T
E
L
KOM
N
I
KA
T
e
le
c
o
m
m
u
n
ic
a
t
io
n
,
C
o
m
p
u
t
i
n
g
,
E
l
ect
ro
n
i
cs
a
n
d
C
o
n
t
ro
l
Vo
l
.
19
, N
o
.
4
,
A
ugus
t
2021
,
pp.
1197~
1207
I
S
S
N
:
1693
-
6930,
a
c
c
r
e
di
t
e
d F
i
r
s
t
G
r
a
de
by K
e
m
e
nr
i
s
t
e
kdi
kt
i
,
D
e
c
r
e
e
N
o:
21/
E
/
K
P
T
/
2018
D
O
I
:
10.
12928/
T
E
L
K
O
M
N
I
K
A
.
v19i
4.
18521
1197
Jou
r
n
al
h
om
e
page
:
ht
t
p:
/
/
j
our
nal
.
uad
.
ac
.
i
d/
i
nde
x
.
php/
T
E
L
K
O
M
N
I
K
A
Op
t
im
a
l
r
es
o
ur
ce a
l
l
o
ca
t
i
o
n f
o
r ro
u
t
e s
el
ec
t
i
o
n i
n
a
d
-
ho
c
net
w
o
r
ks
M
ar
w
a
K
.
F
ar
h
an
1
,
M
u
ayad
S
.
C
r
ooc
k
2
1
C
oll
e
ge
of
I
nf
or
m
a
ti
on E
ng
ine
e
r
in
g,
Al
-
Na
hr
a
i
n
Un
ive
r
sit
y,
B
a
g
hda
d,
I
r
a
q
2
C
ontr
ol a
nd S
y
st
e
m
s E
ng
ine
e
r
in
g De
pa
r
tm
e
nt,
U
ni
ve
r
s
i
ty of
Te
c
h
no
lo
gy,
B
a
g
hda
d,
I
r
a
q
A
rt
i
cl
e I
n
f
o
AB
S
T
RACT
A
r
tic
le
h
is
to
r
y
:
R
ecei
v
ed
N
ov 11,
2020
R
ev
i
s
ed
M
a
r
3,
2021
A
ccep
t
ed
M
a
r
11,
2021
No
wa
da
ys,
t
he
se
le
c
ti
on of
the
op
tim
um
p
a
t
h
i
n
m
obi
l
e
a
d
h
oc
ne
t
wor
ks
(
M
AN
ETS
)
is be
in
g a
n im
p
or
ta
nt i
ss
ue
tha
t sh
ou
ld be
s
olv
e
d sm
a
r
tl
y.
I
n thi
s
pa
pe
r
,
a
n
op
tim
a
l
p
a
t
h se
le
c
t
io
n m
e
t
ho
d
i
s
pr
op
ose
d
f
or
M
A
NE
T
u
si
ng
the
L
a
g
r
a
n
g
e
m
ul
ti
pl
ie
r
a
p
pr
oa
c
h.
The
op
tim
iz
a
t
io
n
pr
obl
e
m
c
o
ns
ide
r
s
t
he
obj
e
c
t
ive
f
unc
ti
on
of
m
a
xim
iz
i
ng
bi
t
r
a
te
,
u
nde
r
the
c
o
ns
tr
a
i
nt
s
of
m
in
im
iz
in
g
the
pa
c
ke
t
lo
ss,
a
n
d la
te
nc
y
.
The
o
bta
in
e
d s
im
u
la
t
io
n
r
e
su
lt
s sh
ow
tha
t the
pr
o
po
se
d L
a
gr
a
n
ge
op
tim
iz
a
t
io
n of
r
a
te
,
de
la
y,
a
nd p
a
c
ke
t
l
os
s a
lg
or
i
thm
(
L
OR
DP
)
im
pr
ove
s
the
se
le
c
ti
on
of
op
tim
a
l
pa
t
h
i
n
c
om
pa
r
i
so
n
to
a
d
-
hoc
on
-
d
e
m
a
n
d
dis
ta
nc
e
ve
c
tor
pr
ot
oc
o
l
(
AOD
V)
.
We
inc
r
e
a
se
d the
pe
r
f
or
m
a
nc
e
of
the
s
ys
te
m
b
y
10.
6
M
bp
s
f
or
b
it
r
a
te
a
nd
0.
13
3
m
s
f
or
la
te
nc
y.
Ke
y
wo
r
d
s
:
AODV
L
ag
r
an
g
e
mu
ltip
lie
r
s
M
ANE
T
O
p
timiz
a
tio
n
T
his
is
a
n
o
pe
n
ac
c
e
s
s
ar
tic
le
u
nde
r
the
CC
B
Y
-
SA
lic
e
n
se
.
C
or
r
e
s
pon
di
n
g A
u
t
h
or
:
M
ar
w
a K
.
F
ar
h
an
C
ol
l
e
ge
of
I
n
f
or
m
a
t
i
on
E
ngi
ne
e
r
i
ng
Al
-
N
a
hr
a
i
n U
ni
ve
r
s
i
t
y
B
a
ghda
d,
I
r
a
q
E
ma
il:
m
ar
w
a
.
k
.
f
ar
h
an
@
g
m
ai
l
.
co
m
1.
I
NT
RO
DUC
T
I
O
N
M
obi
l
e
ad
-
hoc
ne
t
w
or
ks
(
MA
N
E
T
)
i
s a
se
l
f
-
c
onf
i
gur
e
d
a
nd i
nf
r
a
s
t
r
uc
t
ur
e
-
f
r
e
e
ne
t
w
or
k
t
ha
t
i
s
ba
s
e
d
on a
d
-
hoc
c
om
m
uni
c
a
t
i
ons
.
T
he
r
out
i
ng i
n
m
obi
l
e
a
d
-
hoc
ne
t
w
or
ks
i
s
ve
r
y de
f
i
a
nc
e
due
t
o
t
he
p
e
r
s
i
s
t
e
nt
upda
t
e
s
i
n t
opol
ogi
e
s
,
a
nd
a
c
t
i
ve
r
ou
t
e
s
m
a
y be
di
s
c
onne
c
t
e
d
d
u
e t
o
w
i
r
el
es
s
d
ev
i
ce m
o
b
i
l
i
t
y
f
r
o
m
o
n
e p
l
ace
t
o a
not
he
r
[
1]
.
T
he
s
e
w
i
r
e
l
e
s
s
node
s
ope
r
a
t
e
a
s
a
hos
t
a
nd a
s
a
r
out
e
r
t
o
a
l
l
ow
t
he
i
nt
e
r
na
l
c
om
m
u
ni
c
a
t
i
ons
av
ai
l
ab
l
e.
T
h
er
ef
o
r
e,
each
n
o
d
e i
n
t
er
act
s
i
n
t
h
e r
o
u
t
i
n
g
p
r
o
ces
s
t
o
d
el
i
v
er
a p
ack
et
t
o
t
h
e
d
es
t
i
n
at
i
on node
.
M
obi
l
i
t
y,
t
opol
ogy c
ha
nge
s
,
pow
e
r
a
nd r
e
s
our
c
e
s
hor
t
a
ge
,
non
-
cen
t
r
al
i
zed
co
n
t
r
o
l
ar
e al
l
M
A
N
E
T
en
v
i
r
o
n
m
en
t
pr
ope
r
t
i
e
s
.
S
uc
h
c
ha
r
a
c
t
e
r
i
s
t
i
c
s
pr
ovoke
t
he
u
r
ge
t
o
de
s
i
gn a
r
out
i
ng
pr
ot
oc
ol
t
ha
t
a
gr
e
e
s
w
i
t
h
s
om
e
t
e
r
m
s
.
T
he
r
out
e
s
e
l
e
c
t
i
on pr
ot
oc
ol
m
us
t
be
qua
l
i
f
i
e
d
t
o a
da
pt
t
o t
he
s
e
va
r
i
a
t
i
ons
by
c
ont
i
nua
l
l
y m
on
i
t
or
i
ng
t
he
l
i
nk s
t
a
t
e
a
nd pe
r
f
or
m
r
out
e
s
a
c
c
or
di
ngl
y
[
2]
d
i
ffe
re
n
t
i
s
s
u
es
w
er
e ad
d
r
es
s
ed
i
n
t
h
e p
r
i
o
r
r
es
ear
ch
ar
ea,
y
et
s
u
ch
t
h
r
i
v
i
n
g
ne
t
w
or
k bus
i
ne
s
s
s
ubj
e
c
t
s
t
o c
ont
i
nuous
i
m
pr
ove
m
e
nt
s
a
nd e
nha
nc
e
m
e
nt
i
n t
e
r
m
s
of
Q
oS
a
nd
Q
oE
.
T
he
pr
oc
e
s
s
of
t
r
a
ns
f
e
r
r
i
ng da
t
a
pa
c
ke
t
s
f
r
om
s
our
c
e
poi
nt
t
o de
s
t
i
na
t
i
on
poi
nt
t
ha
t
s
ubj
e
c
t
t
o r
e
s
our
c
e
c
ons
t
r
a
i
nt
s
,
s
uc
h a
s
e
ne
r
gy,
de
l
a
y,
bi
t
r
a
t
e
,
pa
c
ke
t
l
os
s
r
a
t
e
,
a
nd
c
os
t
s
houl
d i
nc
l
ude
t
he
us
e
o
f
op
t
i
m
i
z
a
t
i
on
m
e
t
hods
i
n t
he
r
out
i
n
g pr
oc
e
s
s
[
3]
.
T
hus
,
w
e
i
nt
r
oduc
e
a
m
e
t
hod
ba
s
e
d on
L
a
gr
a
nge
o
pt
i
m
i
z
a
t
i
on t
ha
t
s
e
l
e
c
t
s
t
he
op
t
i
m
a
l
r
out
e
f
r
o
m
t
he
de
vi
c
e
t
o ot
he
r
de
vi
c
e
s
i
n a
M
A
N
E
T
.
S
pe
c
i
a
l
l
y de
s
i
gne
d t
o s
a
t
i
s
f
y t
he
de
s
i
r
e
d obj
e
c
t
i
ve
f
unc
t
i
on ba
s
e
d on
s
uppl
e
m
e
nt
a
r
y r
out
i
ng
r
e
qui
r
e
m
e
nt
s
.
W
he
r
e
da
t
a
pa
c
k
e
t
s
a
r
e
s
e
nt
us
i
ng
r
ou
t
e
s
f
r
om
t
he
r
out
i
ng
t
a
bl
e
t
ha
t
a
r
e
s
el
ect
ed
b
as
ed
o
n
t
h
e
r
eq
u
es
t
ed
ch
ar
act
er
i
s
t
i
cs
.
T
h
e ai
m
i
s
t
o
m
ax
i
m
i
ze
t
h
e
b
i
t
r
at
e
f
r
o
m
n
o
d
e
t
o
n
o
d
e an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
19
, N
o
.
4
,
A
ugus
t
2021
:
1197
-
1207
1198
min
imiz
e
th
e
to
ta
l d
e
la
y
a
n
d
p
a
c
k
e
t lo
s
s
p
r
o
b
a
b
ility
in
w
ir
e
le
s
s
d
a
ta
tr
a
n
s
mis
s
io
n
.
T
h
e
r
e
f
o
r
e,
t
h
i
s
p
ap
er
ad
d
r
es
s
e
s
a t
h
eo
r
et
i
cal
an
d
p
r
act
i
cal
s
c
en
ar
i
o
.
P
r
io
r
e
ffo
rt
s
i
n
r
es
ear
ch
f
i
el
d
ha
ve
be
e
n i
nve
s
t
e
d t
o
a
ddr
e
s
s
a
di
ve
r
s
e
i
s
s
ue
i
n t
he
opt
i
m
a
l
r
out
e
pol
i
c
i
e
s
a
nd m
e
t
h
ods
i
n t
e
r
m
s
of
va
r
i
ous
o
b
j
ect
i
v
es
(
min
imiz
in
g
th
e
d
u
r
a
tio
n
o
r
min
imiz
in
g
th
e
e
ne
r
gy c
ons
um
pt
i
on or
e
ve
n nu
m
be
r
of
hops
)
.
