
e
xt
[
4]
.
T
he
ut
il
iz
a
ti
on
of
he
te
r
oge
ne
ous
w
ir
e
le
s
s
n
e
twor
ks
(
H
W
N
s
)
ha
s
a
tt
r
a
c
te
d
s
ig
ni
f
ic
a
nt
a
tt
e
nt
io
n
w
it
hi
n
th
e
r
e
s
e
a
r
c
h
c
om
m
uni
ty
a
nd
m
ul
ti
pl
e
s
e
c
to
r
s
du
e
to
it
s
a
bi
li
ty
to
e
f
f
e
c
ti
ve
ly
a
ddr
e
s
s
ha
ndl
in
g
r
e
s
our
c
e
c
ha
ll
e
nge
s
[
5]
.
R
a
di
o
a
c
c
e
s
s
te
c
hnol
ogy
(
R
A
T
)
s
e
le
c
ti
on
a
n
d
r
e
s
our
c
e
a
ll
oc
a
ti
on
is
a
c
r
uc
ia
l
a
s
pe
c
t
th
a
t
gr
e
a
tl
y
in
f
lu
e
nc
e
s
th
e
e
f
f
ic
ie
nc
y
of
th
e
two
ty
pe
s
of
ne
twor
k
-
le
ve
l
a
lo
ng
w
it
h
us
e
r
-
pe
r
c
e
iv
e
d
e
xpe
r
ie
nc
e
s
[
6]
.
T
hi
s
i
s
pr
im
a
r
il
y
due
to
th
e
in
he
r
e
nt
tr
a
de
of
f
in
vol
ve
d
in
th
e
ha
ndove
r
ope
r
a
ti
on,
a
ll
oc
a
ti
on,
a
nd
c
ol
le
c
ti
ve
us
e
of
r
e
s
our
c
e
s
.
R
e
s
our
c
e
s
h
a
r
in
g
is
a
pr
a
c
ti
c
e
th
a
t
ha
s
be
e
n
obs
e
r
ve
d
to
e
nh
a
nc
e
th
e
e
f
f
e
c
ti
ve
ne
s
s
of
ba
ndw
id
th
ut
il
iz
a
ti
on. Ne
ve
r
th
e
le
s
s
, i
t
is
i
m
por
ta
nt
t
o
not
e
t
ha
t
th
is
pr
a
c
ti
c
e
c
a
n a
ls
o l
e
a
d t
o t
he
oc
c
ur
r
e
nc
e
of
di
s
r
upt
io
n be
twe
e
n a
dj
a
c
e
nt
ne
twor
k
s
. O
n t
he
c
ont
r
a
r
y, di
vi
di
ng r
e
s
our
c
e
s
s
e
r
ve
s
t
o de
c
r
e
a
s
e
u
s
e
of
r
e
s
our
c
e
s
by
e
f
f
e
c
ti
ve
ly
r
e
m
ovi
ng
a
ny
in
te
r
f
e
r
e
nc
e
th
a
t
m
a
y
a
r
is
e
f
r
om
a
dj
a
c
e
nt
ne
twor
ks
[
7]
.
I
t
is
a
nt
ic
ip
a
te
d
th
a
t
f
ut
ur
e
H
W
N
s
w
il
l
ope
r
a
te
a
lo
ngs
id
e
v
a
r
io
us
ne
twor
k
s
th
a
t
ut
il
iz
e
non
-
ove
r
la
ppi
ng
ba
nd
s
of
f
r
e
que
nc
i
e
s
. T
hi
s
c
oe
xi
s
te
nc
e
w
il
l
e
n
a
bl
e
ot
he
r
w
ir
e
le
s
s
a
c
c
e
s
s
s
y
s
te
m
s
to
ut
il
iz
e
a
s
in
gl
e
s
pe
c
tr
um
-
ba
nd
w
it
hout
in
te
r
f
e
r
e
nc
e
.
C
ons
id
e
r
in
g
f
ut
ur
e
de
ve
lo
pm
e
nt
s
in
H
W
N
s
,
it
is
a
nt
ic
ip
a
te
d
th
a
t
a
ha
r
m
oni
ous
c
oe
xi
s
te
nc
e
of
s
ha
r
in
g
r
e
s
our
c
e
s
a
nd
s
pl
it
ti
ng
w
il
l
pr
e
va
il
.
T
he
r
e
f
or
e
,
in
th
e
c
a
s
e
o
f
ge
ne
r
a
l
H
W
N
s
de
pi
c
te
d
in
F
ig
ur
e
1,
it
is
im
pe
r
a
ti
ve
t
o i
nve
s
ti
ga
te
a
nd de
ve
lo
p e
f
f
e
c
ti
ve
h
a
ndove
r
a
nd r
e
s
our
c
e
a
ll
oc
a
ti
on t
e
c
hni
que
s
[
8]
.
F
ig
ur
e
1. A
r
c
hi
te
c
tu
r
e
of
he
te
r
oge
nous
w
ir
e
le
s
s
ne
twor
ks
[
5]
T
he
r
e
s
e
a
r
c
h
w
or
k
is
a
im
e
d
a
t
de
v
e
lo
pi
ng
a
n
e
f
f
e
c
ti
ve
h
a
ndove
r
e
xe
c
ut
io
n
a
lg
or
it
hm
na
m
e
ly
s
ig
na
li
ng
ove
r
he
a
d
m
in
im
iz
a
ti
on
a
w
a
r
e
ha
ndove
r
e
xe
c
ut
io
n
(
S
O
M
A
H
E
)
;
r
e
c
e
nt
ly
,
m
a
c
hi
ne
le
a
r
ni
ng
(
M
L
)
[
7]
‒
[
9]
,
r
e
in
f
o
r
c
e
m
e
nt
le
a
r
ni
ng
(
R
L
)
[
10]
,
[
11]
,
a
nd
de
e
p
le
a
r
ni
ng
(
D
L
)
[
12
]
ha
ve
be
e
n
us
e
d
f
o
r
ha
ndove
r
e
xe
c
ut
io
n
in
H
W
N
s
w
it
h
good
e
f
f
e
c
ts
.
H
ow
e
ve
r
,
th
e
c
ur
r
e
nt
m
e
th
od
is
ti
m
e
c
ons
um
in
g
a
nd
e
xhi
bi
ts
hi
ghe
r
s
ig
na
li
ng
ove
r
he
a
d.
I
n
a
ddr
e
s
s
in
g
th
e
r
e
s
e
a
r
c
h
w
or
k
in
tr
oduc
e
f
e
a
tu
r
e
e
ns
e
m
bl
e
le
a
r
ni
ng
(
F
E
L
)
us
in
g
X
G
B
in
to
S
O
M
A
H
E
to
r
e
duc
e
s
ig
na
li
ng
ov
e
r
he
a
d
a
nd
a
c
hi
e
v
e
a
ut
onomous
ha
ndove
r
e
x
e
c
ut
io
n.
O
nc
e
us
e
r
s
a
r
e
ha
ndove
r
to
ne
w
ne
twor
k
it
c
a
u
s
e
s
in
te
r
f
e
r
e
nc
e
;
in
a
ddr
e
s
s
in
g
t
he
pr
opos
e
d
w
or
k
in
tr
oduc
e
s
a
nove
l
r
e
s
our
c
e
a
ll
oc
a
ti
on t
o ne
w
ly
ha
ndof
f
us
e
r
s
. T
h
e
s
ig
ni
f
ic
a
nc
e
of
S
O
M
A
H
E
i
s
de
s
c
r
ib
e
d a
s
gi
ve
n h
e
r
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
Si
gnal
li
ng ov
e
r
he
ad m
in
imi
z
at
io
n aw
a
r
e
handov
e
r
e
x
e
c
ut
io
n us
in
g e
ns
e
m
bl
e
…
(
B
hav
ana Sr
in
iv
a
s
)
4283
−
T
he
w
or
k i
nt
r
oduc
e
s
a
nove
l
a
ut
onomous
ha
ndove
r
e
xe
c
ut
io
n m
ode
l
th
a
t
r
e
duc
e
s
s
ig
na
li
ng ove
r
he
a
d.
−
T
he
m
ode
l
in
tr
oduc
e
s
F
E
L
us
in
g X
G
B
f
or
be
tt
e
r
ha
ndove
r
de
c
is
io
n
-
m
a
ki
ng.
−
T
he
S
O
M
A
H
E
i
m
pr
ove
s
s
y
s
te
m
t
hr
oughput, r
e
duc
e
s
ha
ndov
e
r
f
a
il
ur
e
, a
nd be
tt
e
r
e
ne
r
gy e
f
f
ic
ie
nc
y.
T
he
m
a
nus
c
r
ip
t
or
ga
ni
z
a
ti
on
is
di
s
c
u
s
s
e
d
a
s
f
ol
lo
w
s
:
s
e
c
ti
on
2
,
di
s
c
us
s
e
s
c
ur
r
e
nt
s
ta
te
-
of
-
th
e
-
a
r
t
m
e
th
ods
of
de
s
ig
ni
ng
e
f
f
e
c
ti
ve
ha
ndove
r
e
xe
c
ut
io
n
a
nd
r
e
s
o
ur
c
e
a
ll
oc
a
ti
on
m
ode
l
f
or
H
W
N
s
.
S
e
c
ti
on
3
di
s
c
us
s
e
s
th
e
w
or
ki
ng
of
pr
opos
e
d
m
e
th
odol
ogy
in
de
s
ig
ni
ng
e
f
f
e
c
ti
ve
ha
ndove
r
e
xe
c
ut
io
n
w
it
h
e
nha
nc
e
d
r
e
s
our
c
e
ut
il
iz
a
ti
on.
S
e
c
ti
on
4
di
s
c
us
s
e
s
th
e
e
xpe
r
im
e
nt
a
nd
r
e
s
ul
t
of
pr
opos
e
d
m
e
th
odol
ogy
ove
r
e
xi
s
ti
ng
m
e
th
odol
ogy;
s
e
c
ti
on
5
,
di
s
c
us
s
e
s
th
e
r
e
s
e
a
r
c
h
nove
lt
y
a
nd
s
ig
ni
f
ic
a
nc
e
a
nd
f
ut
ur
e
r
e
s
e
a
r
c
h
di
r
e
c
ti
on
to
e
nha
nc
e
t
he
m
ode
l.
2.
L
I
T
E
R
A
T
U
R
E
S
U
R
V
E
Y
T
hi
s
s
e
c
ti
on
pr
e
s
e
nt
s
a
r
e
vi
e
w
of
r
e
c
e
nt
r
e
s
e
a
r
c
h
pe
r
ta
in
in
g
to
th
e
de
ve
lo
pm
e
nt
of
a
n
e
f
f
ic
ie
nt
ne
twor
k
s
e
le
c
ti
on
a
nd
r
e
s
our
c
e
s
a
ll
oc
a
ti
on
te
c
hni
que
f
or
H
W
N
s
.
I
n
a
pr
e
vi
ous
r
e
s
e
a
r
c
h
in
ve
s
ti
ga
ti
on
[
13]
,
a
n
a
ppr
oa
c
h f
or
ha
ndl
in
g r
e
s
our
c
e
s
w
a
s
de
v
e
lo
pe
d. T
hi
s
a
ppr
oa
c
h
r
e
li
e
s
on e
xa
c
t
m
e
a
s
ur
e
m
e
nt
s
, but
i
t
s
houl
d be
not
e
d
th
a
t
th
e
m
ode
ls
us
e
d
in
th
is
s
tu
dy
r
e
qui
r
e
m
a
nua
l
opt
im
iz
a
ti
on
of
pa
r
a
m
e
te
r
s
a
s
a
n
a
ddi
ti
ona
l
pr
e
r
e
qui
s
it
e
.
R
e
c
e
nt
r
e
s
e
a
r
c
h
h
a
s
pr
im
a
r
il
y
c
onc
e
nt
r
a
te
d
on
a
d
dr
e
s
s
in
g
Q
oS
pa
r
a
m
e
te
r
s
a
t
th
e
ne
twor
k
le
v
e
l,
a
s
e
vi
de
nc
e
d
by
s
tu
di
e
s
[
13]
,
[
14]
.
A
ddi
ti
ona
ll
y,
th
e
r
e
ha
s
be
e
n
a
not
a
bl
e
e
m
pha
s
is
on
s
a
ti
s
f
yi
ng
us
e
r
pr
e
f
e
r
e
nc
e
r
e
qui
r
e
m
e
nt
s
,
a
s
in
di
c
a
te
d
by
in
ve
s
ti
ga
ti
on
s
[
15]
.
N
e
ve
r
th
e
le
s
s
,
it
ha
s
be
e
n
obs
e
r
ve
d
th
a
t
th
e
s
e
m
ode
ls
e
xhi
bi
t
a
s
ig
ni
f
ic
a
nt
e
ne
r
gy
c
on
s
um
pt
io
n
w
he
n
a
tt
e
m
pt
in
g
to
e
nha
nc
e
th
e
ir
s
uc
c
e
s
s
-
r
a
te
.
I
n
a
pr
io
r
s
tu
dy
[
16]
,
th
e
s
ig
ni
f
ic
a
nc
e
of
r
e
us
in
g
s
ys
te
m
r
e
s
our
c
e
s
in
or
de
r
to
e
nha
nc
e
e
ne
r
gy
c
ons
e
r
va
ti
on
w
a
s
de
m
ons
tr
a
te
d.
A
ddi
ti
ona
ll
y,
th
e
s
tu
dy
a
ls
o
hi
ghl
ig
ht
e
d
th
e
s
ig
ni
f
ic
a
nc
e
of
r
e
s
our
c
e
c
ons
um
pt
io
n
a
nd
it
s
im
pa
c
t
on
ove
r
a
ll
s
ys
te
m
pe
r
f
or
m
a
nc
e
.
I
n
th
e
pa
s
t
f
e
w
ye
a
r
s
,
th
e
r
e
ha
s
be
e
n
a
not
a
bl
e
ut
il
iz
a
ti
on
of
M
L
a
ppr
oa
c
he
s
in
th
e
r
e
a
lm
of
a
ut
om
a
ti
c
de
c
i
s
io
n
m
a
ki
ng,
a
s
in
di
c
a
te
d
by
r
e
f
e
r
e
nc
e
[
17]
.
P
uj
a
r
e
t
al
.