A
s
e
r
i
e
s
of
w
or
ks
ha
ve
be
e
n i
nve
s
t
i
ga
t
e
d a
nd
N
e
w
t
e
c
hnol
ogi
e
s
,
a
s
w
e
l
l
a
s
t
e
c
hni
que
s
,
ha
ve
be
e
n e
xpl
oi
t
e
d i
n t
he
pr
i
or
w
or
k
[
3]
.
A
ut
ho
r
s
of
[
4]
a
d
d
r
e
sse
d
di
r
e
c
t
e
nd
de
vi
c
e
s
c
om
m
uni
c
a
t
i
on i
n
c
a
s
e
of
r
e
s
t
r
i
c
t
e
d c
onne
c
t
i
vi
t
y t
o
t
he
c
e
l
l
ul
a
r
ne
t
w
or
k due
t
o
di
s
a
s
t
e
r
s
or
em
er
g
en
ci
es
.
F
o
r
t
h
e
p
er
f
o
r
m
an
ce ev
al
u
at
i
o
n
o
f
Q
o
S
i
n
ad
-
hoc
ne
t
w
or
ks
a
n
d
c
o
n
s
tr
a
in
t s
a
tis
f
a
c
tio
n
i
n
ad
-
hoc
on
-
de
m
a
nd di
s
t
a
nc
e
ve
c
t
or
pr
ot
oc
ol
(
A
O
D
V
)
p
r
ot
oc
ol
,
t
he
a
ut
hor
s
o
f
[
5]
e
nha
nc
e
d t
he
c
onve
nt
i
ona
l
c
uc
koo
s
ear
ch
a
l
gor
i
t
hm
to
c
hos
e
t
he
Q
oS
pa
t
h ba
s
e
d on
th
e
r
out
i
ng l
oa
d
,
r
e
s
i
dua
l
e
ne
r
gy
,
a
nd hop
c
ount
.
M
or
e
ove
r
,
r
es
ear
ch
er
s
o
f
[
6]
i
nt
r
oduc
e
d a
nove
l
Q
oS
r
out
i
n
g i
n
M
A
N
E
T
s
us
i
ng
em
er
g
en
t
i
n
t
el
l
i
g
en
ce
.
F
or
da
t
a
l
os
s
min
imiz
a
tio
n
,
a
nd e
ne
r
gy
-
e
f
f
i
c
i
e
nt
c
l
us
t
e
r
i
ng w
a
s
i
nt
r
oduc
e
d us
i
ng P
S
O
a
nd f
uz
z
y opt
i
m
i
z
a
t
i
on.
I
n t
e
r
m
s
of
di
s
a
s
t
e
r
r
e
s
pons
e
,
t
he
a
ut
hor
s
of
[
7]
f
oc
us
e
d on D
2D
c
om
m
uni
c
a
t
i
ons
t
o e
xt
e
nd t
he
ba
s
e
s
t
a
t
i
on's
c
ove
r
a
ge
.
T
he
y us
e
d c
ont
r
ol
l
e
r
-
a
s
s
i
s
t
e
d r
out
i
ng t
o i
nc
r
e
a
s
e
t
he
t
ot
a
l
t
hr
oughput
t
o
m
a
xi
m
um
us
i
ng a
nt
c
ol
ony
o
p
timiz
a
tio
n
.
A
l
so
,
t
h
e a
ut
hor
s
o
f
[
8]
f
o
r
m
ul
a
t
e
d
a
qua
l
i
t
y
o
f
e
xpe
r
i
e
nc
e
r
out
i
ng
ove
r
w
i
r
e
le
s
s
m
u
lti
-
hop
ne
t
w
or
ks
unde
r
t
i
m
e
-
c
ons
t
r
a
i
nt
s
.
T
he
y
pr
opos
e
d
a
he
ur
i
s
t
i
c
a
l
gor
i
t
hm
t
o
s
pe
e
d up
f
i
nd
i
ng s
ol
ut
i
ons
.
F
or
w
r
i
t
e
r
s
t
o
en
h
an
ce t
h
e cap
aci
t
y
o
f
t
r
af
f
i
c o
f
f
l
o
ad
i
n
g
f
o
r
c
el
l
u
l
ar
-
D
2D
r
e
l
a
ys
,
a
ut
hor
s
of
[
9]
i
nt
r
oduc
e
d a
t
h
r
ee D
2
D
c
om
m
uni
c
a
t
i
on m
ode
l
.
O
n t
he
ot
he
r
ha
nd,
i
n
[
1
0]
t
he
a
ut
hor
s
ut
i
l
i
z
e
d t
he
O
L
S
R
r
out
i
ng a
l
gor
i
t
hm
t
o
bui
l
d a
m
ul
t
i
-
hop D
2D
c
om
m
uni
c
a
t
i
ons
pl
a
t
f
or
m
ba
s
e
d on
s
m
a
r
t
phone
s
t
o e
xpa
nd
t
he
s
i
ngl
e
-
hop D
2D
s
c
e
na
r
i
os
.
T
he
y
m
e
a
s
ur
e
pe
r
f
or
m
a
nc
e
s
of
e
ne
r
gy c
ons
um
pt
i
on
,
c
ove
r
a
ge
,
ne
t
w
or
k
l
a
t
e
nc
y,
a
nd
l
i
nk
qua
l
i
t
y
.
I
n
[
11]
,
t
he
a
ut
hor
s
in
tr
o
d
u
c
e
d
a
r
e
lia
b
ility
-
a
w
a
r
e
A
O
D
V
by
c
onf
e
r
r
i
n
g s
t
a
bi
l
i
t
y
t
o
pa
t
hs
.
T
h
e s
el
ect
ed
r
o
u
t
es
ar
e
r
es
t
r
i
ct
ed
w
i
t
h
b
a
ndw
i
dt
h
a
nd
en
d
-
to
-
e
nd de
l
a
y.
T
h
e r
es
ear
ch
e
r
i
n
[
1
2]
p
r
opos
e
d t
r
us
t
a
nd
pr
e
ve
nt
pa
r
a
m
e
t
e
r
s
a
ga
i
ns
t
m
a
l
i
c
i
ous
ne
t
w
or
ks
t
o i
de
nt
i
f
y a
s
e
c
ur
e
r
out
e
.
T
he
de
l
i
ve
r
y r
a
t
e
i
nc
r
e
a
s
e
s
m
or
e
s
i
gni
f
i
c
a
nt
l
y w
he
n
m
a
l
i
c
i
ous
node
s
i
nc
r
e
a
s
e
us
i
ng t
he
pr
opos
e
d m
e
t
hod t
ha
n
t
ha
t
of
t
he
A
O
D
V
a
nd
T
V
A
O
D
V
.
T
he
a
ut
ho
r
s
of
[
13]
i
nt
r
oduc
e
d a
hybr
i
d O
L
S
R
v2 t
ha
t
i
s
m
ul
t
i
pa
t
h
e
ne
r
gy a
nd Q
oS
-
a
w
a
r
e
to
s
o
lv
e
th
e
limita
tio
n
o
f
e
n
e
r
g
y
r
e
s
our
c
e
s
,
node
s
m
obi
l
i
t
y,
a
nd
t
r
a
f
f
i
c
c
onge
s
t
i
on.
T
he
r
e
s
e
a
r
c
he
r
i
n
[
14]
p
r
es
en
t
ed
a M
A
T
L
A
B
-
b
as
e
d
ad
-
hoc
on
-
de
m
a
nd di
s
t
a
nc
e
ve
c
t
or
s
i
m
ul
a
t
i
on
pr
e
s
e
nt
e
d
t
o pr
ovi
de
a
m
e
a
ni
ngf
ul
m
e
t
hod of
de
m
ons
t
r
a
t
i
ng ba
s
i
c
r
out
i
ng c
onc
e
pt
s
a
nd
f
a
c
i
l
i
t
a
t
i
ng vi
s
ua
l
l
e
a
r
ni
ng.
T
h
e
a
ut
hor
s
of
[
15]
,
pr
opos
e
d a
vi
r
t
ua
l
A
d
hoc
r
out
i
ng
pr
ot
oc
ol
t
o
i
n
cr
eas
e s
ecu
r
i
t
y
an
d
s
cal
ab
i
l
i
t
y
.
T
he
y a
l
s
o
de
ve
l
ope
d a
s
our
c
e
-
r
out
i
ng pr
ot
oc
ol
t
ha
t
a
c
hi
e
ve
d be
t
t
e
r
s
c
a
la
b
ility
a
n
d
lo
w
e
r
s
c
ons
um
e
d pow
e
r
.
A
l
s
o,
t
he
a
ut
hor
s
of
[
1
6]
i
nt
r
oduc
e
d m
obi
l
i
t
y
-
e
ne
r
gy i
m
pr
ove
d a
nt
c
ol
ony opt
i
m
i
z
a
t
i
on
r
out
i
ng
m
e
t
hod
.
T
he
m
e
t
hod
s
pe
e
de
d up
t
he
r
out
i
ng
a
l
gor
i
t
hm
a
nd
r
e
duc
e
d t
he
r
out
e
di
s
c
ove
r
y pa
c
ke
t
s
.
A
ne
t
w
o
r
k c
odi
ng
-
ba
s
e
d r
out
i
n
g pr
ot
oc
ol
w
a
s
pr
opos
e
d i
n
[
17]
t
o
r
ed
u
ce l
at
en
cy
an
d
t
r
af
f
i
c
l
oa
d f
o
r
t
r
a
ns
m
i
s
s
i
on of
onl
i
ne
ga
m
i
ng
.
T
he
y
pr
opos
e
d a
m
e
di
um
a
c
c
e
s
s
s
c
he
dul
i
ng i
n
d
ev
i
ce t
o
d
ev
i
ce
i
nf
r
a
s
t
r
uc
t
ur
e
a
nd a
l
s
o c
ons
i
de
r
e
d p
r
obl
e
m
s
of
pa
c
ke
t
l
os
s
.
I
n t
e
r
m
s
o
f
opt
i
m
a
l
r
out
e
s
,
t
he
a
ut
hor
s
of
[
18]
p
r
o
pos
e
d a
p
er
f
o
r
m
an
ce
-
on
-
d
e
m
a
nd r
out
i
ng pr
ot
oc
ol
.
T
he
r
ou
t
e
i
s
s
e
l
e
c
t
e
d by hop
num
be
r
a
nd
t
hr
oughpu
t
.
T
he
t
hr
oughput
c
ondi
t
i
on
m
ean
s
to
a
c
h
ie
v
e
th
e
m
in
imu
m
t
hr
e
s
hol
d
w
ith
th
e
hi
ghe
s
t
t
h
r
oughput
of
t
he
e
nt
i
r
e
r
out
e
a
m
ong
c
a
ndi
da
t
e
.
A
ne
w
c
onc
e
pt
of
r
out
e
a
v
a
i
l
a
bi
l
i
t
y
w
as
p
r
es
en
t
ed
i
n
[
19]
a
s
a
m
e
a
s
ur
e
m
e
nt
of
r
out
e
no uni
f
or
m
i
t
y i
n a
M
A
N
E
T
a
s
i
t
r
e
p
r
e
s
e
nt
s
t
he
Q
oS
or
Q
oE
of
vi
de
o s
t
r
e
a
m
i
ng.
T
he
y c
onf
i
r
m
e
d t
w
o Q
oS
m
e
t
r
i
c
s
a
nd f
ounde
d t
ha
t
r
o
u
te
a
v
a
ila
b
ility
i
s
af
f
ect
ed
b
y
ch
an
g
es
i
n vi
de
o qua
l
i
t
y.
M
or
e
on
vi
de
os
ove
r
M
A
N
E
T
s
,
a
ut
hor
s
of
[
2
0]
s
t
r
eam
ed
hi
gh de
f
i
ni
t
i
on
vi
de
os
.
T
he
y
de
s
i
gne
d a
t
r
a
ns
m
i
s
s
i
on s
ys
t
e
m
f
ol
l
ow
e
d by
a
di
s
t
o
r
t
i
on s
ys
t
e
m
t
o
ev
al
u
at
e t
h
e
p
ack
et
-
l
o
s
s
r
at
e an
d
en
d
-
to
-
en
d
de
l
a
y a
nd i
m
pr
ove
d
Q
oS
an
d
Q
o
E
.
A
n opt
i
m
i
z
e
d r
out
i
ng
m
e
t
hod w
a
s
p
r
opos
e
d i
n
[
21]
t
o
en
h
an
ce t
h
e
pe
r
f
or
m
a
nc
e
of
t
he
ne
t
w
or
k
t
h
at
w
as
s
u
b
j
ect
ed
t
o
t
h
e
ma
x
imu
m b
it
r
a
te
,
min
imu
m p
a
c
k
e
t lo
s
s
r
a
te
,
a
n
d
min
imu
m d
e
la
y
.
T
h
e
p
a
th
s
e
le
c
tio
n
r
e
lie
s
o
n
th
e
w
e
i
ght
e
d
S
u
m o
p
timiz
a
t
i
on m
e
t
hod
,
t
he
non
-
dom
i
na
t
e
d
s
or
t
i
ng
-
G
en
et
i
cI
I
,
a
n
d
w
e
i
ght
e
d s
um
-
g
en
et
i
c
o
p
timiz
a
tio
n
.
N
e
tw
o
r
k
a
ssi
st
e
d
-
r
out
i
ng
f
or
de
vi
c
e
-
to
-
d
ev
i
ce
ar
ch
i
t
ect
u
r
es
of
5G
w
a
s
i
nt
r
oduc
e
d i
n
[
22]
t
o e
xt
e
nd t
he
ba
s
e
s
t
a
t
i
ons
c
ove
r
a
ge
.
N
A
R
t
ook i
n
c
ons
i
de
r
a
t
i
on
th
a
t c
o
mmu
n
ic
a
tio
n
s
of
D
2
D
ar
e m
an
ag
ed
b
y
b
as
e s
t
at
i
o
n
s
.
E
ve
nt
ua
l
l
y,
t
h
e r
es
ear
ch
er
s
i
n
[
23]
m
o
d
el
ed
a D
2
D
-
Q
oS
r
out
i
ng.
T
he
y a
s
s
i
gne
d t
he
Q
oS
i
n t
e
r
m
s
of
de
l
a
y,
ba
ndw
i
dt
h
,
an
d
p
ack
et
l
o
s
s
r
at
e
. T
h
e
r
o
u
t
i
n
g
p
at
h
w
as
al
l
o
cat
ed
acc
or
di
ng
t
o
dyna
m
i
c
e
nvi
r
o
nm
e
nt
s
.
M
or
e
ove
r
,
t
he
a
ut
ho
r
s
i
n
[
2
4]
de
ve
l
ope
d a
B
a
ye
s
i
a
n f
r
a
m
e
w
or
k
t
o a
s
s
i
gn t
he
a
m
ount
of
pe
r
m
e
a
bl
e
w
a
t
e
r
i
n a
por
ous
s
t
r
uc
t
ur
e
us
i
ng c
l
us
t
e
r
i
ng a
nd
ge
om
e
t
r
y va
l
ue
s
of
t
he
por
e
-
th
r
o
a
t n
e
tw
o
r
k
.
S
e
v
e
r
a
l c
lu
s
te
r
in
g
c
r
ite
r
ia
w
e
r
e
us
e
d (
e
dge
be
t
w
e
e
nne
s
s
,
s
hor
t
r
a
ndom
w
a
l
ks
,
a
nd gr
e
e
dy a
nd m
ul
t
i
-
le
v
e
l mo
d
u
la
r
ity
o
p
t
im
iz
a
tio
n
).
T
h
ey
’
v
e cr
eat
ed
a
m
i
cr
o
n
et
w
o
r
k
s
d
at
ab
as
e
fo
r
m
i
c
ro
-
s
c
a
l
e
por
ous
s
t
r
uc
t
ur
e
s
t
o
be
t
he
pr
i
m
a
r
y
i
npu
t
f
o
r
t
he
B
a
ye
s
i
a
n m
e
t
hod.
I
n
[
2
5]
,
t
he
a
ut
hor
c
onc
l
ude
d
c
ont
i
nui
t
i
e
s
a
nd di
s
c
ont
i
nui
t
i
e
s
,
i
n
bot
h
t
he
r
e
a
l
i
z
a
t
i
on of
t
e
c
hnol
ogy a
nd s
c
i
e
nc
e
a
s
w
e
l
l
a
s
on
t
he
r
ol
e
of
e
t
hi
c
s
i
n t
hi
s
r
e
vol
ut
i
ona
r
y
pr
oc
e
s
s
.
H
e
a
l
s
o c
onc
l
u
de
d t
ha
t
hum
a
n va
l
ue
s
m
us
t
be
i
nc
or
por
a
t
e
d w
i
t
h t
e
c
hnol
og
y s
houl
d a
nd s
houl
d be
e
nr
i
c
he
d f
r
o
m
s
e
ve
r
a
l
c
ul
t
u
r
a
l
a
r
e
a
s
.
T
he
a
ut
hor
s
of
[
26]
us
e
d s
ta
tis
tic
a
l te
s
ts
to
ad
d
r
es
s
t
h
e
b
e
st
me
th
o
d
o
f
e
s
tima
tio
n
.
T
e
s
ts
lik
e
t
he
L
a
gr
a
nge
mu
ltip
lie
r
B
r
eu
s
ch
-
P
ag
an
t
es
t
,
t
h
e F
t
es
t
,
an
d
t
h
e H
au
s
m
an
t
es
t
w
er
e u
s
ed
a
nd
r
es
u
l
t
ed
i
n
a p
an
el
d
at
a
pr
ot
ot
ype
w
ith
d
i
ffe
re
n
t
e
f
f
e
c
t
s
f
or
i
nd
i
vi
dua
l
s
.
I
n
[
2
7]
,
th
e
w
r
ite
r
’
s
a
na
l
ys
i
s
a
i
m
s
t
o de
ve
l
op a
ba
s
e
f
or
de
c
i
s
i
on
-
m
a
ki
ng
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
O
pt
i
m
al
r
e
s
our
c
e
al
l
oc
at
i
on f
or
r
out
e
s
e
l
e
c
t
i
on i
n ad hoc
ne
t
w
or
k
s
(
M
ar
w
a K
.
F
ar
han
)
1199
in
te
r
ms
o
f
m
e
rg
e
rs
a
nd
a
c
qui
s
i
t
i
ons
f
o
r
m
a
nuf
a
c
t
ur
i
ng c
om
pa
ni
e
s
.
T
hi
s
c
a
n be
done
by
pe
r
f
or
m
i
ng a
s
c
or
e
c
a
r
d
m
ode
l
t
o a
l
l
ow
t
he
pot
e
nt
i
a
l
a
c
qui
r
e
r
t
o
a
ppl
y a
n
o
ve
r
a
l
l
a
na
l
ys
i
s
of
t
he
e
xi
s
t
i
ng
da
t
a
.
I
t
a
l
s
o
r
uns
a
n i
ns
i
ght
f
or
t
he
f
ut
ur
e
us
i
ng a
s
t
a
nda
r
di
z
e
d a
nd e
f
f
i
c
i
e
nt
a
ppr
oa
c
h t
o pos
i
t
i
ve
l
y a
f
f
e
c
t
t
he
a
c
hi
e
ve
-
r
at
e o
f
m
er
g
er
s
an
d
a
c
q
u
is
itio
n t
r
a
ns
a
c
t
i
ons
.
F
i
na
l
l
y,
t
he
a
ut
hor
s
o
f
[
28]
c
om
pos
e
d
cel
l
u
l
ar
ne
t
w
or
ks
of
D
2D
pa
i
r
s
w
he
r
e
r
e
l
a
ys
a
r
e
a
r
r
a
nge
d
i
n
c
lu
s
te
r
s
.
T
h
e
y
in
v
e
s
tig
a
te
D
2
D
c
o
mmu
n
ic
a
tio
n
o
p
tima
l
r
o
u
tin
g
in
th
e
e
x
is
te
n
c
e
o
f
in
te
r
f
e
r
e
n
c
e
.
O
p
tima
l
r
out
i
ng w
a
s
i
nc
l
ude
d t
o
s
e
l
e
c
t
t
he
hi
ghe
s
t
en
d
-
to
-
e
nd S
I
N
R
pa
t
h.
2.
R
ES
EA
R
C
H
M
ETH
O
D
I
n t
hi
s
s
e
c
t
i
on,
w
e
p
r
e
s
e
n
t th
e
s
y
s
te
m
mo
d
e
l a
n
d
f
o
r
mu
la
tio
n
o
f
th
e
p
r
o
b
le
m.
I
n a
not
he
r
m
e
a
ni
ng,
t
he
f
or
m
ul
a
t
i
on
of
t
he
pr
e
s
e
nt
e
d pr
ob
l
e
m
i
s
i
nt
r
odu
c
e
d.
T
he
L
a
gr
a
nge
mu
l
tip
lie
r
m
e
t
hod
i
s
a
dopt
e
d i
n
t
hi
s
f
or
m
ul
a
t
i
on a
s
a
n
obj
e
c
t
i
ve
f
una
t
i
on
a
nd c
ons
t
r
a
i
nt
s
.
2
.1
.
S
y
s
t
e
m
m
o
d
el
W
e
c
ons
i
de
r
a
M
A
N
E
T
t
ha
t
i
s
c
om
pos
e
d of
a
s
e
t
of
node
s
=
{
1¸
2¸
3¸
…
¸
}
t
h
at
ar
e
co
n
n
ect
ed
by a
s
e
t
of
a
va
i
l
a
bl
e
l
i
nks
ℒ
=
{
1¸
2¸
3¸
…
¸
ℒ
}
an
d
r
ep
r
es
en
t
s
a d
ev
i
ce
-
to
-
de
vi
c
e
c
om
m
uni
c
a
t
i
on ove
r
t
he
ad
-
hoc
e
nvi
r
onm
e
nt
.
E
a
c
h s
our
c
e
node
S
r
c
e
m
i
t
s
one
f
l
ow
t
o
t
he
de
s
t
i
na
t
i
on node
D
e
s
t
us
i
ng one
or
m
or
e
of
t
he
a
va
i
l
a
bl
e
l
i
nks
i
n
r
out
i
ng.
D
ue
t
o
t
he
f
r
e
que
nt
upda
t
e
s
i
n t
opol
ogi
e
s
i
n
t
he
ne
t
w
or
k
,
t
he
r
out
i
ng
m
e
t
hod
ne
e
ds
t
o be
i
m
p
r
ove
d a
nd
opt
i
m
i
z
e
d
a
c
c
or
d
i
ngl
y.
A
s
s
um
e
t
ha
t
t
he
s
our
c
e
node
pe
r
f
o
r
m
s
da
t
a
t
r
a
ns
m
i
s
s
i
on w
i
t
h
ꝓ
w
at
t
p
o
w
er
o
v
er
an
ω
ba
ndw
i
dt
h,
a
nd
s
ubj
e
c
t
t
o
σ
w
a
t
t
t
r
a
ns
m
i
s
s
i
on noi
s
e
.