[
17]
in
tr
oduc
e
d
a
ha
ndov
e
r
-
e
xe
c
ut
io
n
a
ppr
oa
c
h
f
or
5G
c
e
ll
ul
a
r
ne
twor
ks
th
a
t
r
e
li
e
s
on
th
e
k
-
ne
a
r
e
s
t
ne
ig
hbor
s
(
K
N
N
)
a
lg
or
it
hm
.
T
he
y
de
m
ons
tr
a
te
d
th
a
t
th
is
a
ppr
oa
c
h
e
x
hi
bi
ts
a
n
e
xc
e
ll
e
nt
pe
r
f
or
m
a
nc
e
in
te
r
m
s
of
ta
r
ge
t
-
di
s
c
ove
r
y,
a
nd
f
ur
th
e
r
hi
ghl
ig
ht
e
d
th
e
a
dva
nt
a
ge
s
of
e
m
pl
oyi
ng
M
L
te
c
hni
que
s
f
or
de
c
is
io
n
-
m
a
ki
ng
pr
oc
e
s
s
e
s
in
th
is
c
ont
e
xt
.
T
he
c
ur
r
e
nt
a
ppr
oa
c
h
ha
s
be
e
n
d
e
ve
l
ope
d
w
it
h
a
f
oc
us
on
hom
og
e
ne
ous
n
e
twor
ks
,
but
it
is
im
por
ta
nt
to
not
e
th
a
t
th
e
r
e
r
e
m
a
in
s
a
ga
p
in
und
e
r
s
ta
ndi
ng
how
M
L
a
ppr
oa
c
he
s
c
a
n
be
e
f
f
e
c
ti
ve
ly
a
ppl
ie
d t
o he
te
r
oge
ne
ous
c
e
ll
ul
a
r
ne
twor
ks
[
18]
, [
19]
.
A
c
c
or
di
ng
to
M
a
ji
d
e
t
al
.
[
20]
,
a
nov
e
l
a
ppr
oa
c
h
i
s
pr
opos
e
d
f
or
pr
e
di
c
ti
ng
ha
ndove
r
-
e
xe
c
ut
io
n
us
in
g
a
n
X
G
B
oos
t
-
ba
s
e
d
a
ppr
oa
c
h.
T
he
pr
im
a
r
y
obj
e
c
ti
ve
is
to
m
it
ig
a
te
th
e
ne
e
d
f
or
r
e
gul
a
r
m
e
a
s
ur
e
m
e
nt
upda
te
s
a
nd
im
pr
ove
th
e
ge
ne
r
a
l
e
f
f
ic
ie
nc
y
of
ne
xt
-
ge
ne
r
a
ti
on
ne
twor
ks
.
T
hi
s
a
ppr
oa
c
h
is
im
pl
e
m
e
nt
e
d
a
nd
e
va
lu
a
te
d i
n t
he
c
ont
e
xt
of
t
he
e
xpe
r
im
e
nt
s
. N
e
ve
r
th
e
le
s
s
, i
t
ha
s
be
e
n obs
e
r
ve
d t
ha
t
tr
a
di
ti
ona
l
M
L
a
lg
or
it
hm
s
c
oul
d
be
le
s
s
e
f
f
e
c
ti
ve
w
h
e
n
de
a
li
ng
w
it
h
d
a
ta
s
e
ts
th
a
t
d
e
m
ons
t
r
a
te
im
ba
la
nc
e
d
be
h
a
vi
or
.
I
bor
r
a
e
t
al
.
[
21]
,
it
w
a
s
de
m
on
s
tr
a
te
d
th
a
t
th
e
c
ur
r
e
nt
a
ppr
oa
c
h
in
que
s
ti
on
doe
s
not
e
f
f
e
c
ti
ve
ly
c
ons
id
e
r
bot
h
Q
oS
a
nd
e
ne
r
gy
c
ons
e
r
va
ti
on
a
s
s
im
ul
ta
ne
ous
f
a
c
to
r
s
.
A
s
upe
r
vi
s
e
d
M
L
a
ppr
oa
c
h
w
a
s
e
m
pl
oye
d
to
de
ve
lo
p
a
n
a
ppr
oa
c
h
f
o
r
c
hoos
in
g
a
n
opt
im
a
l
ne
twor
k
c
onf
ig
ur
a
ti
on
th
a
t
s
a
ti
s
f
ie
s
th
e
r
e
a
l
-
ti
m
e
r
e
qui
r
e
m
e
nt
s
of
a
gi
ve
n
a
ppl
ic
a
ti
on.
C
a
o
e
t
al
.
[
22]
,
th
e
s
im
ul
ta
ne
ous
c
ha
ll
e
nge
s
of
f
r
e
que
nt
ly
lo
a
d
-
ba
la
nc
in
g
a
nd
ha
ndove
r
w
e
r
e
in
ve
s
ti
ga
te
d
in
th
e
c
ont
e
xt
of
a
la
r
ge
c
ove
r
a
ge
n
e
twor
k.
A
de
e
p
-
r
e
in
f
or
c
e
m
e
nt
a
lg
or
it
hm
,
s
pe
c
if
ic
a
ll
y
de
e
p
q
-
ne
twor
ks
(
D
Q
N
s
)
,
w
a
s
e
m
pl
oye
d
in
th
e
ir
s
tu
dy
to
a
ddr
e
s
s
th
e
i
s
s
ue
of
s
ig
na
li
ng
-
ove
r
he
a
d
r
e
duc
ti
on
a
nd
th
r
oughput
m
a
xi
m
iz
a
ti
on
w
hi
le
m
in
im
iz
in
g
ha
ndove
r
-
f
a
il
ur
e
s
[
23]
.
I
n
a
r
e
c
e
nt
s
tu
dy
[
24]
,
th
e
obj
e
c
ti
v
e
w
a
s
to
opt
im
iz
e
th
e
num
be
r
of
in
di
vi
dua
ls
in
a
ne
twor
k
w
hi
le
m
in
im
iz
in
g
ha
ndove
r
-
f
a
il
ur
e
s
,
ta
ki
ng
in
to
a
c
c
ount
th
e
m
obi
li
t
y
pa
tt
e
r
ns
of
in
di
vi
dua
ls
in
a
d
e
ns
e
ly
popula
te
d
ne
twor
k.
A
n
a
nt
ic
ip
a
to
r
y
te
c
hni
que
w
a
s
in
tr
oduc
e
d
by
th
e
r
e
s
e
a
r
c
he
r
s
to
de
te
r
m
in
e
th
e
lo
c
a
ti
on
of
a
n
in
di
vi
dua
l
by
le
v
e
r
a
gi
ng
pr
e
vi
ous
pos
it
io
ns
th
r
ough
th
e
ut
il
iz
a
ti
on
of
a
l
e
a
r
ni
ng me
th
od.
T
h
e
c
h
a
ll
e
n
ge
o
f
f
in
di
n
g t
h
e
b
e
s
t
po
s
s
i
bl
e
ou
tc
om
e
i
n
a
g
a
m
e
-
th
e
or
y a
p
pr
o
a
c
h
[
2
5]
,
[
26]
i
s
m
o
d
e
le
d
a
s
a
pr
e
di
c
ta
bl
e
i
s
s
u
e
th
a
t
i
s
k
no
w
n
to
be
N
P
-
h
a
r
d.
T
h
e
s
t
udy
c
on
duc
te
d
by
M
a
g
oa
r
ou
[
27]
a
im
e
d
to
a
c
qu
ir
e
t
he
ge
o
gr
a
ph
ic
p
os
it
i
on
of
u
s
e
r
s
u
s
in
g
a
n
u
ns
up
e
r
v
is
e
d
l
e
a
r
ni
ng
a
p
pr
o
a
c
h.
T
he
pr
im
a
r
y
obj
e
c
ti
ve
of
th
i
s
r
e
s
e
a
r
c
h
w
a
s
to
a
ddr
e
s
s
t
h
e
p
r
a
c
ti
c
a
l
c
ha
ll
e
ng
e
s
a
s
s
o
c
i
a
t
e
d
w
i
th
h
a
nd
ov
e
r
a
s
w
e
ll
a
s
a
ll
oc
a
ti
on
of
r
e
s
our
c
e
s
, w
h
il
e
t
a
k
in
g
in
to
a
c
c
ou
nt
th
e
i
nf
lu
e
n
c
e
of
s
m
a
ll
s
c
a
le
f
a
de
d
a
pp
e
a
r
a
nc
e
.
T
on
g
e
t
al
.
[
28]
,
th
e
ut
i
li
z
a
t
io
n
of
s
of
t
w
a
r
e
de
f
in
e
d
ne
t
w
or
k
(
S
D
N
)
w
a
s
e
xpl
or
e
d
a
s
a
p
ot
e
nt
i
a
l
s
ol
ut
i
on f
or
t
a
c
kl
i
ng
m
obi
li
t
y
a
lo
ng
w
it
h u
s
e
r
a
ll
oc
a
t
io
n
of
r
e
s
o
ur
c
e
s
c
h
a
ll
e
n
ge
s
in
H
W
N
s
.
T
h
e
r
e
s
e
a
r
c
h
e
r
s
in
th
i
s
s
tu
dy
e
m
pl
o
ye
d
a
th
r
e
e
-
p
ha
s
e
a
pp
r
o
a
c
h
f
or
d
e
v
e
l
opi
ng
m
ul
t
ip
a
th
-
ba
s
e
d
c
om
m
uni
c
a
ti
on.
T
h
e
i
ni
ti
a
l
s
te
p i
nv
ol
v
e
d
ut
i
li
z
i
ng
a
s
t
e
a
dy
s
ta
t
e
s
ys
te
m
f
or
pr
e
d
ic
ti
n
g t
he
u
s
e
r
'
s
lo
c
a
t
io
n
.
I
n
th
e
s
e
c
on
d
s
t
a
g
e
,
a
n
a
m
bi
gu
ou
s
a
n
a
l
yt
i
c
a
l
c
e
nt
r
a
li
z
e
d
pr
oc
e
dur
e
w
a
s
e
m
pl
o
ye
d
f
o
r
c
ho
os
in
g
t
he
a
pp
r
opr
ia
te
n
e
t
w
or
k
.
L
a
s
tl
y, w
i
th
i
n
t
he
th
ir
d
s
te
p,
a
m
u
lt
i
-
pa
th
t
r
a
n
s
m
i
s
s
i
on
pr
ot
o
c
o
l
w
a
s
ut
i
li
z
e
d
t
o e
x
e
c
ut
e
t
h
e
pr
o
c
e
s
s
of
ha
nd
ov
e
r
.
T
h
i
s
a
ppr
o
a
c
h
w
a
s
de
s
c
r
ib
e
d
d
e
ta
il
i
n
[
2
9]
.
I
n
th
e
s
tu
dy
c
on
du
c
t
e
d
by
t
he
r
e
s
e
a
r
c
h
e
r
s
,
va
r
i
ou
s
a
t
tr
ib
ut
e
s
w
e
r
e
ut
i
li
z
e
d
f
or
tr
a
i
ni
n
g a
ne
tw
or
k of
ne
ur
o
n
s
in
or
d
e
r
to
f
a
c
i
li
t
a
t
e
t
h
e
e
x
e
c
ut
i
on
of
h
a
nd
ov
e
r
in
t
he
c
on
te
xt
of
H
W
N
s
.
T
h
e
s
e
a
tt
r
i
but
e
s
e
n
c
o
m
p
a
s
s
e
d
th
e
p
a
c
ke
t
lo
s
s
a
nd
bi
t
-
e
r
r
or
-
r
a
t
e
,
s
i
gn
a
l
-
to
-
n
oi
s
e
r
a
ti
o,
s
h
if
ti
ng
s
p
e
e
d,
l
e
a
s
t
de
la
y,
a
nd
hi
g
he
s
t
tr
a
n
s
m
i
s
s
io
n
r
a
te
.
T
h
e
c
u
r
r
e
nt
r
e
s
our
c
e
s
e
l
e
c
ti
o
n
pr
o
c
e
s
s
la
c
k
s
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4281
-
4290
4284
s
y
s
t
e
m
a
t
ic
a
ppr
oa
c
h
,
a
s
it
r
e
l
ie
s
o
n
a
r
a
n
do
m
m
e
t
ho
d
r
a
t
he
r
t
h
a
n
t
a
ki
ng
in
to
a
c
c
ou
nt
th
e
c
o
n
s
um
e
r
'
s
pr
i
or
it
y
r
e
q
ui
r
e
m
e
nt
s
.
T
h
e
c
on
s
id
e
r
a
ti
o
n
o
f
e
f
f
i
c
i
e
nt
lo
a
d
di
s
tr
ib
ut
io
n
in
or
d
e
r
to
m
in
i
m
i
z
e
h
a
nd
ov
e
r
f
a
i
lu
r
e
a
n
d
m
a
x
im
i
z
e
ut
il
i
z
a
ti
on
of
r
e
s
our
c
e
s
h
a
s
n
ot
be
e
n
a
ddr
e
s
s
e
d
i
n
pr
e
vi
ou
s
s
tu
di
e
s
[
5]
, [
8
]
, [
11]
.
T
h
e
p
r
e
s
e
nt
di
s
c
o
ur
s
e
e
nd
e
a
v
or
s
t
o
s
h
e
d
l
i
ght
on
t
he
s
tu
dy'
s
te
c
h
ni
q
ue
e
m
p
lo
y
e
d
i
n
t
a
c
k
li
ng
di
f
f
e
r
e
n
t
s
t
udy
di
f
f
i
c
ul
ti
e
s
.
3.
P
R
O
P
O
S
E
D
M
E
T
H
O
D
O
L
O
G
Y
T
hi
s
s
e
c
ti
on
di
s
c
us
s
e
s
th
e
m
a
th
e
m
a
ti
c
a
l
m
ode
l
of
pr
opos
e
d
S
O
M
A
H
E
a
lg
or
it
hm
.
T
he
m
ode
l
di
s
c
us
s
e
s
th
e
s
ta
nda
r
d
ha
ndove
r
e
x
e
c
ut
io
n
m
ode
l
a
nd
pr
opo
s
e
d
ha
ndove
r
e
x
e
c
ut
io
n
to
r
e
duc
e
s
ig
na
li
ng
ove
r
he
a
d
us
in
g
nove
l
F
E
L
bui
lt
w
it
h
X
G
B
.
F
in
a
ll
y,
a
n
e
f
f
e
c
t
iv
e
r
e
s
our
c
e
s
e
le
c
ti
on
m
ode
l
i
s
in
tr
oduc
e
d
to
m
a
xi
m
iz
e
t
he
r
e
s
our
c
e
ut
il
iz
a
ti
on f
or
ne
w
ly
ha
ndof
f
m
obi
le
t
e
r
m
in
a
ls
.
3.1. S
ys
t
e
m
an
d
r
ad
io
m
od
e
l
A
s
s
um
e
t
he
r
e
e
xi
s
t
s
a
hi
ghl
y
d
e
ns
e
h
e
te
r
og
e
nou
s
-
w
ir
e
le
s
s
c
om
m
uni
c
a
ti
on
n
e
twor
k
(
H
W
C
N
)
.
I
n
th
i
s
s
c
e
na
r
io
,
le
t
r
e
pr
e
s
e
nt
th
e
r
a
d
iu
s
f
or
ne
t
w
or
k
o
ve
r
la
p
a
m
o
ng
th
e
m
il
li
m
e
te
r
w
a
v
e
(
m
m
W
a
ve
)
(
5G
)
a
n
d
lo
ng
-
te
r
m
e
vol
ut
io
n
(
L
T
E
)
.
F
ur
th
e
r
,
it
i
s
c
ons
id
e
r
e
d
th
a
t
m
obi
l
e
t
e
r
m
in
a
l
s
a
r
e
m
obi
l
e
,
a
n
d
e
nv
ir
onm
e
nt
be
i
ng
dyna
m
i
c
;
he
n
c
e
,
due
to
w
hi
c
h,
th
e
H
W
C
N
e
n
vi
r
onm
e
nt
ha
s
to
im
pl
e
m
e
nt
th
e
p
oi
s
s
on
di
s
tr
i
but
io
n
(
P
D
)
.
M
or
e
o
ve
r
,
t
he
r
e
a
r
e
s
ig
ni
f
ic
a
nt
di
f
f
e
r
e
nc
e
s
be
t
w
e
e
n
th
e
te
c
hnol
ogy
a
n
d
f
r
e
qu
e
nc
y
-
ba
n
ds
us
e
d
by
th
e
5G
a
nd
L
T
E
.
W
h
e
n
e
ve
r
a
m
o
bi
le
te
r
m
in
a
l
i
s
c
onn
e
c
t
e
d
to
a
gi
ve
n
c
e
ll
ul
a
r
-
ne
t
w
or
k,
i
t
s
e
nd
s
d
a
ta
a
s
s
oc
i
a
te
d
w
it
h
th
e
gi
ve
n
do
w
n
-
li
nk
r
a
d
io
-
f
r
e
qu
e
nc
y
it
ha
s
d
e
te
c
te
d
to
th
e
r
e
le
va
nt
ba
s
e
s
t
a
ti
on
s
(
B
S
s
)
.
T
h
e
ke
y
di
s
ti
n
c
ti
on
li
e
s
in
th
e
a
bi
li
t
y
of
B
S
to
m
a
ke
de
c
i
s
io
n
s
w
it
h
r
e
s
pe
c
t
to
th
e
r
e
c
on
f
ig
ur
a
ti
on
of
m
e
a
s
ur
e
m
e
nt
-
ga
p
s
in
L
T
E
.
T
hi
s
c
a
p
a
bi
li
ty
is
f
a
c
il
i
ta
te
d
by
th
e
ut
il
iz
a
ti
on
of
M
L
a
ppr
o
a
c
he
s
,
w
hi
c
h
e
na
b
le
th
e
s
e
t
e
c
h
ni
que
s
t
o
r
e
c
ogni
z
e
t
he
pos
s
ib
il
it
y
th
a
t
th
e
m
m
W
a
ve
-
ba
nd
m
a
y
not
pr
ovi
d
e
s
uf
f
i
c
ie
nt
s
ig
na
l
s
tr
e
n
gt
h
t
o
s
u
s
ta
i
n
c
ont
in
uou
s
c
om
m
uni
c
a
ti
on. M
or
e
ove
r
,
it
i
s
im
pe
r
a
ti
ve
to
not
e
th
a
t
a
s
th
e
m
obi
le
t
e
r
m
in
a
l
a
ppr
oa
c
h
e
s
th
e
pe
r
i
phe
r
y
w
it
hi
n
th
e
ne
t
w
or
k,
it
be
c
om
e
s
n
e
c
e
s
s
a
r
y
f
or
s
a
i
d
de
vi
c
e
to
s
e
a
m
le
s
s
ly
tr
a
n
s
it
io
n
it
s
c
onn
e
c
ti
vi
ty
to
a
n
a
lt
e
r
na
t
iv
e
R
A
T
in
or
de
r
to
s
u
s
ta
i
n
uni
nt
e
r
r
upt
e
d
s
e
r
vi
c
e
pr
ovi
s
io
n
.
T
h
e
a
nt
ic
i
pa
t
e
d
num
b
e
r
of
m
obi
l
e
te
r
m
in
a
l
s
w
hi
c
h
c
oul
d
be
e
f
f
e
c
ti
v
e
ly
m
a
n
a
ge
d
w
it
hi
n a
gi
ve
n
un
it
s
pa
c
e
i
s
c
ha
r
a
c
te
r
iz
e
d
by t
he
m
e
th
od
a
l
ong
s
id
e
th
e
l
e
ve
l
of
in
te
n
s
it
y va
r
i
a
bl
e
. T
hi
s
w
or
k di
s
c
us
s
e
s
t
h
e
pr
oc
e
dur
a
l
f
r
a
m
e
w
o
r
k
pe
r
ta
in
in
g t
o m
obi
le
t
e
r
m
in
a
l
de
not
e
d a
s
,
ope
r
a
ti
ng
w
it
hi
n
th
e
c
ont
e
xt
of
c
e
l
lu
la
r
-
c
om
m
uni
c
a
ti
on
e
nv
ir
onm
e
nt
.
T
he
da
ta
pe
r
ta
i
ni
ng
to
m
o
bi
le
te
r
m
in
a
l
i
s
a
c
qui
r
e
d
by m
e
a
n
s
of
th
e
P
D
, c
ha
r
a
c
te
r
i
z
e
d by
a
m
e
a
n va
lu
e
of
, a
s
e
xpr
e
s
s
e
d b
y
(
1)
.
=
2
(
1)
T
h
e
v
a
r
i
a
b
le
r
e
p
r
e
s
e
nt
s
t
he
r
a
di
u
s
o
f
th
e
c
e
l
lu
l
a
r
-
ne
two
r
k.
T
h
e
pr
e
c
i
s
e
l
oc
a
t
io
n
of
t
he
ℎ
m
ob
il
e
te
r
m
in
a
l
i
s
d
e
te
r
m
i
ne
d u
s
i
ng
a
c
o
nt
i
nu
a
ll
y u
ni
f
o
r
m
d
i
s
tr
i
but
io
n
w
it
hi
n t
he
tw
o
-
di
m
e
n
s
io
na
l
s
p
a
c
e
2
,
e
m
p
lo
y
in
g
pol
a
r
-
c
oor
di
na
t
e
s
(
,
)
.
H
e
r
e
,
r
e
pr
e
s
e
nt
s
th
e
r
a
di
a
l
di
s
t
a
n
c
e
f
r
om
t
he
or
i
gi
n
,
w
i
th
v
a
lu
e
s
r
a
ngi
ng
f
r
o
m
0
t
o
. M
e
a
nw
h
il
e
,
r
e
p
r
e
s
e
nt
s
th
e
a
ng
ul
a
r
po
s
it
io
n
,
r
a
ngi
ng
f
r
o
m
0
t
o
2
. T
h
e
in
de
x
d
e
n
ot
e
s
t
h
e
s
p
e
c
if
i
c
m
o
bi
l
e
te
r
m
in
a
l,
w
it
h
ta
k
in
g
v
a
l
ue
s
f
r
om
1
to
, i
.e
.,
0
≤
≤
,
0
≤
≤
2
a
nd
=
1
,
2
,
3
,
…
,
.
I
n
th
is
w
or
k,
de
c
is
io
n
s
c
onc
e
r
ni
ng
ha
ndof
f
s
a
r
e
m
ode
l
e
d
w
it
h
th
e
he
lp
of
hi
s
to
r
ic
a
l
in
f
or
m
a
ti
on
r
e
ga
r
di
ng
th
e
e
f
f
e
c
ti
ve
ne
s
s
r
a
te
of
h
a
ndof
f
s
f
or
e
a
c
h
in
di
vi
dua
l
m
obi
le
te
r
m
in
a
l,
ut
il
iz
in
g
a
n
a
ppr
oa
c
h
c
a
ll
e
d
f
e
a
tu
r
e
e
ns
e
m
bl
e
le
a
r
ni
ng
X
G
B
(
F
E
L
-
X
G
B
)
.
T
o
e
f
f
e
c
ti
ve
ly
a
ppl
y
F
E
L
-
X
G
B
a
lg
or
it
h
m
,
it
is
im
pe
r
a
ti
ve
th
a
t
th
e
dur
a
ti
on
of
th
e
c
ol
le
c
ti
on
s
e
s
s
io
n,
de
not
e
d
a
s
,
doe
s
not
e
x
c
e
e
d
th
e
c
ha
nn
e
l'
s
-
c
ohe
r
e
nc
e
dur
a
ti
on
(
C
D
)
.
T
o
m
a
in
ta
in
th
e
va
li
di
ty
of
th
e
te
c
hni
que
,
it
is
im
pe
r
a
ti
ve
th
a
t
th
e
dur
a
ti
on
of
s
e
s
s
io
n
r
e
m
a
in
s
w
it
hi
n
th
e
li
m
it
s
of
C
D
.
G
iv
e
n
th
a
t
not
a
ll
m
obi
le
te
r
m
in
a
ls
r
e
qui
r
e
ha
ndof
f
f
unc
ti
on
in
g,
th
e
num
be
r
of
s
a
m
pl
e
s
s
iz
e
obt
a
in
e
d
m
us
t
not
e
xc
e
e
d
th
e
s
um
of
a
ll
ha
ndof
f
a
tt
e
m
pt
s
.
E
a
c
h
ti
m
e
a
m
obi
le
te
r
m
in
a
l
is
c
onne
c
te
d
to
a
ne
twor
k
us
in
g
L
T
E
o
r
s
w
it
c
he
d
to
a
d
if
f
e
r
e
nt
L
T
E
ne
twor
k,
a
ha
ndove
r
im
pl
e
m
e
nt
a
ti
on
a
lg
or
i
th
m
w
a
s
r
un
w
it
hi
n t
he
r
e
c
e
iv
in
g ne
twor
k t
o de
te
r
m
in
e
how
t
o be
s
t
ha
ndl
e
t
he
c
onne
c
ti
on.
3.2. Ob
j
e
c
t
iv
e
f
u
n
c
t
io
n
f
or
au
t
om
at
ic
h
an
d
ove
r
ge
n
e
r
at
io
n
T
o
e
f
f
ic
ie
nt
ly
pe
r
f
or
m
ha
ndove
r
ope
r
a
ti
ons
in
de
ns
e
H
W
C
N
e
nvi
r
onm
e
nt
,
th
e
e
xi
s
ti
ng
w
or
k
ha
s
a
dopt
e
d
a
K
N
N
,
X
G
B
,
RL
,
de
e
p
Q
-
le
a
r
ni
ng,
a
nd
DL
t
e
c
hni
que
.
I
t
is
ha
r
d
to
ha
ndl
e
pr
oc
e
s
s
in
g
s
im
ul
ta
ne
ous
ly
a
t
th
e
s
a
m
e
ti
m
e
,
th
ough,
a
nd
w
he
ne
ve
r
th
e
r
e
e
xi
s
ts
da
ta
im
ba
la
nc
e
,
th
e
a
c
c
ur
a
c
y
a
s
s
o
c
ia
te
d
w
it
h
th
e
ha
ndove
r
im
pl
e
m
e
nt
a
ti
on
a
ppr
oa
c
h
i
s
gr
e
a
tl
y
a
f
f
e
c
te
d.
C
ons
e
que
nt
ly
,
th
e
im
pa
c
t
on
th
e
qu
a
li
ty
of
e
xpe
r
ie
nc
e
(
Q
oE
)
f
or
th
e
us
e
r
is
ob
s
e
r
ve
d.
T
hi
s
s
tu
dy
e
m
pl
oys
th
e
F
E
L
-
X
G
B
c
la
s
s
if
ic
a
ti
on
a
lg
or
it
hm
to
pr
e
di
c
t
ha
ndove
r
e
xe
c
ut
io
ns
due
to
it
s
a
bi
li
ty
to
c
a
r
r
y
out
in
di
v
id
ua
l
tr
e
e
s
in
a
di
s
tr
ib
ut
e
d
w
a
y.
T
he
pr
opos
e
d
a
ppr
oa
c
h
hol
ds
pr
om
is
e
f
or
a
ddr
e
s
s
in
g
th
e
r
e
a
l
-
ti
m
e
de
m
a
nds
of
H
W
N
s
th
a
t
op
e
r
a
te
w
it
h
di
s
tr
ib
ut
e
d
B
S
s
.
T
he
te
c
hni
que
e
xhi
bi
ts
th
e
pot
e
nt
ia
l
f
or
s
c
a
la
bi
li
ty
a
s
it
pe
r
ta
in
s
to
in
put
s
iz
e
,
a
ll
ow
in
g
f
or
th
e
a
c
c
om
m
oda
ti
on
of
la
r
ge
r
da
ta
s
e
ts
.
A
ddi
ti
ona
ll
y,
it
de
m
ons
tr
a
te
s
th
e
c
a
pa
bi
li
ty
to
e
f
f
e
c
ti
ve
ly
c
a
pt
ur
e
a
nd
de
ve
lo
p
s
tr
ong
c
or
r
e
la
ti
ons
w
it
hi
n
th
e
c
ol
le
c
ti
on
o
f
f
e
a
tu
r
e
s
.