W
e
i
nc
l
ude
a
f
a
di
ng
f
a
c
t
or
t
o
r
e
f
l
e
c
t
t
he
e
f
f
e
c
t
of
c
ha
nne
l
f
a
di
ng
,
a
t
w
hi
c
h t
he
t
r
a
ns
m
i
t
t
i
ng node
r
out
e
s
da
t
a p
ack
et
s
t
o
t
h
e r
ecei
v
i
n
g
n
o
d
e.
E
a
c
h of
t
he
node
s
i
nj
e
c
t
da
t
a
pa
c
ke
t
s
i
nt
o t
he
ne
t
w
or
k w
i
t
h s
pe
c
i
f
i
c
pow
e
r
ꝓ
a
nd i
s
e
xpos
e
d t
o a
n
a
m
ount
of
noi
s
e
σ
.
W
e
a
s
s
i
gne
d a
s
ur
ve
yi
ng
pr
oc
e
dur
e
,
f
or
e
a
c
h of
t
he
hops
,
t
ha
t
e
xpl
o
r
e
a
l
l
a
va
i
l
a
bl
e
ℒ
l
i
nks
c
onne
c
t
e
d t
o
t
he
c
or
r
e
s
pondi
ng hop.
T
hi
s
s
ur
ve
yi
ng pr
oc
e
dur
e
a
ddr
e
s
s
e
s
pa
r
a
m
e
t
e
r
s
t
ha
t
m
e
a
s
ur
e
t
he
s
i
gni
f
i
c
a
nc
e
of
e
ve
r
y c
onne
c
t
e
d l
i
nk r
e
l
a
t
e
d t
o t
ha
t
hop.
T
he
p
r
oc
e
s
s
of
s
ur
ve
yi
ng e
ndur
e
s
c
a
l
c
ul
a
t
i
ons
f
or
t
he
b
it r
a
te
Ʀ
to
b
e
tr
a
n
s
mitte
d
b
y
,
th
e
to
ta
l d
e
la
y
δ
c
ons
um
e
d,
a
nd t
he
pr
oba
bi
l
i
t
y of
pa
c
ke
t
l
os
s
ψ
unde
r
t
a
ke
n by
t
h
e s
p
eci
f
i
ed
l
i
n
k
.
F
u
r
t
h
er
m
o
r
e
,
w
e at
t
ai
n
an
o
b
j
ect
i
v
e f
u
n
ct
i
o
n
ℒ
.
ℱ
co
m
p
u
t
at
i
o
n
f
o
r
each
o
f
t
h
es
e co
n
n
ect
ed
l
i
n
k
s
t
h
at
r
ef
l
ect
t
h
e
s
at
i
s
f
act
i
o
n
o
f
a s
o
u
r
ce
w
i
t
h
t
h
e
r
es
o
u
r
ce al
l
o
cat
i
o
n
.
B
as
ed
o
n t
he
a
bove
-
m
e
nt
i
one
d
c
om
put
a
t
i
on,
e
a
c
h hop i
s
r
e
s
pons
i
bl
e
f
or
s
e
l
e
c
t
i
ng t
he
opt
i
m
um
l
i
nk t
ha
t
l
e
a
ds
t
o t
he
de
s
t
i
na
t
i
on n
ode
D
e
s
t
r
e
qui
r
e
d by
t
he
s
our
c
e
node
S
r
c
.
T
he
de
c
i
s
i
on i
s
m
a
de
ba
s
e
d on t
he
m
a
xi
m
um
s
c
or
e
d obj
e
c
t
i
ve
f
un
c
t
i
on a
nd
its
c
o
r
r
e
la
te
d
lin
k
i
s
e
le
c
te
d
.
O
u
r
f
o
r
mu
la
te
d
o
p
timiz
a
tio
n
mo
d
e
l
in
s
ect
i
o
n
3
.
2 i
s
ba
s
e
d on
hop
-
by
-
h
op
opt
i
m
i
z
a
t
i
on c
ont
r
ol
,
t
he
r
e
f
or
e
,
i
t
s
s
ui
t
a
bl
e
f
or
s
c
e
na
r
i
os
of
di
ve
r
s
e
r
out
e
s
.
T
a
bl
e
1 s
um
m
a
r
i
z
e
s
t
he
m
a
i
n
not
a
t
i
ons
a
nd t
he
i
r
c
o
r
r
e
s
pondi
ng de
f
i
ni
t
i
on
t
ha
t
a
r
e
us
e
d t
hr
ough
out
t
he
pa
pe
r
.
T
a
bl
e
1.
L
i
s
t
of
not
a
t
i
ons
S
ym
bol
S
e
ma
n
tic
s
ꝓ
P
o
w
e
r
a
v
a
ila
b
le
f
o
r
d
a
ta
tr
a
n
s
mis
s
io
n
ω
B
a
ndw
i
dt
h a
l
l
oc
a
t
e
d f
or
t
he
ne
t
w
or
k
σ
N
oi
s
e
pow
e
r
ge
ne
r
a
t
e
d by t
he
c
ha
nne
l
R
a
ndom
va
r
i
a
bl
e
r
e
pr
e
s
e
nt
s
t
h
e
c
ha
nne
l
f
a
di
ng
ℒ
N
um
be
r
of
a
va
i
l
a
bl
e
l
i
nks
ℒ
.
ℱ
L
a
gr
a
ngi
a
n obj
e
c
t
i
ve
f
unc
t
i
on
Ʀ
B
it r
a
te
c
a
lc
u
la
te
d
f
o
r
tr
a
n
s
mis
s
io
n
δ
T
o
t
al
d
el
ay
cal
cu
l
at
ed
ψ
P
r
oba
bi
l
i
t
y of
pa
c
ke
t
s
l
os
s
c
a
l
c
ul
a
t
e
d
λ
,
μ
L
a
g
r
a
n
g
e
mu
ltip
lie
r
s
φ
L
e
ngt
h of
t
he
phys
i
c
a
l
m
e
di
um
ζ
P
r
opa
ga
t
i
on s
pe
e
d of
t
he
m
e
di
um
α
P
ack
et
av
er
ag
e ar
r
i
v
al
r
at
e
N
um
be
r
of
N
ode
s
i
n t
he
a
d
-
hoc
ne
t
w
or
k
2.
1.
1.
T
r
a
n
sm
i
ssi
o
n
ra
t
e
We
c
ons
i
gn
ꝓ
de
not
e
t
he
pow
e
r
us
e
d
f
or
t
r
a
ns
m
i
s
s
i
on ove
r
a
ba
ndw
i
dt
h
ω
o
f
th
e
lin
k
ℒ
,
an
d
l
et
r
ef
l
ect
f
ad
i
n
g
f
act
o
r
,
w
h
er
eas
σ
i
s
t
he
noi
s
e
pow
e
r
.
T
he
r
e
l
a
t
i
ons
hi
p be
t
w
e
e
n t
he
t
r
a
ns
m
i
s
s
i
on r
a
t
e
a
nd a
l
l
oc
a
t
e
d
pow
e
r
i
n
f
a
di
ng c
ha
nne
l
s
t
ypi
c
a
l
l
y a
c
onc
a
ve
f
unc
t
i
on.
T
he
t
r
a
ns
m
i
s
s
i
on r
a
t
e
Ʀ
e
xpr
e
s
s
e
d be
l
ow
[
29]
:
Ʀ
=
ω
log
2
1
+
ꝓ
σ
(
1
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
19
, N
o
.
4
,
A
ugus
t
2021
:
1197
-
1207
1200
2.
1.
2.
T
o
t
al
no
da
l
de
l
a
y
N
ow
w
e
i
nvol
ve
t
he
de
l
a
y pr
ope
r
t
y t
o be
r
e
f
l
e
c
t
e
d
i
n t
he
c
ons
t
r
a
i
nt
s
of
t
he
obj
e
c
t
i
ve
f
unc
t
i
on.
A
s
pa
c
ke
t
s
s
t
a
r
t
t
he
i
r
j
our
ne
y f
r
om
t
he
s
our
c
e
node
S
r
c
t
hr
ough a
m
u
l
t
i
-
hop r
e
a
c
hi
ng t
he
de
s
t
i
na
t
i
on no
de
D
e
s
t
.
Wh
er
eas
at
each
n
o
d
e,
p
ack
et
s
en
d
u
r
e a
n
o
d
al
d
el
a
y t
ha
t
c
ons
i
s
t
s
of
s
e
ve
r
a
l
t
ype
s
o
f
de
l
a
ys
a
l
ong
t
he
pa
t
h.
T
he
m
os
t
i
nf
l
ue
nt
i
a
l
a
r
e
t
he
t
r
a
ns
m
i
s
s
i
on de
l
a
y,
p
r
opa
g
a
t
i
on de
l
a
y,
a
nd
que
ui
ng
de
l
a
y,
a
nd
t
oge
t
he
r
a
c
c
um
ul
a
t
e
t
o
gi
ve
a
t
ot
a
l
noda
l
de
l
a
y
[
30]
.
D
e
l
a
y i
s
a
n i
n
f
l
ue
nt
i
a
l
de
s
i
gn c
ons
i
de
r
a
t
i
on i
n s
om
e
r
e
a
l
-
time
a
p
p
lic
a
tio
n
s
.
T
h
u
s
,
t
he
t
ot
a
l
noda
l
de
l
a
y
δ
ove
r
a
l
i
nk
ℒ
i
s e
x
p
r
e
sse
d
a
s
[
30]
:
δ
ℒ
n
o
d
al
=
δ
ℒ
t
r
a
n
sm
i
ssi
o
n
+
δ
ℒ
p
r
o
p
ag
at
i
o
n
+
δ
ℒ
q
u
e
ue
(
2
)
w
h
er
e
a
s
t
he
a
m
ount
of
t
i
m
e
r
e
qui
r
e
d
t
o pus
h a
l
l
of
t
he
pa
c
ke
t
’
s
bi
t
s
i
nt
o t
he
l
i
nk
ℒ
r
e
p
r
e
s
e
n
ts
th
e
tr
a
n
s
mis
s
io
n
de
l
a
y.
I
t
de
pe
nds
on t
he
l
e
ngt
h
of
t
he
pa
c
ke
t
of
ᴌ
b
it
s
a
t a
tr
a
n
s
mis
s
io
n
r
a
te
Ʀ
of
t
he
l
i
nk
a
s
s
how
n be
l
ow
[
30]
:
δ
t
r
a
n
sm
i
ssi
o
n
=
ᴌ
Ʀ
(
3
)
a
nd
t
he
t
i
m
e
i
m
pos
e
d t
o
s
pr
e
a
d
f
r
om
t
he
be
gi
nni
ng
of
t
he
l
i
nk
ℒ
t
o
t
he
ne
xt
-
hop
e
xhi
bi
t
s
t
he
pr
opa
ga
t
i
on
de
l
a
y.
A
s
b
i
t
s
ar
e t
r
an
s
m
i
t
t
ed
o
v
er
a d
i
s
t
an
ce
φ
be
t
w
e
e
n t
w
o hops
a
t
a
phys
i
c
a
l
m
e
di
um
w
i
t
h a
pr
opa
ga
t
i
on s
pe
e
d
ζ
on a
l
i
nk
ℒ
.
T
he
p
r
opa
ga
t
i
on de
l
a
y i
s
w
r
i
t
t
e
n a
s
[
3
0]
:
δ
p
r
o
p
ag
at
i
o
n
=
φ
ζ
(
4
)
a
s
pa
c
ke
t
s
s
uf
f
e
r
out
put
bu
f
f
e
r
que
ui
ng
d
e
la
y
w
h
ic
h
is
th
e
p
e
r
io
d
o
f
w
a
itin
g
to
b
e
t
r
a
n
s
mitte
d
o
n
to
th
e
lin
k
ℒ
.
S
uc
h de
l
a
y i
s
va
r
i
a
bl
e
a
nd
r
e
l
i
e
s
on t
he
c
onge
s
t
i
on l
e
ve
l
o
f
t
he
ne
t
w
or
k
.