T
he
F
E
L
-
X
G
B
te
c
hni
que
a
im
s
to
r
e
duc
e
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
Si
gnal
li
ng ov
e
r
he
ad m
in
imi
z
at
io
n aw
a
r
e
handov
e
r
e
x
e
c
ut
io
n us
in
g e
ns
e
m
bl
e
…
(
B
hav
ana Sr
in
iv
a
s
)
4285
obj
e
c
ti
ve
va
r
ia
bl
e
by
in
c
or
por
a
ti
ng
a
di
s
ti
ngui
s
ha
bl
e
nor
m
a
li
z
a
ti
on
te
r
m
a
lo
ng
w
it
h
a
c
onve
x
lo
s
s
-
f
unc
ti
on.
S
pe
c
if
ic
a
ll
y,
it
in
c
lu
de
s
a
n
e
xpr
e
s
s
io
n
‖
‖
1
,
w
hi
c
h
r
e
pr
e
s
e
nt
s
th
e
1
nor
m
of
th
e
va
r
ia
bl
e
w
e
ig
ht
e
d
by
th
e
va
r
ia
bl
e
'
s
pa
r
a
m
e
te
r
,
w
hi
le
th
e
e
xpr
e
s
s
io
n
1
2
‖
‖
2
2
,
w
hi
c
h
r
e
pr
e
s
e
n
ts
ha
lf
th
e
s
qua
r
e
of
th
e
2
nor
m
o
f
w
e
ig
ht
e
d
by
th
e
va
r
ia
bl
e
.
T
he
ov
e
r
a
ll
obj
e
c
ti
ve
w
or
k
is
f
or
m
ul
a
te
d
a
s
‖
‖
1
+
1
2
‖
‖
2
2
+
.
T
he
s
iz
e
of
e
a
c
h
le
a
f
is
r
e
pr
e
s
e
nt
e
d
by
th
e
va
r
ia
bl
e
,
w
hi
le
th
e
ve
c
to
r
c
ons
is
ts
of
th
e
c
or
r
e
s
ponding
w
e
ig
ht
s
of
th
e
le
a
ve
s
in
th
e
s
pe
c
if
ic
gr
a
di
e
nt
-
boos
ti
ng
tr
e
e
.
T
h
e
ut
il
iz
a
ti
on
o
f
a
nor
m
a
li
z
a
ti
on
te
r
m
s
e
r
ve
s
to
m
it
ig
a
te
th
e
is
s
ue
of
ove
r
-
f
it
ti
ng
a
nd
e
nha
nc
e
th
e
opt
im
iz
a
ti
on
of
c
om
put
a
ti
on
c
ha
ll
e
nge
s
.
A
ppl
yi
ng
th
e
s
ubs
e
que
nt
e
qua
ti
on, we
c
a
n c
h
a
r
a
c
te
r
iz
e
t
he
×
0
m
a
tr
ix
t
ha
t
r
e
pr
e
s
e
nt
s
t
he
t
a
r
ge
t
c
ol
le
c
ti
on of
f
e
a
tu
r
e
s
.
=
[
]
=
1
(
2)
T
he
va
r
ia
bl
e
r
e
pr
e
s
e
nt
s
t
he
numbe
r
of
f
e
a
tu
r
e
s
e
ts
t
ha
t
a
r
e
t
a
ke
n i
nt
o c
ons
id
e
r
a
ti
on f
or
t
he
l
e
a
r
ni
n
g
pr
oc
e
s
s
.
O
n
th
e
ot
he
r
h
a
nd,
de
not
e
s
a
m
ul
ti
-
di
m
e
ns
io
na
l
f
e
a
tu
r
e
-
ve
c
to
r
th
a
t
is
g
e
ne
r
a
te
d
f
or
a
s
pe
c
if
ic
ti
m
e
in
s
ta
nc
e
.
T
he
s
upe
r
vi
s
e
d
-
la
be
le
d
-
ve
c
to
r
,
de
not
e
d
a
s
,
is
a
bi
n
a
r
y
va
r
ia
bl
e
th
a
t
r
e
pr
e
s
e
nt
s
th
e
e
xe
c
ut
io
n
of
ha
ndove
r
.
S
pe
c
if
ic
a
ll
y,
th
e
va
lu
e
1
in
di
c
a
te
s
th
a
t
h
a
ndove
r
ha
s
be
e
n
c
a
r
r
ie
d
out
,
w
hi
le
th
e
va
lu
e
0
s
ig
ni
f
ie
s
th
a
t
ha
ndove
r
ha
s
not
b
e
e
n
c
a
r
r
ie
d
out
.
T
hi
s
w
or
k
e
xa
m
in
e
s
a
n
ove
r
a
ll
of
5
f
e
a
tu
r
e
s
.
T
he
s
e
f
e
a
tu
r
e
s
in
c
lu
de
th
e
c
oor
di
na
te
s
of
m
obi
le
te
r
m
in
a
ls
,
th
e
phys
ic
a
l
di
s
ta
n
c
e
b
e
twe
e
n
m
obi
le
te
r
m
in
a
ls
a
lo
ngs
id
e
th
e
B
S
,
th
e
r
e
f
e
r
e
nc
e
s
ym
bol
r
e
c
e
iv
e
d
pow
e
r
(
R
S
R
P
)
in
bot
h
m
m
w
a
ve
a
nd
L
T
E
te
c
hnol
ogi
e
s
,
th
e
m
odi
f
ic
a
ti
on
of
R
S
R
P
m
e
a
s
ur
e
m
e
nt
s
de
pe
ndi
ng
upon
th
e
1
in
te
r
f
a
c
e
,
a
lo
ngs
id
e
t
he
m
odi
f
ic
a
ti
on
of
R
S
R
P
m
e
a
s
ur
e
m
e
nt
s
de
pe
ndi
ng
upon
th
e
2
in
te
r
f
a
c
e
.
T
h
e
c
ol
le
c
ti
on
of
th
e
th
r
e
e
r
e
m
a
in
in
g
pa
r
a
m
e
te
r
s
is
a
bl
e
to
be
a
c
hi
e
v
e
d
by
di
r
e
c
tl
y
a
c
c
e
s
s
in
g
m
obi
le
te
r
m
in
a
ls
.
O
n
th
e
ot
he
r
ha
nd,
th
e
in
it
ia
l
two
f
e
a
tu
r
e
s
a
r
e
a
c
hi
e
va
bl
e
by
a
na
ly
z
in
g
th
e
r
a
di
o
r
e
s
our
c
e
c
ont
r
ol
(
R
R
C
)
m
e
s
s
a
ge
s
.
T
hi
s
is
pos
s
ib
le
by
ut
il
iz
in
g
th
e
de
te
c
te
d
a
r
r
iv
a
l
ti
m
e
va
r
ia
nc
e
or
th
e
gl
oba
l
na
vi
ga
ti
ng
s
tr
uc
tu
r
e
,
a
s
m
e
nt
io
ne
d
in
r
e
f
e
r
e
nc
e
[
26]
.
T
he
opt
im
iz
a
ti
on
of
hype
r
-
pa
r
a
m
e
te
r
s
in
vol
ve
s
th
e
ut
il
iz
a
ti
on
of
th
e
gr
id
s
e
a
r
c
hi
ng
m
e
th
odol
ogy,
w
hi
c
h
ta
ke
s
in
to
a
c
c
ount
k
-
f
ol
d
c
r
os
s
va
li
da
ti
on
(K
-
C
V
)
a
s
a
m
e
a
ns
to
e
nha
nc
e
th
e
a
c
c
ur
a
c
y
of
ha
ndove
r
e
xe
c
ut
io
n.
A
c
c
or
di
ng
to
th
e
f
in
di
ngs
p
r
e
s
e
nt
e
d
in
[
10]
, t
he
c
om
put
a
ti
ona
l
in
vol
ve
d i
n t
he
pr
e
s
e
nt
e
d t
e
c
hni
que
c
a
n
be
a
c
c
ur
a
te
ly
c
ha
r
a
c
te
r
iz
e
d by
(
3)
.
=
(
(
↑
+
l
o
g
)
)
(
3)
T
he
va
r
ia
bl
e
↑
r
e
pr
e
s
e
nt
s
th
e
pa
r
a
m
e
te
r
th
a
t
c
ha
r
a
c
te
r
iz
e
s
th
e
hi
ghe
s
t
pos
s
ib
le
de
pt
h
of
th
e
boos
te
d
tr
e
e
s
.
W
hi
le
th
e
va
r
ia
bl
e
de
not
e
s
th
e
pa
r
a
m
e
te
r
th
a
t
qua
nt
if
ie
s
th
e
to
ta
l
s
iz
e
of
th
e
tr
e
e
s
.
T
he
m
e
a
s
ur
e
m
e
nt
of
c
om
put
a
ti
on
c
om
pl
e
xe
s
is
c
onduc
t
e
d
in
r
e
la
ti
on
w
it
h
th
e
ove
r
a
ll
s
iz
e
of
m
obi
le
te
r
m
in
a
l
e
nc
om
pa
s
s
e
d
th
r
oughout a
c
e
ll
ul
a
r
ne
twor
k, a
s
w
e
ll
a
s
t
he
e
xpr
e
s
s
e
d f
r
e
que
nc
y a
t
w
hi
c
h t
he
s
e
m
e
a
s
ur
e
m
e
nt
s
a
r
e
t
a
ke
n.
3.3. Han
d
ove
r
m
e
c
h
an
is
m
A
s
s
p
e
c
if
i
e
d
in
th
e
L
T
E
pr
ot
o
c
ol
s
pe
c
if
ic
a
ti
on
,
th
e
m
o
bi
le
te
r
m
in
a
l
i
s
r
e
s
pon
s
ib
le
f
or
m
e
a
s
ur
in
g
th
e
R
S
R
P
a
s
s
oc
ia
t
e
d
w
it
h
th
e
L
T
E
n
e
twor
k
.
T
hi
s
R
S
R
P
va
l
ue
i
s
c
om
pa
r
e
d
a
g
a
in
s
t
th
e
ha
ndov
e
r
qua
li
ty
-
s
pe
c
if
ie
r
in
or
de
r
to
in
it
ia
t
e
th
e
R
R
C
X
2
pr
oc
e
s
s
.
T
he
L
T
E
n
e
twor
k
un
d
e
r
goe
s
a
r
e
c
onf
ig
ur
a
ti
on
of
th
e
R
R
C
pr
ot
oc
ol
,
w
hi
c
h
is
pr
im
a
r
il
y
dr
iv
e
n
by
th
e
ut
il
iz
a
ti
on
of
m
e
a
s
ur
e
m
e
nt
-
g
a
ps
.
M
or
e
ov
e
r
,
it
ha
s
be
e
n
obs
e
r
ve
d
th
a
t
m
obi
l
e
te
r
m
in
a
l
s
e
xhi
bi
t
a
n
in
c
r
e
a
s
e
d
R
S
R
P
c
om
pa
r
e
d
to
th
e
ha
n
do
ve
r
qu
a
li
ty
-
s
pe
c
if
ie
r
.
T
h
is
va
r
i
a
ti
on
i
n
s
ig
na
l
s
tr
e
ng
th
pr
om
pt
s
th
e
in
it
i
a
ti
on
of
R
R
C
X
1
pr
o
c
e
s
s
.
W
he
n
th
e
pow
e
r
of
m
m
W
a
v
e
e
xc
e
e
d
s
th
e
pr
e
de
t
e
r
m
in
e
d
qua
li
ty
th
r
e
s
h
ol
d,
m
obi
l
e
te
r
m
in
a
l
i
ni
ti
a
t
e
s
R
R
C
Y
2.
S
ub
s
e
q
ue
nt
ly
,
it
a
r
bi
tr
a
r
y
c
ho
os
e
s
a
s
lo
t
w
it
hi
n
t
he
m
m
W
a
v
e
c
ha
nn
e
l
to
f
a
c
il
it
a
t
e
c
om
m
uni
c
a
ti
on,
r
e
s
ul
ti
ng
i
n
a
n
e
f
f
ic
ie
nt
ha
nd
ove
r
pr
oc
e
s
s
.
T
h
e
c
o
m
pr
e
h
e
n
s
iv
e
de
pi
c
t
io
n
of
th
e
c
onv
e
nt
io
na
l
a
nd
pr
opo
s
e
d
ha
n
dov
e
r
pr
oc
e
dur
e
f
or
e
xe
c
u
ti
on
i
s
il
lu
s
tr
a
t
e
d
in
F
ig
ur
e
s
2 a
nd
3.
F
ig
ur
e
s
2
a
n
d
3
il
lu
s
tr
a
te
s
th
a
t
ha
n
dove
r
a
tt
e
m
pt
s
a
r
e
m
a
d
e
dur
in
g
pha
s
e
X
,
w
hi
le
ha
nd
ove
r
e
xe
c
ut
io
n
oc
c
ur
s
dur
in
g ph
a
s
e
Y
, w
h
e
n t
h
e
L
T
E
ne
t
w
or
k
c
hoo
s
e
s
t
o
pe
r
m
it
ha
n
do
ve
r
.
T
he
ut
il
iz
a
ti
on
of
th
e
c
onve
nt
io
na
l
a
ppr
oa
c
h
m
a
y
r
e
s
ul
t
in
a
m
a
r
gi
na
l
in
c
r
e
a
s
e
in
ove
r
he
a
d.
T
hi
s
w
or
k i
nt
r
oduc
e
s
a
n a
lg
or
it
hm
f
o
r
e
xe
c
ut
in
g ha
ndove
r
s
us
in
g M
L
t
e
c
hni
que
s
. T
he
de
te
r
m
in
a
ti
on of
w
he
th
e
r
t
o
a
c
knowle
dge
m
e
a
s
ur
e
m
e
nt
s
f
r
om
m
obi
le
te
r
m
in
a
l
de
vi
c
e
s
or
ut
il
iz
e
M
L
-
ba
s
e
d
h
a
ndove
r
s
u
c
c
e
s
s
r
a
te
a
c
c
ur
a
c
y
le
ve
l
s
is
a
c
c
om
pl
is
he
d.
T
he
r
e
c
e
iv
e
r
ope
r
a
ti
ng
c
h
a
r
a
c
te
r
is
ti
c
(
R
O
C
)
c
ur
ve
is
a
w
id
e
ly
u
s
e
d
to
ol
in
pr
e
di
c
ti
ve
m
ode
li
ng
r
e
s
e
a
r
c
h.
I
n
th
is
s
tu
dy,
w
e
e
m
pl
oye
d
th
e
R
O
C
c
ur
ve
to
e
s
ti
m
a
te
th
e
li
ke
li
hood
of
a
ha
ndove
r
e
it
he
r
f
a
il
in
g
or
s
uc
c
e
e
di
ng.
T
o
e
n
s
ur
e
r
obus
tn
e
s
s
a
n
d
a
c
c
ur
a
c
y
of
our
pr
e
di
c
ti
ons
,
w
e
e
m
pl
oye
d
a
10
-
f
ol
d
K
-
C
V
a
ppr
oa
c
h.
A
ddi
ti
ona
ll
y,
w
e
e
m
pl
oye
d
a
n
F
E
L
-
X
G
B
,
f
or
c
a
lc
ul
a
ti
ng
th
e
R
O
C
c
ur
ve
.
I
n
th
e
e
ve
nt
th
a
t
th
e
de
te
r
m
in
e
d
pow
e
r
obt
a
in
e
d
by
th
e
L
T
E
s
y
s
te
m
is
f
ound
to
be
be
lo
w
th
e
m
in
im
um
r
e
qui
r
e
m
e
nt
f
or
ha
ndove
r
qua
li
ty
,
a
lo
ng
w
i
th
th
e
a
nt
ic
ip
a
te
d
obt
a
in
e
d
pow
e
r
w
it
hi
n
th
e
m
m
w
a
ve
s
ys
te
m
is
de
te
r
m
in
e
d
t
o
be
gr
e
a
te
r
th
a
n
th
e
e
s
ta
bl
i
s
he
d
ha
ndov
e
r
qua
li
ty
-
s
pe
c
if
ie
r
,
th
e
ha
ndove
r
a
lg
or
it
hm
w
il
l
c
ont
in
ue
w
it
h
it
s
nor
m
a
l
e
xe
c
ut
io
n
m
e
th
od.
I
n
th
e
in
s
ta
nc
e
th
a
t
th
e
pr
oj
e
c
te
d
obt
a
in
e
d
pow
e
r
of
m
m
W
a
ve
is
lo
w
e
r
th
a
n
th
e
ha
ndove
r
qua
li
ty
-
s
pe
c
if
ie
r
,
th
e
L
T
E
ne
twor
k
is
goi
ng
to
bl
oc
k
th
e
m
obi
le
te
r
m
in
a
ls
de
m
a
nd
to
t
r
a
ns
it
io
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4281
-
4290
4286
th
r
ough
L
T
E
to
m
m
W
a
ve
.
T
he
ut
il
iz
a
ti
on
of
m
obi
le
te
r
m
in
a
ls
ha
s
s
how
n
pr
om
is
e
in
m
it
ig
a
ti
ng
pot
e
nt
ia
l
ha
ndove
r
f
a
il
ur
e
s
.
T
h
e
s
ugge
s
te
d
ha
ndov
e
r
e
xe
c
ut
io
n
a
ppr
oa
c
h
f
or
H
W
N
s
m
a
ki
ng
us
e
of
a
F
E
L
-
X
G
B
m
e
th
od i
s
de
pi
c
te
d i
n F
ig
ur
e
3.
F
ig
ur
e
2. C
onve
nt
io
na
l
ha
ndove
r
e
xe
c
ut
io
n a
ppr
oa
c
h
F
ig
ur
e
3. P
r
opos
e
d ha
ndove
r
e
xe
c
ut
io
n a
ppr
oa
c
h
3.4. F
e
at
u
r
e
e
n
s
e
m
b
le
X
G
B
c
la
s
s
i
f
ie
r
m
od
e
l
T
he
X
G
B
c
la
s
s
if
ie
r
a
ppr
oa
c
h
e
m
pl
oys
a
s
e
que
nt
ia
l
le
a
r
ni
ng
s
ys
te
m
th
a
t
ta
k
e
s
in
to
a
c
c
ount
th
e
s
e
c
ond
-
or
de
r
de
r
iv
a
ti
ve
s
.
I
n
or
de
r
to
e
f
f
e
c
ti
ve
ly
a
dj
us
t
to
th
e
a
ut
om
a
te
d
ha
ndove
r
-
e
xe
c
ut
io
n
te
c
hni
que
s
,
it
is
ne
c
e
s
s
a
r
y
to
obt
a
in
th
e
two
-
de
r
iv
a
ti
ve
s
(
in
pa
r
ti
c
ul
a
r
,
th
e
H
e
s
s
ia
n
a
nd
s
ig
m
oi
d)
of
th
e
lo
s
s
-
f
unc
ti
on
w
it
h
th
e
a
im
of
pr
e
di
c
ti
on.
A
s
s
um
e
th
a
t
th
e
va
r
ia
bl
e
is
us
e
d
to
r
e
pr
e
s
e
nt
th
e
s
iz
e
of
th
e
s
a
m
pl
e
in
f
or
m
a
ti
on
be
in
g
c
ons
id
e
r
e
d,
w
hi
le
th
e
va
r
ia
bl
e
is
us
e
d
to
r
e
pr
e
s
e
nt
th
e
s
iz
e
of
th
e
f
e
a
tu
r
e
be
in
g
c
ons
id
e
r
e
d.
T
he
in
it
ia
l
pr
e
di
c
ti
ons
,
de
not
e
d
a
s
,
a
r
e
tr
a
ns
f
or
m
e
d
us
in
g
th
e
s
ig
m
oi
d
e
qua
ti
on
to
obt
a
in
pr
oba
bi
li
ty
-
ba
s
e
d
pr
e
di
c
ti
ons
,
de
not
e
d
a
s
̂
=
(
)
.
T
he
s
ig
m
oi
d
ope
r
a
ti
on,
de
not
e
d
a
s
(
∙
)
,
is
e
m
pl
oye
d
to
d
e
f
in
e
th
e
tr
a
ns
f
or
m
a
ti
on.
A
ddi
ti
ona
ll
y,
th
e
s
ym
bol
s
a
lo
ng
w
it
h
a
r
e
e
m
pl
oye
d
to
de
not
e
two
di
s
ti
nc
t
lo
s
s
-
f
unc
ti
ons
in
th
e
c
ont
e
xt
of
th
is
s
tu
dy.
A
c
c
or
di
ng
to
[
6]
,
[
9]
,
it
c
a
n
be
us
e
d
to
de
r
iv
e
th
e
a
ddi
ti
ve
-
le
a
r
ni
ng
pr
oc
e
s
s
a
s
s
how
n i
n (
4)
.
(
)
=
∑
(
,
(
−
1
)
+
(
)
)
+
(
)
−
1
(
4)
T
he
va
r
ia
bl
e
de
not
e
s
th
e
r
e
pe
ti
ti
on
num
be
r
ut
il
iz
e
d
du
r
in
g
th
e
le
a
r
ni
ng
pr
oc
e
s
s
of
th
e
a
ut
om
a
ti
c
a
ll
y
ha
ndove
r
-
e
xe
c
ut
io
n
a
ppr
oa
c
h.
B
y
e
m
pl
oyi
ng
th
e
T
a
yl
or
'
s
s
e
c
ond
-
or
de
r
e
xpa
ndi
ng
te
c
hni
que
on (
4)
, t
he
r
e
s
ul
ti
ng e
qua
ti
on c
a
n be
de
r
iv
e
d a
s
s
how
n i
n (
5)
. T
he
va
r
ia
bl
e
ℎ
r
e
pr
e
s
e
nt
s
t
he
gr
a
di
e
nt
-
f
unc
ti
on,
th
a
t
w
il
l
be
de
te
r
m
in
e
d
by
ut
il
iz
in
g
th
e
gi
ve
n
(
6)
.
T
h
e
va
r
ia
b
le
r
e
pr
e
s
e
nt
s
h
e
s
s
ia
n
-
f
unc
ti
on,
th
a
t
w
il
l
be
de
te
r
m
in
e
d by uti
li
z
in
g t
he
gi
ve
n (
7)
.
(
)
≈
∑
[
(
,
(
−
1
)
)
+
ℎ
(
)
+
1
2
(
(
)
)
2
]
+
(
)
−
1
(
5)
∝
∑
[
ℎ
(
)
+
1
2
[
]
2
]
+
(
)
−
1
ℎ
=
ℳ
(
6
)
=
2
ℳ
2
(
7
)
T
he
s
c
a
l
a
r
pa
r
a
m
e
te
r
s
ℎ
a
lo
ng
w
it
h
a
r
e
ut
il
iz
e
d
w
it
hi
n
th
e
c
ont
e
xt
of
s
pe
c
if
ic
boos
ti
ng
tr
e
e
s
,
w
hi
c
h
a
r
e
c
om
m
onl
y
e
m
pl
oye
d
f
or
a
dd
r
e
s
s
in
g
bi
na
r
y
is
s
ue
s
.
T
he
X
G
B
oos
t
a
lg
or
it
hm
doe
s
not
in
he
r
e
nt
l
y
pr
ovi
de
a
ut
om
a
te
d
di
f
f
e
r
e
nc
e
s
ope
r
a
ti
on
c
a
pa
bi
li
ti
e
s
.
T
he
r
e
f
or
e
,
th
is
w
or
k
in
tr
oduc
e
s
a
m
a
nua
l
a
ppr
oa
c
h
f
or
c
a
lc
ul
a
ti
ng
th
e
ne
c
e
s
s
a
r
y
de
r
iv
a
ti
ve
s
.
T
he
c
hoi
c
e
of
th
e
s
ig
m
oi
d
-
f
unc
ti
on
to
be
th
e
a
c
ti
va
ti
ng
f
unc
ti
on
dur
in
g
th
e
lo
s
s
-
f
unc
ti
on
c
a
n
b
e
e
xpl
a
in
e
d
by
it
s
d
e
r
iv
a
ti
ve
,
w
hi
c
h
i
s
e
xpl
a
in
e
d
in
(
8)
.
I
n
(
9)
a
ll
ow
s
f
or
th
e
de
r
iv
a
ti
on
of
a
w
e
ig
he
d l
os
s
-
f
unc
ti
on t
ha
t
is
s
p
e
c
if
ic
a
ll
y de
s
ig
ne
d f
or
t
he
pur
pos
e
of
ha
ndove
r
s
e
le
c
ti
on i
n a
n H
W
N
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
Si
gnal
li
ng ov
e
r
he
ad m
in
imi
z
at
io
n aw
a
r
e
handov
e
r
e
x
e
c
ut
io
n us
in
g e
ns
e
m
bl
e
…
(
B
hav
ana Sr
in
iv
a
s
)
4287
̂
=
(
)
=
(
)
(
1
−
(
)
)
=
̂
(
1
−
̂
)
(
8)
ℳ
=
−
∑
(
l
o
g
(
̂
)
+
(
1
−
)
l
o
g
(
1
−
̂
)
)
=
1
(
9)
T
he
s
ym
bol
is
u
s
e
d
to
r
e
pr
e
s
e
nt
th
e
bi
a
s
v
a
r
ia
bl
e
,
w
hi
c
h
is
e
m
pl
oye
d
to
c
ha
r
a
c
te
r
iz
e
im
ba
la
n
c
e
d
f
e
a
tu
r
e
s
.
I
n
th
e
e
ve
nt
th
a
t
th
e
va
lu
e
of
e
xc
e
e
ds
1
,
it
is
vi
ta
l
to
in
c
or
por
a
te
th
e
e
xt
r
a
lo
s
s
va
r
ia
bl
e
.
T
hi
s
is
a
c
hi
e
ve
d
by
c
a
te
gor
iz
in
g
th
e
va
lu
e
of
1
a
s
0
.
I
n
th
e
s
a
m
e
m
a
nne
r
,
w
he
n
th
e
va
lu
e
of
is
gi
ve
n
he
r
e
,
th
e
opt
im
iz
a
ti
on
of
th
e
lo
s
s
-
f
unc
ti
on
s
iz
e
is
f
oc
us
e
d
on
e
ns
ur
in
g
t
ha
t
da
ta
-
f
e
a
tu
r
e
s
th
r
ough
a
n
id
e
nt
if
ie
r
of
0
a
r
e
pr
ope
r
ly
c
a
te
gor
iz
e
d. T
he
e
va
lu
a
ti
on of
de
r
iv
a
ti
ve
of
f
ir
s
t
-
or
de
r
i
s
done
us
in
g t
he
gi
ve
n
as
(
10)
.
ℳ
=
−
(
−
̂
)
(
10)
T
he
a
f
or
e
m
e
nt
io
ne
d
e
qua
ti
on
be
a
r
s
r
e
s
e
m
bl
a
nc
e
to
th
e
e
xpr
e
s
s
io
n
ℳ
te
r
m
c
om
m
onl
y
e
m
pl
oye
d
in
th
e
c
om
put
a
ti
on
of
ge
ne
r
a
l
c
r
os
s
-
e
nt
r
opy
-
lo
s
s
.
A
not
a
bl
e
di
s
ti
nc
ti
on
in
th
is
c
ont
e
xt
is
th
e
ut
il
iz
a
ti
on
of
th
e
va
r
ia
bl
e
to
r
e
gul
a
te
th
e
c
ur
r
e
nt
va
r
ia
bl
e
.
T
he
s
e
c
ond
-
de
r
iv
a
ti
ve
is
c
a
lc
ul
a
te
d
by
ta
ki
ng
a
not
he
r
de
r
iv
a
ti
ve
w
it
h r
e
f
e
r
e
nc
e
t
o
ba
s
e
d on (
10)
i
n t
he
f
ol
lo
w
in
g m
a
nne
r
.
ℳ
2
2
=
−
(
1
−
̂
)
(
̂
)
(
11)
3.5. Re
s
ou
r
c
e
al
lo
c
at
io
n
m
od
e
l
In
(
12)
e
xpl
a
in
s
how
th
e
r
a
ti
o
of
a
ve
r
a
g
e
oc
c
upi
e
d
s
lo
t
s
to
th
e
ove
r
a
ll
s
lo
t
s
w
it
hi
n
th
e
a
s
s
oc
ia
te
d
ne
twor
k
is
c
a
lc
ul
a
te
d
to
de
te
r
m
in
e
us
in
g
th
e
s
t
a
nda
r
d
ut
il
iz
a
t
io
n
of
r
e
s
our
c
e
s
a
ppr
oa
c
h.
w
he
r
e
is
th
e
num
be
r
of
a
c
ti
ve
ly
be
in
g
us
e
d
s
lo
ts
,
a
nd
is
th
e
num
be
r
of
s
lo
ts
a
c
c
e
s
s
ib
le
w
he
n
th
e
s
te
a
dy
-
s
ta
te
li
ke
li
hood
is
c
ons
id
e
r
e
d.