U
nl
i
ke
p
r
e
vi
ous
l
y m
e
nt
i
one
d de
l
a
ys
,
t
he
que
ui
ng de
l
a
y va
r
i
e
s
f
r
om
one
pa
c
ke
t
t
o a
not
he
r
.
A
s
pa
c
ke
t
s
a
r
r
i
ve
a
t
a
n
em
p
t
y
q
u
eu
e at
t
h
e s
am
e t
i
m
e,
t
h
e
f
i
r
s
t
p
ack
et
s
u
f
f
er
s
zer
o
q
u
eu
i
n
g
d
el
ay
s
,
w
h
i
l
e t
h
e
l
as
t
p
ack
et
s
u
f
f
er
s
q
u
eu
i
n
g
d
el
ay
as
i
t
w
ai
t
s
f
o
r
t
h
e ear
l
i
er
pa
c
ke
t
s
t
o be
t
r
a
ns
m
i
t
t
e
d.
T
he
r
e
f
or
e
,
a
n
a
ve
r
a
ge
que
ui
ng de
l
a
y i
s
c
ons
i
de
r
e
d.
i
t
e
xpr
e
s
s
e
d by t
he
l
e
n
gt
h o
f
t
h
e
p
ack
et
o
f
ᴌ
b
i
t
s
an
d
t
h
e av
er
ag
e r
at
e at
w
h
i
ch
p
ack
et
s
ar
r
i
v
e at
t
h
e q
u
eu
e
α
at
a
Ʀ
t
r
a
n
sm
i
ssi
o
n
r
a
t
e
a
s
f
ol
l
ow
s
[
30]
:
δ
q
u
e
ue
=
ᴌ
×
α
Ʀ
(
5
)
T
h
er
ef
o
r
e,
t
h
e t
o
t
al
d
el
ay
r
ep
r
es
en
t
s
t
h
e s
u
m
o
f
al
l
as
s
h
o
w
n
[
30]
:
δ
n
o
d
al
=
ᴌ
Ʀ
+
φ
ζ
+
ᴌ
∗
α
Ʀ
(
6
)
δ
n
o
d
al
=
ᴌ
(
1
+
α
)
Ʀ
+
φ
ζ
(
7
)
2.
1.
3.
P
r
ob
ab
i
l
i
t
y o
f
p
ac
k
e
t
l
os
s
A
not
he
r
f
e
a
t
ur
e
t
o be
i
nvol
ve
d i
n
t
he
c
ons
t
r
a
i
nt
s
o
f
t
he
ob
j
e
c
t
i
ve
f
unc
t
i
on
.
W
he
r
e
t
he
pr
oba
bi
l
i
t
y o
f
p
ack
et
l
o
s
s
ψ
can
b
e
fo
rm
e
d
as
a
f
unc
t
i
on
of
ꝓ
t
r
a
ns
m
i
s
s
i
on pow
e
r
u
tiliz
e
d
i
n
s
e
ndi
ng pa
c
ke
t
s
ove
r
a
l
i
nk
ℒ
.
A
s
t
h
e p
ack
et
l
o
s
s
i
s
a
r
at
i
o
o
f
r
ecei
v
ed
o
v
e
r
s
e
nt
va
l
ue
s
w
i
t
h
e
xpone
nt
be
ha
vi
or
;
i
n
t
h
i
s
c
a
s
e
,
i
t
’
s
t
he
t
r
a
ns
m
i
s
s
i
on pow
e
r
ꝓ
,
s
u
b
t
r
act
ed
f
r
o
m
u
n
i
t
y
t
o
m
eas
u
r
e t
h
e p
r
o
b
ab
i
l
i
t
y
.
We as
s
u
m
ed
t
h
at
a p
ack
et
i
s
r
ecei
v
ed
e
rro
r
-
f
r
ee w
i
t
h
Ʀ
t
r
a
ns
m
i
s
s
i
on r
a
t
e
a
nd ba
ndw
i
dt
h
ω
ov
e
r
t
he
l
i
nk
ℒ
.
Ɛ
[
H
]
r
ep
r
es
en
t
s
t
h
e ex
p
ect
ed
ch
an
n
el
s
t
at
e
t
h
at
i
s
f
i
x
ed
f
o
r
each
p
ack
et
an
d
i
s
r
eal
i
zed
a
t
th
e
t
r
a
n
s
mitte
r
s
id
e
[
31]
,
[
3
2]
:
=
1
−
−
/
ꝓ
(
8
)
=
×
Ɛ
[
]
×
(
2
Ʀ
/
−
1
)
(
9
)
2.
2.
P
r
o
b
l
e
m fo
r
mu
l
a
ti
o
n
N
o
w
w
e
f
o
r
mu
la
te
th
e
o
p
timu
m
p
a
th
mu
lti
-
o
b
j
ect
i
v
e p
r
o
b
l
em
.
We f
i
r
s
t
d
ef
i
n
e a v
ar
i
ab
l
e t
h
at
ai
m
s
t
o
ma
x
imiz
e
th
e
b
it r
a
te
a
n
d
min
i
miz
e
th
e
to
ta
l n
e
tw
o
r
k
d
e
la
y
a
s
w
e
ll a
s
p
a
c
k
e
t lo
s
s
.
T
he
pr
i
m
e
obj
e
c
t
i
ve
f
unc
t
i
on
is
b
it
r
a
te
ma
x
imiz
a
tio
n
f
o
r
num
be
r
o
f
a
va
i
l
a
bl
e
pa
t
hs
i
n a
ne
t
w
or
k
a
nd
c
a
n be
e
xpr
e
s
s
e
d m
a
t
he
m
a
t
i
c
a
l
l
y by
:
m
ax
∑
Ʀ
i
=
1
(
10
)
T
h
a
t s
u
b
je
c
t to
th
e
c
o
n
s
tr
a
in
ts
o
f
to
ta
l
d
e
la
y
min
i
miz
a
tio
n
a
n
d
p
a
c
k
e
t lo
s
s
min
im
iz
a
tio
n
f
o
r
e
a
c
h
p
at
h
s
an
d
can
b
e ch
ar
act
er
i
zed
b
y
:
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
O
pt
i
m
al
r
e
s
our
c
e
al
l
oc
at
i
on f
or
r
out
e
s
e
l
e
c
t
i
on i
n ad hoc
ne
t
w
or
k
s
(
M
ar
w
a K
.
F
ar
han
)
1201
m
in
∑
δ
i
+
m
in
∑
ψ
i
=
1
=
1
(
11
)
U
tiliz
in
g
th
e
mu
lt
i
-
obj
e
c
t
i
ve
a
ppr
oa
c
h t
o m
ode
l
t
hi
s
i
de
a
t
o ge
t
:
m
ax
∑
Ʀ
i
=
1
−
m
in
∑
δ
i
−
m
in
∑
ψ
i
=
1
=
1
(
12
)
B
y a
ppl
yi
ng t
he
L
a
gr
a
nge
opt
i
m
i
z
a
t
i
on m
e
t
hod t
o t
he
m
ode
l
i
de
a
t
o e
ns
ur
e
t
ha
t
t
he
gr
a
di
e
nt
of
bi
t
r
a
t
e
i
s
pr
opor
t
i
ona
l
t
o t
he
gr
a
di
e
nt
s
of
t
he
t
ot
a
l
de
l
a
y a
nd pa
c
ke
t
l
os
s
.
T
he
pr
opor
t
i
ona
l
i
t
y va
r
i
a
bl
e
s
a
r
e
c
a
l
l
e
d
L
a
gr
a
nge
m
ul
t
i
pl
i
e
r
s
a
nd a
r
e
de
not
e
d
by
λ
a
nd
μ
.
T
he
L
a
gr
a
nge
opt
i
m
i
z
a
t
i
on f
unc
t
i
on
i
s
e
xpr
e
s
s
e
d a
s
:
∇
bi
t
r
at
e
=
λ
∇
to
ta
l
d
el
ay
+
μ
∇
p
ac
k
et
los
s
(
13
)
ℒ
.
ℱ
=
∇
bi
t
r
at
e
−
λ
∇
to
ta
l
d
el
ay
−
μ
∇
p
ac
k
et
los
s
(
14
)
T
he
c
ha
l
l
e
nge
i
s
t
o
f
i
nd a
n
e
qua
t
i
on f
o
r
m
ul
a
f
or
a
l
l
t
hr
e
e
,
t
he
ob
j
e
c
t
i
ve
a
nd c
ons
t
r
a
i
nt
s
,
t
ha
t
c
ons
i
s
t
of
c
om
m
on pa
r
a
m
e
t
e
r
s
.
T
he
r
e
a
s
on be
hi
nd t
ha
t
i
s
be
c
a
us
e
w
h
e
n t
a
ki
ng pa
r
t
i
a
l
de
r
i
va
t
i
ve
f
or
e
qua
t
i
ons
w
i
t
h
p
a
r
a
me
te
r
in
c
o
mmo
n
w
ill
n
o
t
r
e
s
u
lt
in
a
n
il
v
al
u
e,
d
u
e
t
o
f
act
t
h
at
d
er
i
v
at
i
v
es
o
f
co
n
s
t
an
t
s
ar
e zer
o
.
T
h
o
s
e
c
om
m
on gr
ound pa
r
a
m
e
t
e
r
s
a
r
e
t
he
pow
e
r
(
ꝓ
)
a
nd ba
ndw
i
dt
h
(
ω
)
.
F
r
o
m
a
m
at
h
em
at
i
cal
p
er
s
p
ect
i
v
e,
t
h
e d
es
i
r
ed
L
ag
r
an
g
i
an
f
u
n
ct
i
o
n
i
s
ch
ar
act
er
i
zed
as
:
ℒ
ℱ
(
ꝓ
⸴
ω
)
=
∑
[
Ʀ
i
(
ꝓ
⸴
ω
)
]
=
1
−
∑
[
λ
i
δ
i
(
ꝓ
⸴
ω
)
]
=
1
−
∑
[
μ
i
ψ
i
(
ꝓ
⸴
ω
)
]
=
1
(
15)
w
h
er
e
Ʀ
(
ꝓ
⸴
ω
)
r
ep
r
es
en
t
t
h
e g
r
ad
i
en
t
s
o
f
a b
i
t
r
at
e as
a r
es
u
l
t
o
f
t
h
e
f
i
r
s
t
-
or
de
r
de
r
i
va
t
i
on
c
onc
e
r
ni
ng t
he
pow
e
r
(
ꝓ
)
a
nd ba
ndw
i
dt
h
(
ω
)
r
es
p
ect
i
v
el
y
.
M
o
r
eo
v
er
,
δ
(
ꝓ
⸴
ω
)
de
not
e
s
t
he
gr
a
di
e
nt
s
of
t
ot
a
l
de
l
a
y by t
a
ki
ng o
f
t
he
fi
rs
t
-
or
de
r
de
r
i
va
t
i
ve
of
t
ot
a
l
de
l
a
y
r
e
l
a
t
i
ve
t
o
t
he
pow
e
r
(
ꝓ
)
a
nd ba
ndw
i
dt
h
(
ω
)
r
es
p
ect
i
v
el
y
.
F
u
r
t
h
er
m
o
r
e
,
ψ
(
ꝓ
⸴
ω
)
r
e
pr
e
s
e
nt
t
he
g
r
a
di
e
nt
s
of
pa
c
ke
t
l
os
s
pr
oba
bi
l
i
t
y
a
s
a
r
e
s
ul
t
of
t
he
f
i
r
s
t
-
or
de
r
de
r
i
va
t
i
on c
onc
e
r
ni
ng
t
he
pow
e
r
(
ꝓ
)
a
nd ba
ndw
i
dt
h
(
ω
)
r
es
p
ect
i
v
el
y
.
N
ow
,
s
e
t
t
i
ng up
t
he
L
a
gr
a
nge
mu
ltip
lie
r
s
obj
e
c
t
i
ve
f
unc
t
i
on.