T
he
pr
opos
e
d
S
O
M
A
H
E
m
ode
l
is
a
bl
e
to
r
e
duc
e
th
e
ove
r
a
ll
ti
m
e
to
pe
r
f
or
m
ha
ndove
r
w
it
h
m
in
im
a
l
f
a
il
ur
e
a
nd be
tt
e
r
r
e
s
our
c
e
s
e
le
c
ti
on i
.e
., hi
ghe
r
t
hr
oughput a
s
s
how
n i
n ne
xt
s
e
c
ti
on.
=
∑
∈
(
12)
4.
S
I
M
U
L
A
T
I
O
N
S
T
U
D
Y
I
n
th
i
s
s
e
c
t
io
n
,
w
e
w
i
ll
a
n
a
l
yz
e
th
e
e
xp
e
r
i
m
e
nt
a
l
r
e
s
ul
t
s
of
th
e
ha
nd
ov
e
r
e
f
f
i
c
i
e
n
c
y
of
t
h
e
pr
op
os
e
d
S
O
M
A
H
E
a
n
d
e
xi
s
t
in
g
te
c
h
ni
q
ue
na
m
e
ly
c
e
nt
r
a
li
z
e
d
tr
a
i
ni
n
g
w
it
h
de
c
e
nt
r
a
l
iz
e
d
e
x
e
c
ut
io
n
de
e
p
q
-
n
e
t
w
or
k
(
C
T
D
E
-
D
Q
N
)
[
11
]
.
T
h
e
C
#
-
ba
s
e
d
S
I
M
I
T
S
s
im
ul
a
to
r
h
a
v
e
b
e
e
n
u
s
e
d
to
c
on
du
c
t
s
im
u
la
ti
o
n
a
n
d
s
tu
d
y
th
e
pe
r
f
or
m
a
n
c
e
of
S
O
M
A
H
E
a
n
d
C
T
D
E
-
D
Q
N
.
T
he
p
yt
h
on
X
G
B
-
w
r
a
pp
e
r
f
un
c
t
io
n
h
a
v
e
b
e
e
n
u
s
e
d
in
t
o
S
I
M
I
T
S
s
im
ul
a
to
r
f
or
h
yp
e
r
p
a
r
a
m
e
t
e
r
op
ti
m
iz
a
ti
on
t
o
p
e
r
f
o
r
m
a
ut
ono
m
o
us
ha
ndo
ve
r
e
xe
c
u
ti
o
n. T
h
e
n
e
t
w
or
k
p
a
r
a
m
e
te
r
us
e
d
f
or
r
un
ni
n
g
t
he
s
im
u
la
ti
o
n s
t
udy
i
s
g
iv
e
n i
n T
a
b
le
1 l
ik
e
[
1
1]
c
on
s
i
de
r
in
g
r
a
di
o
pr
op
a
g
a
ti
on
m
o
de
l
d
e
f
i
ne
d
in
[
20]
.
A
s
tu
dy
w
a
s
c
a
r
r
i
e
d
out
t
o
a
s
s
e
s
s
t
h
e
p
e
r
f
or
m
a
n
c
e
m
e
tr
ic
s
r
e
l
a
t
e
d
to
h
a
n
dov
e
r
op
e
r
a
ti
on
s
,
s
p
e
c
if
i
c
a
ll
y
th
e
h
a
n
dov
e
r
f
a
il
ur
e
s
,
ov
e
r
a
ll
e
ne
r
gy
-
c
o
n
s
um
pt
i
on
,
a
nd
t
h
r
ou
gh
p
ut
b
y
va
r
yi
ng
th
e
t
im
e
-
s
lo
t
s
s
iz
e
.
T
a
bl
e
1
.
N
e
twor
k pa
r
a
m
e
te
r
N
e
t
w
or
k P
a
r
a
m
e
t
e
r
V
a
l
ue
N
um
be
r
of
B
S
s
(
N
)
9
N
um
be
r
of
M
T
s
(
U
)
30
T
r
a
ns
m
i
t
pow
e
r
of
t
he
m
a
c
r
o B
S
(
P
0
)
43 dB
m
T
r
a
ns
m
i
t
pow
e
r
of
s
m
a
l
l
B
S
s
(
P
1 t
o P
n)
23 dB
m
E
l
e
c
t
r
i
c
a
l
c
i
r
c
ui
t
pow
e
r
of
B
S
s
(
pc
i
r
c
ui
t
)
2 W
B
a
ndw
i
dt
h f
or
e
a
c
h B
S
(
W
)
20 M
H
z
N
e
t
w
or
k a
r
e
a
[
-
0.5, 0.5]
* [
--
0.5, 0.5]
km
2
T
he
r
m
a
l
noi
s
e
pow
e
r
s
pe
c
t
r
a
l
de
ns
i
t
y
-
174 dB
m
/
H
z
N
um
be
r
of
bi
t
s
pe
r
pa
c
ke
t
(
V
)
100 kbi
t
s
P
a
t
h
-
l
os
s
e
xpone
nt
(
y)
3.8
C
or
r
e
l
a
t
i
on c
oe
f
f
i
c
i
e
nt
of
s
m
a
l
l
f
a
di
ng (
p)
0.9
T
hr
e
s
hol
d l
e
ve
l
of
a
va
i
l
a
bl
e
l
i
nk qua
l
i
t
y (
9t
h)
-
140 dB
m
T
i
m
e
s
l
ot
dur
a
t
i
on T
s
1 s
e
c
T
hr
oughput
r
e
ui
r
e
m
e
nt
R
t
h
100
kbps
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4281
-
4290
4288
4.1. Re
s
u
lt
s
an
d
d
is
c
u
s
s
io
n
I
n
th
is
s
e
c
ti
on
th
e
pe
r
f
or
m
a
nc
e
of
pr
opos
e
d
S
O
M
A
H
E
a
nd
e
xi
s
ti
ng
C
T
D
E
-
D
Q
N
is
s
tu
di
e
d
in
te
r
m
s
of
ha
ndove
r
f
a
il
u
r
e
,
e
ne
r
gy
c
ons
um
pt
io
n
a
nd
th
r
oughput
pe
r
f
or
m
a
nc
e
by
va
r
yi
ng
th
e
ti
m
e
-
s
lo
ts
s
iz
e
f
r
o
m
2000
to
8000
s
e
c
onds
[
t]
.
F
ig
ur
e
4
s
how
s
th
e
h
a
ndove
r
f
a
il
ur
e
pe
r
f
or
m
a
nc
e
e
xpe
r
ie
nc
e
d
us
in
g
S
O
M
A
H
E
a
nd
C
T
D
E
-
D
Q
N
.
T
he
r
e
s
ul
ts
de
m
ons
tr
a
te
th
a
t
th
e
S
O
M
A
H
E
out
pe
r
f
or
m
s
C
T
D
E
-
D
Q
N
c
ons
id
e
r
in
g
a
ll
s
c
e
na
r
io
s
in
te
r
m
s
of
ha
ndove
r
f
a
il
ur
e
r
e
duc
ti
on;
th
us
,
a
n
a
ve
r
a
ge
ha
ndove
r
f
a
il
ur
e
r
e
duc
ti
on
of
47.48%
is
e
xpe
r
ie
nc
e
d by S
O
M
A
H
E
i
n c
om
pa
r
is
on w
it
h e
xi
s
ti
ng C
T
D
E
-
D
Q
N
m
ode
l.
F
ig
ur
e
4. H
a
ndove
r
f
a
il
ur
e
pe
r
f
or
m
a
nc
e
F
ig
ur
e
5
s
h
ow
s
th
e
e
ne
r
gy
c
o
ns
u
m
pt
io
n
pe
r
f
or
m
a
nc
e
e
xpe
r
i
e
n
c
e
d
u
s
in
g
S
O
M
A
H
E
a
nd
C
T
D
E
-
D
Q
N
.
T
he
r
e
s
ul
t
s
de
m
ons
tr
a
te
th
a
t
th
e
S
O
M
A
H
E
out
pe
r
f
or
m
s
C
T
D
E
-
DQN
c
o
ns
i
de
r
in
g
a
ll
s
c
e
na
r
i
os
i
n
te
r
m
s
of
e
ne
r
gy
c
o
ns
u
m
pt
io
n
r
e
du
c
ti
on
.
T
hu
s
,
a
n
a
v
e
r
a
g
e
e
ne
r
gy
c
o
ns
u
m
pt
io
n
r
e
duc
ti
on
of
10.
55%
i
s
e
xp
e
r
ie
n
c
e
d
by
S
O
M
A
H
E
in
c
om
p
a
r
is
o
n
w
it
h
e
xi
s
t
in
g
C
T
D
E
-
D
Q
N
m
od
e
l.
F
ig
ur
e
6
s
h
ow
s
th
e
th
r
o
ughpu
t
p
e
r
f
or
m
a
n
c
e
e
xpe
r
i
e
nc
e
d
u
s
in
g
S
O
M
A
H
E
a
nd
C
T
D
E
-
D
Q
N
.
T
h
e
r
e
s
ul
t
s
de
m
on
s
tr
a
t
e
th
a
t
th
e
S
O
M
A
H
E
out
p
e
r
f
or
m
s
C
T
D
E
-
D
Q
N
c
on
s
id
e
r
i
ng
a
l
l
s
c
e
n
a
r
io
s
in
t
e
r
m
s
of
t
hr
oughp
u
t
im
pr
ove
m
e
nt
.
T
h
us
,
a
n
a
ve
r
a
ge
t
hr
oughp
ut
im
pr
ove
m
e
nt
of
26.27%
is
e
xp
e
r
ie
n
c
e
d
by S
O
M
A
H
E
i
n
c
om
p
a
r
is
on
w
it
h
e
xi
s
ti
ng
C
T
D
E
-
D
Q
N
m
od
e
l.
F
ig
ur
e
5. E
ne
r
gy c
ons
um
pt
io
n pe
r
f
or
m
a
nc
e
F
ig
ur
e
6. T
hr
oughput per
f
or
m
a
nc
e
0
10
20
30
40
50
2000
4000
6000
8000
N
u
m
be
r
of
H
a
n
dov
e
r
s
f
a
i
l
e
d
T
im
e
-
s
lot
s
[
t]
H
a
n
dov
e
r
f
a
i
l
u
r
e
C
T
D
E
-
D
Q
N
S
O
M
A
H
E
0
5
10
15
20
2000
4000
6000
8000
E
n
e
r
g
y
c
on
s
um
e
d
(
J
ou
l
e
s
)
T
i
m
e
-
s
l
ot
s
[
t
]
E
n
e
r
g
y
C
on
s
um
pti
o
n
C
T
D
E
-
D
Q
N
S
O
M
A
H
E
0
5
10
15
20
2000
4000
6000
8000
T
h
r
ou
g
h
pu
t
(
M
bps
)
T
i
m
e
-
s
l
ot
[
t
]
T
h
r
ou
g
h
pu
t
S
OM
AH
E
C
T
DE
-
DQN
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
Si
gnal
li
ng ov
e
r
he
ad m
in
imi
z
at
io
n aw
a
r
e
handov
e
r
e
x
e
c
ut
io
n us
in
g e
ns
e
m
bl
e
…
(
B
hav
ana Sr
in
iv
a
s
)
4289
5.
C
O
N
C
L
U
S
I
O
N
T
he
pr
e
s
e
nt
s
tu
dy
in
it
ia
ll
y
c
a
r
r
ie
d
out
a
n
e
xt
e
ns
iv
e
in
ve
s
ti
ga
ti
on
of
di
ve
r
s
e
ha
ndove
r
m
e
c
ha
ni
s
m
s
th
a
t
a
r
e
c
ur
r
e
nt
ly
in
us
e
w
it
hi
n
H
W
N
s
.
B
a
s
e
d
on
th
e
f
in
di
ngs
of
th
e
s
ur
ve
y,
it
is
e
vi
de
nt
th
a
t
th
e
c
ur
r
e
n
t
ha
ndove
r
te
c
hni
que
ha
s
pr
ove
n
to
be
in
e
f
f
e
c
ti
ve
in
a
c
hi
e
vi
ng
a
s
a
ti
s
f
a
c
to
r
y
ba
la
nc
e
a
m
ong
m
e
e
ti
ng
pe
r
f
or
m
a
nc
e
r
e
qui
r
e
m
e
nt
s
a
nd
a
c
hi
e
vi
ng
e
ne
r
gy
r
e
duc
ti
on
obj
e
c
ti
ve
s
.
M
or
e
ov
e
r
,
th
e
c
ur
r
e
nt
te
c
hni
que
gi
ve
s
r
is
e
to
e
n
e
r
gy
-
ove
r
he
a
d
due
to
th
e
pr
e
s
e
nc
e
of
f
ur
th
e
r
s
ig
na
li
n
g
-
ove
r
he
a
d.
T
he
m
ode
li
ng
of
a
n
a
ut
om
a
ti
c
a
ll
y
ha
ndove
r
te
c
hni
que
th
a
t
in
c
or
por
a
te
s
M
L
te
c
hni
que
s
ha
s
b
e
e
n
unde
r
ta
ke
n
in
or
de
r
to
m
it
ig
a
te
th
e
pr
oc
e
s
s
of
s
ig
na
li
ng
ove
r
he
a
d.
T
he
K
N
N
-
ba
s
e
d
h
a
ndove
r
te
c
hni
que
e
x
hi
bi
te
d
a
la
c
k
of
a
c
c
ur
a
c
y
in
it
s
pr
e
di
c
ti
on
c
a
pa
bi
li
ti
e
s
due
to
th
e
te
c
hni
que
'
s
in
a
bi
li
ty
to
e
f
f
e
c
ti
ve
ly
a
ddr
e
s
s
f
e
a
tu
r
e
im
ba
la
nc
e
c
onc
e
r
ns
.
I
n
or
de
r
to
ta
c
kl
e
th
e
s
tu
dy'
s
c
ha
ll
e
nge
s
a
t
ha
nd,
th
is
pa
pe
r
in
tr
oduc
e
s
a
nove
l
ha
ndove
r
a
nd
r
e
s
our
c
e
a
ll
oc
a
ti
on
te
c
hni
que
th
a
t
le
ve
r
a
ge
s
th
e
pow
e
r
of
M
L
,
s
pe
c
if
ic
a
ll
y
th
e
F
E
L
-
X
G
B
a
lg
or
it
hm
.
I
n
th
is
w
or
k,
a
n
F
E
L
-
X
G
B
a
lg
or
it
hm
is
pr
opos
e
d
to
ta
c
kl
e
th
e
is
s
ue
of
b
e
tt
e
r
hype
r
pa
r
a
m
e
te
r
opt
im
iz
a
ti
on
to
r
e
duc
e
ha
ndove
r
f
a
il
ur
e
w
it
h
be
tt
e
r
r
e
s
our
c
e
opt
im
iz
a
ti
on.
T
he
ha
ndove
r
te
c
hni
que
s
ugge
s
te
d
in
th
is
s
tu
dy,
w
hi
c
h
r
e
li
e
s
on
F
E
L
-
X
G
B
,
de
m
ons
tr
a
te
s
s
ig
ni
f
ic
a
nt
ly
im
pr
ove
d
a
c
c
ur
a
c
ie
s
in
t
e
r
m
s
of
ha
ndove
r
f
a
il
ur
e
r
e
duc
ti
on
in
c
ont
r
a
s
t
to
th
e
ha
ndove
r
t
e
c
hni
que
d
e
pe
ndi
ng
on
D
Q
N
.
E
xpe
r
im
e
nt
s
a
r
e
c
a
r
r
ie
d
out
to
a
s
s
e
s
s
th
e
h
a
ndove
r
pe
r
f
or
m
a
nc
e
by
va
r
yi
ng
th
e
ti
m
e
-
s
lo
t
s
iz
e
.
T
h
e
obt
a
in
e
d
out
c
om
e
de
m
ons
tr
a
te
s
th
a
t
th
e
ha
ndov
e
r
f
a
il
ur
e
is
r
e
duc
e
d
by
47.48%
th
r
ough
th
e
ut
il
iz
a
ti
on
of
th
e
F
E
L
-
X
G
B
-
ba
s
e
d
S
O
M
A
H
E
ha
ndove
r
te
c
hni
que
in
c
om
pa
r
is
on
to
th
e
C
T
D
E
-
DQN
-
ba
s
e
d
ha
ndove
r
te
c
hni
que
.
I
n
a
ddi
ti
on,
it
ha
s
be
e
n
obs
e
r
ve
d
th
a
t
th
e
im
pl
e
m
e
nt
a
ti
on
of
a
S
O
M
A
H
E
ha
ndove
r
te
c
hni
que
ba
s
e
d
on
F
E
L
-
X
G
B
de
m
ons
tr
a
te
s
a
s
ig
ni
f
ic
a
nt
r
e
duc
ti
on
in
e
ne
r
gy
c
on
s
um
pt
io
n,
s
pe
c
if
ic
a
ll
y
by
10.55%
,
in
c
om
pa
r
is
on
to
a
ha
ndove
r
te
c
hni
que
de
p
e
nde
nt
on
C
T
D
E
-
D
Q
N
.
T
h
e
F
E
L
-
X
G
B
-
ba
s
e
d
S
O
M
A
H
E
ha
ndove
r
te
c
h
ni
que
ha
s
be
e
n
obs
e
r
ve
d
to
yi
e
ld
a
n
ove
r
a
ll
th
r
oughput
im
pr
ove
m
e
nt
of
26.27%
,
in
c
om
pa
r
is
on
a
ga
in
s
t
th
e
C
T
D
E
-
DQN
-
ba
s
e
d
ha
ndove
r
t
e
c
hni
que
.
T
hi
s
im
pr
ove
m
e
nt
c
a
n
be
a
tt
r
ib
ut
e
d
to
th
e
de
c
r
e
a
s
e
in
f
ur
th
e
r
s
ig
na
li
ng
ove
r
he
a
d
a
nd
be
tt
e
r
r
e
s
our
c
e
opt
im
iz
a
ti
on
s
tr
a
te
gy.
T
he
pr
e
s
e
nt
e
d
S
O
M
A
H
E
a
ppr
oa
c
h
de
m
ons
tr
a
te
s
r
obus
tn
e
s
s
a
nd
s
c
a
la
bi
li
ty
,
r
e
ga
r
dl
e
s
s
of
th
e
va
r
yi
ng
ti
m
e
-
s
lo
ts
s
iz
e
c
ons
id
e
r
in
g
la
r
ge
r
num
be
r
o
f
m
obi
le
te
r
m
in
a
ls
w
or
ki
ng
w
it
hi
n
a
H
W
N
s
.
F
ut
u
r
e
w
or
k
w
oul
d
c
ons
id
e
r
a
s
s
e
s
s
th
e
p
e
r
f
or
m
a
nc
e
of
pr
opos
e
d
m
ode
l
c
o
ns
id
e
r
in
g
m
or
e
di
ve
r
s
e
m
obi
li
ty
s
c
e
n
a
r
io
s
a
nd
r
a
di
o
pr
opa
ga
ti
on
m
ode
ls
;
a
lo
ngs
id
e
m
or
e
e
nha
nc
e
d
r
e
s
our
c
e
opt
im
iz
a
ti
on
s
tr
a
te
gy
s
houl
d
be
de
ve
lo
p
e
d
c
ons
id
e
r
in
g m
or
e
di
ve
r
s
e
s
c
e
na
r
io
t
o m
it
ig
a
te
i
nt
e
r
f
e
r
e
nc
e
e
f
f
e
c
t
unde
r
l
a
r
ge
de
ns
it
y m
obi
le
t
e
r
m
in
a
ls
.
R
E
F
E
R
E
N
C
E
S
[
1]
A
.
H
a
ghr
a
h,
M
.
P
.
A
bdol
l
a
hi
,
H
.
A
z
a
r
ha
va
,
a
nd
J
.
M
.
N
i
ya
,
“
A
s
ur
ve
y
on
t
he
h
a
ndove
r
m
a
na
ge
m
e
nt
i
n
5G
-
N
R
c
e
l
l
ul
a
r
ne
t
w
or
ks
:
a
s
pe
c
t
s
,
a
ppr
oa
c
he
s
a
nd
c
ha
l
l
e
nge
s
,”
E
ur
a
s
i
p
J
ou
r
nal
on
W
i
r
e
l
e
s
s
C
om
m
uni
c
a
t
i
ons
and
N
e
t
w
or
k
i
ng
,
vol
.
2023,
no.
1,
J
un.
2023,
doi
:
10.1186/
s
13638
-
023
-
02261
-
4.
[
2]
Y
.
U
l
l
a
h,
M
.
B
.
R
os
l
e
e
,
S
.
M
.
M
i
t
a
ni
,
S
.
A
.
K
ha
n,
a
nd
M
.
H
.
J
us
oh,
“
A
s
ur
ve
y
on
ha
ndove
r
a
nd
m
obi
l
i
t
y
m
a
na
ge
m
e
nt
i
n
5G
he
t
ne
t
s
:
c
ur
r
e
nt
s
t
a
t
e
, c
ha
l
l
e
nge
s
, a
nd f
ut
ur
e
di
r
e
c
t
i
on
s
,”
Se
ns
o
r
s
, vol
. 23, no. 1
1, 2023, doi
:
10.3390/
s
23115081.
[
3]
M
.
E
m
r
a
n,
V
.
T
ha
ya
na
nt
ha
n,
M
.
U
m
a
i
r
,
I
.
K
ot
ul
i
a
k,
M
.
S
.
Q
ur
e
s
hi
,
a
nd
M
.
B
.
Q
ur
e
s
hi
,
“
T
he
ha
ndove
r
a
nd
pe
r
f
or
m
a
nc
e
a
na
l
ys
i
s
of
l
t
e
ne
t
w
or
k
w
i
t
h
t
r
a
di
t
i
ona
l
a
n
d
S
D
N
a
ppr
oa
c
he
s
,”
W
i
r
e
l
e
s
s
C
om
m
uni
c
at
i
o
ns
and
M
obi
l
e
C
om
put
i
ng
,
vol
.
2022,
pp.
1
–
9,
A
pr
.
2022, doi
:
10.1155/
2022/
7387737.
[
4]
A
.
U
.
R
e
hm
a
n,
M
.
B
.
R
o
s
l
e
e
,
a
nd
T
.
J
.
J
i
a
t
,
“
A
s
ur
ve
y
of
ha
ndov
e
r
m
a
na
ge
m
e
nt
i
n
m
obi
l
e
he
t
ne
t
s
:
c
ur
r
e
nt
c
h
a
l
l
e
nge
s
a
nd
f
ut
ur
e
di
r
e
c
t
i
ons
,”
A
ppl
i
e
d Sc
i
e
nc
e
s
, vol
. 13, no. 5, M
a
r
. 2023, doi
:
10.3390/
a
pp13053367.
[
5]
J
.
T
a
nve
e
r
,
A
.
H
a
i
de
r
,
R
.
A
l
i
,
a
nd
A
.
K
i
m
,
“
A
n
ove
r
vi
e
w
of
r
e
i
n
f
or
c
e
m
e
nt
l
e
a
r
ni
ng
a
l
gor
i
t
hm
s
f
or
ha
ndove
r
m
a
na
ge
m
e
nt
i
n
5G
ul
t
r
a
-
de
ns
e
s
m
a
l
l
c
e
l
l
ne
t
w
or
ks
,”
A
ppl
i
e
d Sc
i
e
nc
e
s
, vol
. 12, no. 1, J
a
n. 2022, doi
:
10.3390/
a
pp12010426.
[
6]
J
.
L
i
m
a
,
A
.
M
e
de
i
r
os
,
E
.
A
gui
a
r
,
V
.
J
r
.,
a
nd
T
.
G
ue
r
r
a
,
“
U
s
e
r
-
l
e
ve
l
ha
ndove
r
de
c
i
s
i
on
m
a
ki
ng
ba
s
e
d
on
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
h
e
s
,”
J
ou
r
nal
of
C
om
m
uni
c
at
i
on and I
nf
or
m
at
i
on Sy
s
t
e
m
s
, vol
. 37, no.
1, pp. 104
–
108, 2022, doi
:
10.14209/
j
c
i
s
.2022.11.
[
7]
M
.
K
.
D
a
houd
a
,
S
.
J
i
n,
a
nd
I
.
J
oe
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
-
ba
s
e
d
s
ol
ut
i
ons
f
or
ha
ndove
r
de
c
i
s
i
on
s
i
n
non
-
t
e
r
r
e
s
t
r
i
a
l
ne
t
w
or
ks
,”
E
l
e
c
t
r
oni
c
s
, vol
. 12, no. 8, A
pr
. 2023, doi
:
10.3390/
e
l
e
c
t
r
oni
c
s
12081759.
[
8]
A
.
P
r
i
ya
nka
,
P
.
G
a
ut
ha
m
a
r
a
ya
t
hi
r
um
a
l
,
a
nd
C
.
C
ha
ndr
a
s
e
ka
r
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
a
l
gor
i
t
hm
s
i
n
pr
oa
c
t
i
ve
de
c
i
s
i
on
m
a
ki
ng
f
or
ha
ndove
r
m
a
na
ge
m
e
nt
f
r
om
5G
&
be
yond
5G
,”
E
gy
pt
i
an
I
nf
or
m
at
i
c
s
J
our
nal
,
vol
.
24,
no.
3,
2023,
doi
:
10.1016/
j
.e
i
j
.2023.100389.
[
9]
E
.
K
i
m
a
nd I
.
J
oe
,
“
H
a
ndove
r
t
r
i
gge
r
i
ng
p
r
e
di
c
t
i
on
w
i
t
h
t
he
t
w
o
-
s
t
e
p
X
G
B
O
O
S
T
e
ns
e
m
bl
e
a
l
gor
i
t
hm
f
or
c
ondi
t
i
ona
l
ha
ndove
r
i
n
non
-
t
e
r
r
e
s
t
r
i
a
l
ne
t
w
or
k
s
,”
E
l
e
c
t
r
oni
c
s
, vol
. 12, no. 16, A
ug. 2023, doi
:
10.3390/
e
l
e
c
t
r
oni
c
s
12163435.
[
10]
S
.
K
hos
r
a
vi
,
H
.
S
.
G
ha
di
kol
a
e
i
,
J
.
Z
a
nde
r
,
a
nd
M
.
P
e
t
r
ova
,
“
R
e
i
nf
or
c
e
m
e
nt
l
e
a
r
ni
ng
-
ba
s
e
d
j
oi
nt
ha
ndove
r
a
nd
be
a
m
t
r
a
c
ki
ng
i
n
m
i
l
l
i
m
e
t
e
r
-
w
a
ve
ne
t
w
or
ks
,”
i
n
I
E
E
E
W
i
r
e
l
e
s
s
C
om
m
uni
c
at
i
ons
and
N
e
t
w
o
r
k
i
ng
C
onf
e
r
e
nc
e
,
W
C
N
C
,
vol
.
2023
-
M
a
r
c
h,
pp.
1
–
6
,
M
a
r
. 2023,
doi
:
10.1109/
W
C
N
C
55385.2023.10118761.
[
11]
Y
.
S
ong,
S
.
H
.
L
i
m
,
a
nd
S
.
W
.
J
e
on,
“
H
a
ndov
e
r
de
c
i
s
i
on
m
a
ki
ng
f
or
de
n
s
e
H
e
t
N
e
t
s
:
a
r
e
i
nf
or
c
e
m
e
nt
l
e
a
r
ni
ng
a
ppr
oa
c
h
,”
I
E
E
E
A
c
c
e
s
s
, vol
. 11, pp. 24737
–
24751, 2023, doi
:
10.1109/
A
C
C
E
S
S
.2023.3254557.
[
12]
Y
.
J
a
ng,
S
.
M
.
R
a
z
a
,
M
.
K
i
m
,
a
nd
H
.