T
he
f
i
r
s
t
s
t
e
p i
s
t
o c
ons
t
r
uc
t
t
he
obj
e
c
t
i
ve
f
unc
t
i
on by
pr
e
pa
r
i
ng
t
he
f
i
r
s
t
de
r
i
va
t
i
ve
s
f
or
t
he
m
a
i
n f
unc
t
i
on
(
bi
t
r
a
t
e
)
a
nd t
he
c
ons
t
r
a
i
nt
s
(
t
ot
a
l
d
e
l
a
y a
nd
pa
c
ke
t
l
os
s
pr
oba
bi
l
i
t
y)
c
onc
e
r
ni
ng
ꝓ
,
a
nd c
onc
e
r
ni
n
g
ω
as
a s
eco
n
d
s
t
ep
.
∂
ℒ
.
ℱ
∂
ꝓ
=
∑
∂
Ʀ
i
∂
ꝓ
=
1
−
∑
λ
i
∂
δ
i
∂
ꝓ
=
1
−
∑
μ
i
∂
ψ
i
∂
ꝓ
=
1
(
15
)
∂
ℒ
.
ℱ
∂
ω
=
∑
∂
Ʀ
i
∂
ω
=
1
−
∑
λ
i
∂
δ
i
∂
ω
=
1
−
∑
μ
i
∂
ψ
i
∂
ω
=
1
(
16
)
2.
2.
1.
T
r
an
s
m
i
s
s
i
on
r
at
e
A
s
s
um
i
ng t
he
e
qua
t
i
on i
n
[
2
9]
f
i
ndi
ng
th
e
p
a
r
tia
l
d
e
r
iv
a
tiv
e
o
f
Ʀ
c
onc
e
r
ni
ng
ꝓ
,
ω
:
∂
Ʀ
∂
ꝓ
=
1
1
+
ꝓ
σ
×
ω
×
σ
×
L
n
2
(
17
)
∂
Ʀ
∂
ω
=
(
−
ꝓ
σ
×
ω
1
+
ꝓ
σ
×
ω
×
L
n
2
)
)
+
(
log
2
1
+
ꝓ
σ
×
ω
)
(
18
)
2.
2.
2.
T
ot
al
n
od
al
d
e
l
ay
C
ons
i
de
r
i
ng t
he
e
qua
t
i
on i
n
[
30]
,
T
o
ta
l d
e
la
y
is
t
r
e
a
te
d
a
s
th
e
f
ir
s
t c
o
n
s
tr
a
in
t to
th
e
o
b
je
c
tiv
e
.
T
h
e
c
ons
t
r
a
i
nt
s
houl
d be
e
qua
l
t
o
z
e
r
o
by
m
ovi
ng
pa
r
a
m
e
t
e
r
s
t
o
t
he
l
e
f
t
-
ha
nd s
i
de
of
t
he
e
qua
t
i
on a
s
s
ho
w
n be
l
ow
:
∑
δ
=
1
≤
∆
T
(
19
)
∑
δ
=
1
−
∆
T
≤
0
(
20
)
∂
δ
∂
ꝓ
=
∂
∂
ꝓ
ᴌ
(
1
+
α
)
×
ω
−
1
×
log
2
1
+
ꝓ
σ
×
ω
−
1
+
φ
ŝ
−
∆
T
(
21
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
19
, N
o
.
4
,
A
ugus
t
2021
:
1197
-
1207
1202
∂
δ
∂
ꝓ
=
−
ᴌ
(
1
+
α
)
ω
2
×
σ
×
L
n
2
×
1
+
ꝓ
σ
×
ω
lo
g
2
1
+
ꝓ
σ
×
ω
2
(
23)
∂
δ
∂
ω
=
ꝓ
×
ᴌ
×
(
1
+
α
)
σ
×
ω
3
×
L
n
2
1
+
ꝓ
σ
×
ω
lo
g
2
1
+
ꝓ
σ
×
ω
2
−
ᴌ
×
(
1
+
α
)
ω
2
×
lo
g
2
1
+
ꝓ
σ
×
ω
(
22
)
2.
2.
3.
P
r
ob
a
b
i
l
i
t
y
of
p
a
ck
et
l
o
s
s
P
a
c
ke
t
l
os
s
i
s
ha
ndl
e
d a
s
t
he
s
e
c
ond c
ons
t
r
a
i
nt
t
o t
he
bi
t
r
a
t
e
m
a
xi
m
i
z
a
t
i
on obj
e
c
t
i
ve
.
T
he
c
ons
t
r
a
i
nt
s
houl
d be
e
qua
l
t
o
zer
o
b
y
m
o
v
i
n
g
p
ar
am
et
er
s
t
o
t
h
e l
ef
t
-
ha
nd s
i
de
of
t
he
e
qua
t
i
on a
s
s
how
n be
l
ow
:
∑
ψ
=
1
≤
ψ
T
(
25)
∑
ψ
=
1
−
ψ
T
≤
0
(
26)
C
ons
i
de
r
i
ng t
he
e
qua
t
i
on i
n
[
7]
,
[
32]
,
P
a
c
ke
t
l
os
s
i
s
ha
ndl
e
d a
s
t
he
s
e
c
ond c
on
s
tr
a
in
t to
th
e
b
it
r
a
te
m
a
xi
m
i
z
a
t
i
on obj
e
c
t
i
ve
.
T
he
c
ons
t
r
a
i
nt
s
houl
d
be
e
qua
l
t
o z
e
r
o
by
m
ovi
ng pa
r
a
m
e
t
e
r
s
t
o
t
he
l
e
f
t
-
ha
nd s
i
de
of
t
he
e
qua
t
i
on a
s
s
how
n be
l
ow
:
∂
ψ
∂
ꝓ
=
∂
∂
ꝓ
[
1
−
ex
p
(
−
σ
×
ω
ꝓ
×
2
lo
g
2
1
+
ꝓ
σ
×
ω
+
σ
×
ω
ꝓ
)
]
−
ψ
T
(
23
)
∂
ψ
∂
ꝓ
=
−
ex
p
(
−
σ
×
ω
ꝓ
×
2
lo
g
2
1
+
ꝓ
σ
×
ω
+
σ
×
ω
ꝓ
)
×
σ
×
ω
ꝓ
×
2
l
o
g
2
1
+
ꝓ
σ
×
ω
−
σ
×
ω
ꝓ
−
1
ꝓ
+
ꝓ
2
σ
×
ω
(
24
)
∂
ψ
∂
ω
=
∂
∂
ω
[
1
−
ex
p
(
−
σ
×
ω
ꝓ
×
2
lo
g
2
1
+
ꝓ
σ
×
ω
+
σ
×
ω
ꝓ
)
]
−
ψ
T
(
25
)
∂
ψ
∂
ω
=
−
ex
p
(
−
σ
×
ω
ꝓ
×
2
lo
g
2
1
+
ꝓ
σ
×
ω
+
σ
×
ω
ꝓ
)
×
(
−
σ
×
ω
×
2
l
og
2
1
+
ꝓ
σ
×
ω
+
(
σ
×
ω
)
+
ꝓ
ꝓ
×
ω
+
ꝓ
2
σ
(
26
)
∂
ℒ
⋅
ℱ
∂
ω
=
∑
−
ꝓ
σ
×
ω
1
+
ꝓ
σ
×
ω
×
L
n
2
)
)
+
(
log
2
1
+
ꝓ
σ
×
ω
i
i
−
∑
λ
i
ꝓ
×
ᴌ
×
(
1
+
α
)
σ
×
ω
3
×
L
n
2
1
+
ꝓ
σ
×
ω
lo
g
2
1
+
ꝓ
σ
×
ω
2
−
ᴌ
×
(
1
+
α
)
ω
2
×
lo
g
2
1
+
ꝓ
σ
×
ω
i
i
+
∑
μ
i
ex
p
−
σ
×
ω
ꝓ
×
2
lo
g
2
1
+
ꝓ
σ
×
ω
+
σ
×
ω
ꝓ
×
i
(
−
σ
×
ω
×
2
l
og
2
1
+
ꝓ
σ
×
ω
+
(
σ
×
ω
)
+
ꝓ
ꝓ
×
ω
+
ꝓ
2
σ
i
(
27
)
T
he
ne
xt
s
t
e
p i
s
t
o s
ol
ve
t
he
s
y
s
t
e
m
of
L
a
gr
a
nge
mu
ltip
lie
r
s
'
obj
e
c
t
i
ve
f
unc
t
i
on e
qua
t
i
ons
.
T
ha
t
c
a
n
be
a
c
hi
e
ve
d by
s
e
t
t
i
ng t
hos
e
e
qua
t
i
ons
e
qua
l
t
o
z
e
r
o,
t
he
n
s
ol
ve
t
o
f
i
nd
λ
a
nd
μ
in
te
r
ms
o
f
a
ll o
th
e
r
p
a
r
a
me
te
r
s
,
an
d
cal
cu
l
at
e t
h
o
s
e m
u
l
t
i
p
l
i
er
s
f
o
r
each
p
at
h
.
F
i
na
l
l
y,
e
va
l
ua
t
i
ng
μ
a
nd
λ
a
nd t
he
n pl
uggi
ng t
hos
e
v
al
u
es
b
ac
k
i
nt
o t
he
obj
e
c
t
i
ve
f
unc
t
i
on a
nd
i
n ou
r
pr
og
r
a
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
O
pt
i
m
al
r
e
s
our
c
e
al
l
oc
at
i
on f
or
r
out
e
s
e
l
e
c
t
i
on i
n ad hoc
ne
t
w
or
k
s
(
M
ar
w
a K
.
F
ar
han
)
1203
2
.
3.
L
O
RDP
a
l
gor
i
t
h
m
I
n t
hi
s
s
e
c
t
i
on,
w
e
pr
ovi
de
a
n a
l
gor
i
t
hm
,
na
m
e
d
L
O
R
D
P
,
t
o c
om
put
e
t
he
opt
i
m
a
l
s
ol
ut
i
on f
or
t
he
obj
e
c
t
i
ve
f
unc
t
i
on,
w
hi
c
h
i
s
t
he
be
s
t
pos
s
i
bl
e
s
ol
ut
i
on a
s
t
he
pr
obl
e
m
c
a
n be
pr
ove
d.
D
e
vot
i
ng
t
he
n
um
e
r
i
c
a
l
e
va
l
ua
t
i
on of
t
he
pe
r
f
or
m
a
nc
e
o
f
t
he
L
a
gr
a
ngi
a
n
opt
i
m
i
zat
i
o
n
o
f
r
at
e,
d
el
ay
,
an
d
p
ack
et
l
os
s
a
l
gor
i
t
hm
(
L
O
R
D
P
)
d
es
i
g
n
ed
s
ch
em
es
as
co
m
p
ar
ed
t
o
ad
-
h
o
c
on
-
de
m
a
nd di
s
t
a
nc
e
ve
c
t
or
(
AODV)
.
F
i
g
u
r
e
1
s
h
o
ws
t
h
e
f
l
ow
c
ha
r
t
of
t
he
L
O
R
D
P
a
l
gor
i
t
hm
.
3.
R
ES
U
LTS
A
ND ANAL
YS
I
S
T
o
te
s
t th
e
o
b
ta
in
e
d
o
p
timiz
a
tio
n
f
o
r
mu
la
o
n
th
e
p
r
a
c
tic
a
l s
id
e
,
a
s
imu
la
tio
n
e
x
a
min
a
tio
n
is
p
e
r
f
o
r
me
d
i
n t
e
r
m
s
of
A
O
D
V
a
s
a
c
onve
nt
i
ona
l
m
e
t
hod a
nd
t
he
pr
opos
e
d L
O
R
D
P
a
l
gor
i
t
hm
.
F
or
e
a
s
e
of
r
e
a
di
ng,
w
e
c
ons
i
de
r
c
a
s
e
s
t
udi
e
s
i
n
i
m
pl
e
m
e
nt
i
ng
t
he
unde
r
l
yi
ng a
l
gor
i
t
h
m
s
.
W
e
a
s
s
um
e
d a
r
a
ndom
node
di
s
t
r
i
b
ut
i
on o
f
9 node
s
,
t
he
n a
dopt
t
he
s
our
c
e
a
nd
de
s
t
i
na
t
i
on no
de
s
a
nd t
he
i
r
de
di
c
a
t
e
d pa
t
h
a
s
s
how
n i
n
F
i
gur
e
2
f
o
r
cas
e
s
t
udy 1 a
nd
F
i
g
ur
e
3
f
or
c
a
s
e
s
t
udy 2.
F
i
gur
e
1.
L
O
R
D
P
a
l
gor
i
t
hm
F
i
gur
e
2.
L
O
R
D
P
vs
A
O
D
V
s
e
l
e
c
t
e
d pa
t
h f
r
om
s
our
c
e
node
7 t
o
de
s
t
i
na
t
i
on node
6
F
i
gur
e
3
.
L
OR
DP
v
s
AODV s
e
l
e
c
t
e
d
p
a
t
h
f
r
o
m
s
our
c
e
node
3 t
o de
s
t
i
na
t
i
on node
9
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
19
, N
o
.
4
,
A
ugus
t
2021
:
1197
-
1207
1204
3.
1.
C
a
se
st
u
d
y
1
C
ons
i
de
r
i
ng node
7 a
s
t
he
s
our
c
e
node
w
hi
l
e
node
6
a
s
t
he
de
s
t
i
na
t
i
on node
,
w
i
t
h
5.
5
m
e
t
e
r
s
e
a
c
h node
ap
ar
t
.
F
i
g
u
r
e 2
r
e
pr
e
s
e
nt
s
t
he
opt
i
m
a
l
pa
t
h s
e
l
e
c
t
e
d by L
O
R
D
P
ve
r
s
us
A
O
D
V
’
s
.
T
hi
s
f
i
gur
e
s
how
s
t
he
di
s
t
r
i
but
i
on of
node
s
i
n a
ddi
t
i
on
t
o
t
he
s
e
l
e
c
t
e
d pa
t
h.
3.
1.
1.
T
r
a
n
sm
i
ssi
o
n
ra
t
e
T
h
e v
al
u
es
o
f
b
i
t
r
at
e
ar
e
m
eas
u
r
ed
p
er
-
hop a
nd
a
s
a
ve
r
a
ge
,
a
s
one
c
a
n obs
e
r
ve
t
ha
t
t
he
L
OR
DP
a
c
hi
e
ve
d a
hi
ghe
r
bi
t
r
a
t
e
t
ha
n t
hos
e
c
or
r
e
s
pondi
n
g t
o
A
O
D
V
.
T
hi
s
i
s
p
r
o
v
e
d i
n
i
n
F
i
gur
e
s
4
a
nd
5
t
h
at
ar
e
r
ep
r
es
en
t
ed
as
b
ar
s
h
ap
es
.
S
uc
h
e
nha
nc
e
m
e
nt
de
l
i
ve
r
e
d by
m
ul
t
i
-
obj
e
c
t
i
ve
opt
i
m
i
z
a
t
i
on t
ha
t
a
s
s
um
e
d t
he
bi
t
r
a
te
a
s
th
e
ma
in
p
r
io
r
ity
.
3.
1.
2.
T
o
t
al
no
da
l
de
l
a
y
T
h
e
va
l
ue
s
of
t
ot
a
l
noda
l
de
l
a
y
ar
e
al
s
o
m
eas
u
r
ed
p
er
-
hop a
nd a
s
a
ve
r
a
ge
. O
n
e can
s
ee t
h
at
t
h
e L
O
R
D
P
a
c
hi
e
ve
d l
e
s
s
t
ot
a
l
noda
l
de
l
a
y
t
ha
n
t
hos
e
c
or
r
e
s
pondi
ng t
o
A
O
D
V
,
a
s
s
how
n
i
n
F
i
gur
e
s
6
a
nd
7
t
h
at
ar
e
r
ep
r
es
en
t
ed
as
b
ar
s
h
ap
e
.
T
ot
a
l
noda
l
de
l
a
y i
s
a
s
s
i
gne
d a
s
t
he
f
i
r
s
t
c
ons
t
r
a
i
nt
i
n
t
hi
s
pa
t
h opt
i
m
i
z
a
t
i
on.
F
i
gur
e
4
.
L
O
R
D
P
vs
A
O
D
V
R
a
t
e
va
l
ue
s
pe
r
hop
f
r
om
s
our
c
e
node
7
t
o de
s
t
i
na
t
i
on node
6
F
i
gur
e
5
.
L
O
R
D
P
v
s
A
O
D
V
av
er
ag
e r
at
e
f
r
o
m
s
our
c
e
node
7 t
o de
s
t
i
na
t
i
on node
6
F
i
gur
e
6
.
L
O
R
D
P
vs
A
O
D
V
t
ot
a
l
noda
l
de
l
a
y f
r
o
m
s
our
c
e
node
7 t
o de
s
t
i
na
t
i
on node
6
F
i
gur
e
7
.
L
O
R
D
P
vs
A
O
D
V
a
ve
r
a
ge
noda
l
de
l
a
y
f
r
om
s
our
c
e
node
7
t
o de
s
t
i
na
t
i
on node
6
3.
1.
3.
P
r
ob
ab
i
l
i
t
y o
f
p
ac
k
e
t
l
os
s
M
or
e
ove
r
,
bot
h a
l
go
r
i
t
hm
s
;
L
O
R
D
P
a
nd A
O
D
V
,
h
a
ve
a
c
hi
e
ve
d t
he
s
a
m
e
va
l
ue
s
how
n i
n F
i
gu
r
e
8
.
T
he
pr
oba
bi
l
i
t
y
of
pa
c
ke
t
l
os
s
i
s
a
s
s
i
gne
d a
s
t
he
s
e
c
ond c
ons
t
r
a
i
nt
i
n t
hi
s
pa
t
h opt
i
m
i
z
a
t
i
on
.
T
hi
s
i
s
d
ue
t
o
h
ig
h
imp
o
r
ta
n
c
e
o
f
th
is
p
a
r
a
me
te
r
in
s
e
le
c
tio
n
o
f
o
p
tim
al
p
at
h
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
O
pt
i
m
al
r
e
s
our
c
e
al
l
oc
at
i
on f
or
r
out
e
s
e
l
e
c
t
i
on i
n ad hoc
ne
t
w
or
k
s
(
M
ar
w
a K
.
F
ar
han
)
1205
3.
2.
C
as
e
s
t
u
d
y
2
C
ons
i
de
r
i
ng node
3 a
s
t
he
s
our
c
e
node
w
hi
l
e
node
9
a
s
t
he
de
s
t
i
na
t
i
on node
,
w
i
t
h
5.
5
m
e
t
e
r
s
e
a
c
h node
a
pa
r
t
.
F
i
gur
e
3
r
e
pr
e
s
e
nt
s
t
he
opt
i
m
a
l
pa
t
h s
e
l
e
c
t
e
d by L
O
R
D
P
ve
r
s
us
A
O
D
V
’
s
.
I
n a
ddi
t
i
on,
t
he
di
s
t
r
i
but
i
on
of
node
s
i
s
a
l
s
o s
how
n i
n t
h
i
s
f
i
gur
e
.
3.
2.
1.
T
r
an
s
m
i
s
s
i
on
ra
t
e
T
he
va
l
ue
s
of
bi
t
r
a
t
e
ar
e
m
e
a
s
ur
e
d pe
r
hop
. A
s
o
n
e can
o
b
s
er
v
e t
h
at
t
h
e L
O
R
D
P
ach
i
ev
ed
a h
i
g
h
er
bi
t
r
a
t
e
t
ha
n
t
hos
e
c
or
r
e
s
pondi
ng
t
o A
O
D
V
a
s
s
how
n i
n
F
i
gur
e
s
9
a
nd
10.
T
h
es
e f
i
g
u
r
es
ar
e
r
ep
r
es
en
t
ed
as
b
ar
s
h
ap
e t
o
s
h
o
w
t
h
e p
er
f
o
r
m
an
ce i
n
cl
ear
w
ay
.
3.
2.
2.
T
o
t
al
no
da
l
de
l
a
y
T
he
va
l
ue
s
of
t
ot
a
l
noda
l
de
l
a
y
ar
e
al
s
o
m
eas
u
r
e
d
p
er
-
hop a
nd a
s
a
ve
r
a
ge
.
I
t
can
b
e s
een
th
a
t th
e
L
O
R
D
P
ach
i
ev
ed
l
es
s
t
o
t
al
n
o
d
al
d
el
ay
t
h
a
n t
hos
e
c
or
r
e
s
pondi
ng
t
o
AODV
a
s
s
how
n i
n F
i
gu
r
e
s
11 a
nd 12.
T
h
e
to
ta
l n
o
d
a
l
d
e
la
y
imp
r
o
v
e
me
n
t d
u
e
to
th
e
f
ir
s
t
c
o
n
s
tr
a
in
t in
th
is
p
a
th
o
p
timiz
a
tio
n
.
F
i
gur
e
8
.
L
O
R
D
P
vs
A
O
D
V
pr
oba
bi
l
i
t
y
of
pa
c
ke
t
l
os
s
f
r
om
s
our
c
e
node
7 t
o
de
s
t
i
na
t
i
on node
6
F
i
g
u
r
e
9
.
L
OR
DP
v
s
AODV
R
a
t
e
va
l
ue
s
pe
r
hop
f
r
om
s
our
c
e
node
3
t
o de
s
t
i
na
t
i
on node
9
F
i
gur
e
10.
L
O
R
D
P
vs
A
O
D
V
a
ve
r
a
ge
r
a
t
e
f
r
om
s
our
c
e
node
3 t
o de
s
t
i
na
t
i
on node
9
F
i
gur
e
11.
L
O
R
D
P
vs
A
O
D
V
t
ot
a
l
noda
l
de
l
a
y
f
r
om
s
our
c
e
node
3
t
o de
s
t
i
na
t
i
on node
9
3.
2.
3.
P
r
ob
ab
i
l
i
t
y o
f
p
ac
k
e
t
l
os
s
O
n t
he
ot
he
r
ha
nd,
bot
h a
l
gor
i
t
h
m
s
;
L
O
R
D
P
a
nd A
O
D
V
,
ha
ve
a
c
hi
e
ve
d t
he
s
a
m
e
va
l
u
e
s
m
e
a
s
ur
e
d i
n
F
i
gur
e
8
f
o
r
c
a
s
e
s
t
udy 1
. W
h
er
e
t
he
pr
oba
bi
l
i
t
y
of
pa
c
ke
t
l
os
s
i
s
a
s
s
i
gne
d a
s
t
he
s
e
c
ond c
ons
t
r
a
i
nt
i
n
t
hi
s
pa
t
h
o
p
timiz
a
tio
n
.
T
h
e
si
m
i
l
ar
v
al
u
es
ar
e
g
i
v
en
t
o
ev
al
u
at
e t
h
e o
t
h
er
p
ar
am
et
er
s
an
d
t
h
ei
r
ef
f
ect
s
o
n
t
h
e o
p
t
i
m
i
zat
i
o
n
pr
obl
e
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
19
, N
o
.
4
,
A
ugus
t
2021
:
1197
-
1207
1206
F
i
gur
e
12.
L
O
R
D
P
vs
A
O
D
V
a
ve
r
a
ge
noda
l
de
l
a
y
f
r
om
s
our
c
e
node
3
t
o de
s
t
i
na
t
i
on node
9
4.