C
hoo,
“
P
r
oa
c
t
i
ve
h
a
ndove
r
de
c
i
s
i
on
f
or
U
A
V
s
w
i
t
h
de
e
p
r
e
i
nf
or
c
e
m
e
nt
l
e
a
r
ni
ng
,”
Se
n
s
or
s
,
vol
. 22, no. 3
,
F
e
b. 2022, doi
:
10.3390/
s
22031200.
[
13]
M
.
O
z
t
ur
k,
M
.
A
kr
a
m
,
S
.
H
us
s
a
i
n,
a
nd
M
.
A
.
I
m
r
a
n,
“
N
ove
l
Q
oS
-
a
w
a
r
e
pr
oa
c
t
i
ve
s
pe
c
t
r
um
a
c
c
e
s
s
t
e
c
hni
que
s
f
or
c
ogni
t
i
ve
r
a
di
o
us
i
ng m
a
c
hi
ne
l
e
a
r
n
i
ng,”
I
E
E
E
A
c
c
e
s
s
, vol
. 7, pp. 70811
–
70827, 2019, doi
:
10.1109/
A
C
C
E
S
S
.2019.2918380.
[
14]
A
.
Z
hu,
S
.
G
uo,
B
.
L
i
u,
M
.
M
a
,
J
.
Y
a
o,
a
nd
X
.
S
u,
“
A
da
pt
i
ve
m
ul
t
i
s
e
r
vi
c
e
h
e
t
e
r
oge
ne
ous
ne
t
w
or
k
s
e
l
e
c
t
i
on
s
c
he
m
e
i
n
m
obi
l
e
e
dge
c
om
put
i
ng,”
I
E
E
E
I
nt
e
r
ne
t
of
T
hi
ngs
J
our
nal
, vol
. 6, no. 4, pp. 6862
–
687
5, A
ug. 2019, doi
:
10.1109/
J
I
O
T
.2019.2912155.
[
15]
J
.
C
he
n,
Y
.
W
a
ng,
Y
.
L
i
,
a
nd
E
.
W
a
ng,
“
Q
oE
-
a
w
a
r
e
i
nt
e
l
l
i
ge
nt
ve
r
t
i
c
a
l
ha
ndof
f
s
c
he
m
e
ove
r
he
t
e
r
oge
ne
ou
s
w
i
r
e
l
e
s
s
a
c
c
e
s
s
ne
t
w
or
ks
,”
I
E
E
E
A
c
c
e
s
s
, vol
. 6, pp. 38285
–
38293, 2018, doi
:
10.1109/
A
C
C
E
S
S
.2018.2853730.
[
16]
S
.
H
.
S
r
e
e
dha
r
a
,
V
.
K
um
a
r
,
a
nd
S
.
S
a
l
m
a
,
“
E
f
f
i
c
i
e
nt
bi
g
da
t
a
c
l
us
t
e
r
i
ng
us
i
ng
a
dhoc
f
uz
z
y
c
m
e
a
ns
a
nd
a
ut
o
-
e
nc
ode
r
C
N
N
,”
i
n
L
e
c
t
ur
e
N
ot
e
s
i
n N
e
t
w
o
r
k
s
and S
y
s
t
e
m
s
, vol
. 563, pp. 353
–
368
, 2023,
doi
:
10.1007/
978
-
981
-
19
-
7402
-
1_25.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
:
4281
-
4290
4290
[
17]
P
.
P
uj
a
r
,
A
.
K
um
a
r
,
a
nd
V
.
K
um
a
r
,
“
P
l
a
nt
l
e
a
f
de
t
e
c
t
i
on
t
hr
ough
m
a
c
hi
ne
l
e
a
r
ni
ng
ba
s
e
d
i
m
a
ge
c
l
a
s
s
i
f
i
c
a
t
i
on
a
ppr
oa
c
h,”
I
A
E
S
I
nt
e
r
nat
i
onal
J
our
nal
of
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
, vol
. 13, no. 1, pp. 1139
–
1148, M
a
r
. 2024, doi
:
10.11591/
i
j
a
i
.v13.i
1.pp1139
-
1148.
[1
8
]
B
.
M
a
,
W
.
G
uo,
a
nd
J
.
Z
ha
ng,
“
A
s
ur
ve
y
of
on
l
i
ne
da
t
a
-
dr
i
ve
n
pr
oa
c
t
i
ve
5G
ne
t
w
or
k
opt
i
m
i
s
a
t
i
on
us
i
ng
m
a
c
hi
ne
l
e
a
r
n
i
ng
,”
I
E
E
E
A
c
c
e
s
s
, vol
. 8, pp. 35606
–
35637, 2020, doi
:
10.1109/
A
C
C
E
S
S
.2020.2975004.
[
19
]
L
.
Y
a
n
e
t
al
.
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
-
ba
s
e
d
ha
ndove
r
s
f
or
s
ub
-
6
G
H
z
a
nd
m
m
W
a
ve
i
nt
e
gr
a
t
e
d
ve
hi
c
ul
a
r
ne
t
w
or
ks
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on W
i
r
e
l
e
s
s
C
om
m
uni
c
at
i
ons
, vol
. 18, no. 10, pp. 4873
–
4885, O
c
t
. 2019, doi
:
10.1109/
T
W
C
.2019.2930193.
[2
0
]
S
. I
. M
a
j
i
d, S
.
W
. S
ha
h,
S
. N
. K
.
M
a
r
w
a
t
, A
. H
a
f
e
e
z
,
H
. A
l
i
, a
nd
N
. J
a
n,
“
U
s
i
n
g
a
n e
f
f
i
c
i
e
nt
t
e
c
hni
qu
e
ba
s
e
d on
dyna
m
i
c
l
e
a
r
ni
ng
pe
r
i
od
f
or
i
m
pr
ov
i
ng
de
l
a
y
i
n
AI
-
ba
s
e
d
ha
ndove
r
,”
M
ob
i
l
e
I
nf
or
m
at
i
on
Sy
s
t
e
m
s
,
vol
.
2021,
pp.
1
–
19,
A
ug.
2021,
doi
:
10.1155/
2021/
2775278.
[2
1
]
R
.
S
.
-
I
bor
r
a
,
L
.
B
.
-
E
s
c
obe
do,
a
nd
J
.
S
a
nt
a
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
-
ba
s
e
d
r
a
di
o
a
c
c
e
s
s
t
e
c
hnol
ogy
s
e
l
e
c
t
i
on
i
n
t
he
i
nt
e
r
ne
t
of
m
ovi
ng
t
hi
ngs
,”
C
hi
na C
om
m
uni
c
at
i
on
s
, vol
. 18, no. 7, pp. 13
–
24, J
ul
. 2021, doi
:
10.23919/
J
C
C
.2021.07.002.
[2
2
]
Y
.
C
a
o,
S
.
Y
.
L
i
e
n,
Y
.
C
.
L
i
a
ng,
K
.
C
.
C
he
n,
a
nd
X
.
S
he
n,
“
U
s
e
r
a
c
c
e
s
s
c
ont
r
ol
i
n
ope
n
r
a
di
o
a
c
c
e
s
s
ne
t
w
or
ks
:
a
f
e
de
r
a
t
e
d
de
e
p
r
e
i
nf
or
c
e
m
e
nt
l
e
a
r
ni
ng
a
ppr
oa
c
h,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
W
i
r
e
l
e
s
s
C
om
m
uni
c
at
i
ons
,
vol
.
21,
no.
6,
pp.
3721
–
3736,
J
un.
2022,
doi
:
10.1109/
T
W
C
.2021.3123500.
[2
3
]
B
.
G
a
l
ki
n,
E
.
F
ons
e
c
a
,
R
.
A
m
e
r
,
L
.
A
.
D
a
s
i
l
va
,
a
nd
I
.
D
u
s
pa
r
i
c
,
“
R
E
Q
I
B
A
:
r
e
gr
e
s
s
i
on
a
nd
de
e
p
q
-
l
e
a
r
ni
ng
f
or
i
nt
e
l
l
i
ge
nt
U
A
V
c
e
l
l
ul
a
r
us
e
r
t
o
ba
s
e
s
t
a
t
i
on
a
s
s
oc
i
a
t
i
on
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
V
e
hi
c
ul
ar
T
e
c
hnol
ogy
,
vol
.
71,
no.
1,
pp.
5
–
20,
J
a
n.
2022,
doi
:
10.1109/
T
V
T
.2021.3126536.
[2
4
]
B
.
Z
ha
ng,
S
.
L
i
u,
J
.
L
.
Y
u,
a
nd
Z
.
H
a
n,
“
A
l
e
a
r
ni
ng
a
i
de
d
l
ong
-
t
e
r
m
us
e
r
a
s
s
oc
i
a
t
i
on
s
c
he
m
e
f
or
ul
t
r
a
-
de
ns
e
n
e
t
w
or
ks
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on V
e
hi
c
ul
ar
T
e
c
hnol
ogy
, vol
. 71, no. 1, pp. 820
–
830, J
a
n. 2022, doi
:
10.1109/
T
V
T
.2021.3127367.
[2
5
]
J
.
C
he
n
e
t
al
.
,
“
J
o
i
nt
t
a
s
k
a
s
s
i
gnm
e
nt
a
nd
s
pe
c
t
r
um
a
l
l
oc
a
t
i
on
i
n
he
t
e
r
oge
ne
ous
ua
v
c
om
m
uni
c
a
t
i
on
ne
t
w
or
ks
:
a
c
oa
l
i
t
i
on
f
or
m
a
t
i
on
ga
m
e
-
t
he
or
e
t
i
c
a
pp
r
oa
c
h
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
W
i
r
e
l
e
s
s
C
om
m
u
ni
c
at
i
ons
,
vol
.
20,
no.
1,
pp.
440
–
452,
J
a
n.
2021,
doi
:
10.1109/
T
W
C
.2020.3025316.
[2
6
]
H
.
H
u,
C
.
M
a
,
a
nd
B
.
T
a
ng,
“
C
ha
nn
e
l
a
l
l
oc
a
t
i
on
s
c
he
m
e
f
or
ul
t
r
a
-
de
ns
e
f
e
m
t
oc
e
l
l
ne
t
w
or
ks
:
ba
s
e
d
on
c
oa
l
i
t
i
on
f
or
m
a
t
i
on
ga
m
e
a
nd
m
a
t
c
hi
ng
g
a
m
e
,”
i
n
2019
I
E
E
E
M
T
T
-
S I
nt
e
r
nat
i
onal
W
i
r
e
l
e
s
s
Sy
m
pos
i
um
,
I
W
S
2019
-
P
r
oc
e
e
di
ngs
,
M
a
y
2019,
pp.
1
–
3.
D
oi
:
10.1109/
I
E
E
E
-
I
W
S
.2019.8803970.
[2
7
]
L
. L
.
M
a
goa
r
ou,
“
E
f
f
i
c
i
e
nt
c
ha
nne
l
c
ha
r
t
i
ng vi
a
pha
s
e
-
i
ns
e
n
s
i
t
i
ve
di
s
t
a
nc
e
c
om
put
a
t
i
on
,”
I
E
E
E
W
i
r
e
l
e
s
s
C
om
m
uni
c
at
i
ons
L
e
t
t
e
r
s
,
vol
. 10, no. 12, pp. 2634
–
2638, D
e
c
. 2021, doi
:
10.1109/
L
W
C
.2021.3109295.
[2
8
]
H
.
T
ong,
T
.
W
a
ng,
Y
.
Z
hu,
X
.
L
i
u,
S
.
W
a
ng,
a
nd
C
.
Y
i
n,
“
M
obi
l
i
t
y
-
a
w
a
r
e
s
e
a
m
l
e
s
s
ha
ndove
r
w
i
t
h
M
P
T
C
P
i
n
s
of
t
w
a
r
e
-
de
f
i
ne
d
he
t
ne
t
s
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
N
e
t
w
o
r
k
and
Se
r
v
i
c
e
M
anage
m
e
nt
,
v
ol
.
18,
no.
1,
pp.
498
–
510,
M
a
r
.
2021,
doi
:
10.1109/
T
N
S
M
.2021.3050627.
[
2
9
]
X
.
T
a
n,
G
.
C
he
n,
a
nd
H
.
S
un,
“
V
e
r
t
i
c
a
l
ha
ndove
r
a
l
go
r
i
t
hm
ba
s
e
d
on
m
ul
t
i
-
a
t
t
r
i
but
e
a
nd
ne
ur
a
l
ne
t
w
or
k
i
n
he
t
e
r
oge
ne
ous
i
nt
e
gr
a
t
e
d
ne
t
w
or
k,”
E
ur
as
i
p
J
our
nal
on
W
i
r
e
l
e
s
s
C
om
m
uni
c
at
i
ons
and
N
e
t
w
or
k
i
ng
,
vol
.
2020,
no.
1,
D
e
c
.
2020,
doi
:
10.1186/
s
13638
-
020
-
01822
-
1.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Bhavana
Srinivas
earned
her
Bachelor'
s
of
Engineering
BE
degre
e
in
ECE
fro
m
Nagarjuna
College
of
Engineering
&
Technology,
Bangalore
in
200
9.
She
has
obtained
her
master'
s
degree
in
M.Tech
.
(ECE)
from
Dayanandhsagar
College
of
Engineering
in
2012
and
currently
she
is
a
research
scholar
at
New
Horizon
Coll
ege
of
Engine
ering
doing
her
Ph
.
D
.
in
Electronics
and
Communicat
ion
Enginee
ring.
She
has
attende
d
many
workshop
s
and
induction pro
grams cond
ucted by va
rious univer
sities. Her
areas o
f interes
t are
communi
cation
and machi
ne learnin
g
. She can be contacted a
t email:
bhavanas_12@
rediff
mail.com.
Nadig
Vijayendra
Uma
Reddy
is
p
rofessor
and
h
ead
(AI
a
nd
ML),
Ne
w
Horizon College
of
Engineering
, Bangalore with
an experience of
26
years in Teaching. She is
qualified in Bachelor and Master De
grees in Electronics
and
Commun
ication Engineering, and
Ph.D
.
in
m
ixed mode
VLSI
design
and executive
PG degr
ee in
AI &
ML. He
r are
as of
inter
est
are
analog
and
mixed
mode
VLSI
design
,
data
analytics,
mac
hine
learning,
artificial
intelligence
,
and
wireless
sensor
networks.
She
can
be
contacted
at
email:
nvumareddy@
gmail.com
.
Evaluation Warning : The document was created with Spire.PDF for Python.