CO
NCL
US
I
O
N
T
hi
s
pa
pe
r
a
ddr
e
s
s
e
d r
out
i
ng
-
e
f
f
i
c
i
e
nt
s
c
he
dul
i
ng
pr
obl
e
m
s
ove
r
a
d
hoc
c
ha
nne
l
s
t
o
m
a
xi
m
i
z
e
t
he
bi
t
r
a
t
e
unde
r
t
he
t
ot
a
l
noda
l
de
l
a
y a
nd p
r
oba
bi
l
i
t
y o
f
pa
c
ke
t
l
os
s
c
ons
t
r
a
i
nt
s
.
w
e
pr
opos
e
d a
n opt
i
m
a
l
r
out
i
ng
a
l
gor
i
t
hm
t
ha
t
r
uns
i
n
be
t
w
e
e
n node
s
t
o
m
a
xi
m
i
z
e
t
he
bi
t
r
a
t
e
a
nd m
i
ni
m
i
z
e
t
he
noda
l
de
l
a
y a
nd
pa
c
ke
t
l
os
s
pr
oba
bi
l
i
t
y s
t
a
r
t
i
ng f
r
om
t
he
s
our
c
e
node
a
nd r
e
a
c
hi
ng t
he
de
s
t
i
na
t
i
on node
us
i
ng t
he
L
a
gr
a
nge
m
u
ltip
lie
r
m
e
t
hod.
T
he
opt
i
m
a
l
r
out
i
ng
r
e
pr
e
s
e
nt
e
d t
he
be
s
t
pos
s
i
bl
e
s
ol
ut
i
on t
ha
t
ve
r
i
f
i
e
s
t
he
obj
e
c
t
i
ve
f
unc
t
i
on.
W
e
co
n
s
i
d
er
t
h
e cas
e
w
h
er
e each
p
ack
et
i
s
s
en
t
o
v
er
a
n
ad
d
i
t
i
v
e w
h
i
t
e G
au
s
s
i
an
n
o
i
s
e ch
an
n
el
.
S
i
m
u
l
at
i
o
n
r
es
u
l
t
s
m
an
i
f
es
t
t
he
e
f
f
i
c
i
e
nc
y o
f
t
he
pr
opos
e
d a
l
gor
i
t
hm
s
i
n m
a
xi
m
i
z
a
t
i
on o
f
t
he
ob
j
e
c
t
i
ve
f
unc
t
i
on
.
R
EF
ER
EN
C
ES
[1
]
R.
R.
Ro
y
, “
Ha
nd
bo
ok
of
m
ob
ile
a
d
hoc
ne
tw
or
k
s f
or
m
obi
li
ty m
ode
ls
,
”
B
os
to
n,
MA:
Sp
ri
nge
r US
,
20
11
.
[2
]
L
.
M
c
na
m
a
r
a
,
B
.
P
a
sz
tor
,
N.
Tr
igo
ni,
S
.
W
a
ha
r
te
,
a
n
d S
.
S
tojm
e
n
ov
ic
,
“
M
obi
le
a
d hoc
ne
tw
or
k
in
g
:
c
ut
ti
ng
e
d
ge
dir
e
c
ti
on
s
,”
Wi
le
y
-
I
EE
E
Pr
e
ss
,
20
13.
[3
]
M
.
K.
F
a
r
ha
n a
nd M
.
S
.
Cro
o
ck
,
“
R
o
ut
in
g
Te
c
h
ni
que
s
S
tud
y
f
or
D
2D
in
M
a
ne
t
B
a
se
d
En
vir
onm
e
nt
:
A
S
ur
ve
y
a
nd
Ope
n I
ss
ue
s
,”
I
n
t.
J
.
I
nnov
.
E
ng.
Sc
i.
Re
s.
,
v
ol.
3,
n
o.
4
,
pp.
13
-
23
,
20
19
.
[4
]
P
.
M
a
se
k,
A.
M
ut
ha
n
na
,
a
nd J.
H
ose
k,
“
S
ui
ta
bi
li
ty
of
M
AN
ET
r
o
ut
in
g pr
ot
oc
o
ls f
or
the
ne
xt
-
ge
ne
r
a
t
io
n na
t
i
ona
l
se
c
ur
it
y a
nd
pu
bl
ic
sa
f
e
ty s
ys
te
m
s
,”
S
pr
in
ge
r I
nt.
P
ub
l.
Sw
i
tz
,
pp
.
2
42
-
25
3
,
A
ug 20
15,
do
i
: 10.
10
07
/9
78
-
3
-
31
9
-
103
53
-
2.
[5
]
V.
V.
M
a
ndha
r
e
,
V.
R
.
Thool,
a
n
d R
.
R
.
M
a
ntha
l
ka
r
,
“
QoS
R
ou
ti
ng e
n
ha
nc
e
m
e
nt u
si
ng m
e
ta
he
ur
is
tic
a
p
pr
oa
c
h i
n
m
obi
le
a
d
-
hoc
ne
t
wor
k
,”
vo
l.
11
0
,
pp.
1
80
-
1
91
,
D
e
c
.
20
16
,
d
oi
: 10.
1
01
6/
j.
c
om
ne
t.
20
16.
0
9.
0
23
.
[6
]
S
.
C
ha
vha
n a
nd P
.
Ve
n
ka
ta
r
a
m
,
“
Em
e
r
ge
nt
in
te
l
li
ge
nc
e
ba
se
d Q
oS
r
o
ut
in
g
in
M
AN
E
T
,”
Pr
oc
e
d
ia
C
om
pu
t.
Sc
i.
,
vol.
5
2,
no.
1
,
pp
.
6
59
-
6
64
,
20
15
,
do
i
:
10.
10
16
/j.
pr
oc
s.
2
015.
05.
0
68.
[7
]
M
.
Ta
nha
,
D.
S
a
jja
d
i,
F
.
Tong,
a
nd J.
P
a
n,
“
Di
sa
s
te
r
m
a
na
ge
m
e
nt a
n
d r
e
sp
on
se
f
or
m
ode
r
n c
e
l
lu
la
r
ne
t
wor
ks u
s
in
g
f
l
ow
-
ba
se
d m
ul
ti
-
ho
p de
v
ic
e
-
to
-
de
vic
e
c
om
m
un
ic
a
t
io
ns
,
”
20
16 I
EE
E 84
th Ve
hic
ul
ar T
e
c
hn
ol
ogy
C
o
nfe
re
nc
e
(
V
TC
-
Fa
ll
)
,
20
17
,
do
i:
10.
11
09
/V
TC
F
a
l
l.
20
16.
78
80
96
0.
[8
]
P
.
T.
A.
Qua
n
g,
K.
P
ia
m
r
a
t,
K.
D.
S
in
gh,
a
n
d C
.
Vi
ho,
“
Q
-
R
oS
A
: Q
oE
-
a
wa
r
e
r
o
ut
in
g
f
or
S
VC
vi
de
o str
e
a
m
in
g
o
ve
r
a
d
-
h
oc
ne
tw
or
k
s
,”
201
6 13
th I
E
EE An
nu.
C
on
su
m.
C
omm
un.
N
e
tw
.
C
onf.
,
CCN
C
,
2016,
pp.
68
7
-
692
,
doi
: 1
0.
11
09
/C
C
NC
.
20
16.
7
44
48
63.
[9
]
R
.
M
a
,
N.
Xia
,
H.
H.
C
he
n,
C
.
Y.
C
h
iu,
a
n
d C
.
S
.
Ya
ng,
“
M
od
e
se
le
c
t
io
n,
r
a
di
o r
e
s
our
c
e
a
ll
oc
a
t
io
n,
a
nd
po
we
r
c
oor
di
na
t
io
n
in D2
D
c
om
m
un
ic
a
t
io
ns
,”
I
E
EE Wi
re
l
.
C
ommu
n.
,
vol.
24,
no.
3
,
pp.
112
-
1
21
,
Jun
e
20
17
,
doi
: 1
0.
11
09
/M
W
C
.
20
17.
1
50
03
85
W
C
.
[
10]
H.
Qin,
Z
.
M
i,
C
.
Dong,
F
.
P
e
ng,
a
nd P
.
S
he
ng,
“
An e
xp
e
r
im
e
n
ta
l st
ud
y on m
ul
ti
ho
p D2
D c
om
m
u
nic
a
ti
on
s ba
se
d on
sm
a
r
t
ph
one
s
,”
I
EE
E
Ve
h.
T
e
c
h
no
l.
C
on
f.
,
Ju
l.
20
16,
do
i:
10.
11
09
/V
TC
S
pr
in
g.
2
01
6.
75
04
12
8.
[
11]
S
.
Tya
gi,
S
.
S
om
,
a
nd
Q.
P
.
R
a
na
,
“
A
R
e
lia
bi
li
ty
ba
se
d
Va
r
ia
nt of
A
OD
V in
M
A
NE
Ts
:
P
r
o
pos
a
l,
Ana
ly
si
s
a
n
d
C
om
pa
r
is
on
,”
E
lse
v
ie
r,
P
roc
e
di
a
C
om
pu
t.
Sc
i.
,
vo
l
.
79
,
pp.
90
3
-
9
11
,
20
16
,
do
i:
10.
10
16
/j.
pr
oc
s.
20
16.
0
3.
1
12.
[
12]
N.
M
ova
he
d
ia
n A
tta
r
,
“
D
yna
m
ic
de
te
c
t
io
n of
se
c
ur
e
r
o
ut
e
s in a
d hoc
ne
t
wor
ks
,
”
E
me
r
g.
Sc
i.
J
.
,
vol.
1,
n
o.
4
,
J
a
n
ua
r
y
201
8
,
d
oi
: 10.
2
89
91
/i
js
e
-
0
11
27.
[
13]
W
.
A.
Ja
b
ba
r
,
W
.
K.
S
a
a
d,
a
nd M
.
I
sm
a
il,
“
M
EQ
S
A
-
O
L
S
R
v2: A m
ul
tic
r
ite
r
ia
-
ba
se
d h
ybr
id m
ul
ti
pa
t
h pr
ot
oc
o
l
f
or
e
n
e
r
g
y
-
e
f
f
ic
ie
n
t
a
nd QoS
-
a
wa
r
e
da
ta
r
o
ut
in
g i
n M
AN
ET
-
W
S
N c
o
nve
r
ge
nc
e
sc
e
na
r
io
s of
I
oT
,
”
I
E
EE
A
c
c
e
s
s
,
vo
l.
6,
pp.
76
54
6
-
7
65
72
, N
o
v
.
20
18
,
do
i
: 10.
11
09
/AC
C
ES
S
.
20
18.
2
88
28
53.
[
14]
S
.
M
ille
r
,
“
An Ac
c
e
s
si
ble
,
O
pe
n
-
S
our
c
e
,
R
e
a
l
tim
e
A
O
DV S
im
ula
ti
on i
n M
ATL
AB
,
”
Mis
so
ur
i U
niv
.
Sc
i.
T
e
c
hno
l
,
201
7
.
[
15]
M
.
Abo
lha
sa
n,
M
.
Ab
do
lla
hi,
W
.
Ni,
A.
Ja
m
a
li
po
ur
,
N.
S
ha
r
ia
ti,
a
n
d
J.
L
ipm
a
n,
“
A
r
o
ut
in
g f
r
a
m
e
wor
k f
or
of
f
l
oa
d
in
g
tr
a
f
f
ic
f
r
om
c
e
ll
ula
r
ne
t
wor
ks to
SD
N
-
ba
se
d m
u
lt
i
-
h
op d
e
vic
e
-
to
-
de
v
ic
e
ne
tw
or
k
s
,
”
I
EE
E T
r
an
s.
N
e
tw
.
Se
rv
.
Ma
n
ag.
,
vol.
1
5,
no.
4
,
pp
.
15
16
–
15
31
,
20
18
,
do
i
: 1
0.
11
09
/
TNS
M
.
201
8.
28
75
69
6.
Evaluation Warning : The document was created with Spire.PDF for Python.