I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
,
p
p
.
1543
~
1
5
6
1
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
1
6
i
3
.
pp
1
5
4
3
-
1
5
6
1
1543
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
AI
-
ena
bled
energ
y
-
a
wa
re r
o
uting a
ppro
a
ch f
o
r
fut
ur
e
-
wirel
ess
sens
o
r net
wo
rks
Sh
a
m
s
her
Sin
g
h,
M
a
nd
ee
p
K
um
a
r
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
S
R
P
A
A
d
a
r
s
h
B
h
a
r
t
i
y
a
C
o
l
l
e
g
e
,
P
a
t
h
a
n
k
o
t
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
2
1
,
2
0
2
6
R
ev
is
ed
Ma
r
2
3
,
2
0
2
6
Acc
ep
ted
Ap
r
2
7
,
2
0
2
6
Ne
x
t
-
g
e
n
e
ra
ti
o
n
wire
les
s
se
n
so
r
n
e
two
r
k
s
(W
S
Ns
)
d
e
m
a
n
d
i
n
telli
g
e
n
t,
e
n
e
rg
y
-
a
wa
re
c
o
m
m
u
n
ica
ti
o
n
m
e
c
h
a
n
ism
s
c
a
p
a
b
le
o
f
su
sta
in
i
n
g
lo
n
g
-
term
o
p
e
ra
ti
o
n
in
e
n
v
iro
n
m
e
n
ts
wit
h
v
a
ry
i
n
g
c
o
n
d
it
i
o
n
s
a
n
d
stric
t
re
so
u
rc
e
li
m
it
a
ti
o
n
s.
Trad
it
i
o
n
a
l
r
o
u
ti
n
g
p
ro
t
o
c
o
ls
o
fte
n
fa
il
to
o
p
t
imiz
e
e
n
e
rg
y
c
o
n
su
m
p
ti
o
n
u
n
d
e
r
v
a
ry
i
n
g
n
e
two
rk
d
e
n
sit
ies
,
h
e
tero
g
e
n
e
o
u
s
traffic
p
a
tt
e
rn
s,
a
n
d
e
n
v
ir
o
n
m
e
n
tal
u
n
c
e
rtain
ti
e
s.
T
h
is
re
se
a
rc
h
p
r
o
p
o
se
s
a
n
AI
-
e
n
a
b
led
e
n
e
rg
y
-
e
fficie
n
t
r
o
u
ti
n
g
p
ro
t
o
c
o
l
(AI
-
E
ERP
)
d
e
sig
n
e
d
to
e
n
h
a
n
c
e
n
e
two
r
k
li
fe
ti
m
e
,
sta
b
il
i
ty
,
a
n
d
d
a
ta
d
e
li
v
e
ry
p
e
rfo
rm
a
n
c
e
i
n
n
e
x
t
-
g
e
n
e
ra
ti
o
n
WS
Ns
.
T
h
e
p
ro
t
o
c
o
l
in
teg
ra
tes
m
a
c
h
in
e
lea
rn
in
g
–
b
a
se
d
n
o
d
e
se
lec
ti
o
n
,
a
d
a
p
ti
v
e
c
lu
ste
rin
g
,
a
n
d
p
re
d
icti
v
e
re
sid
u
a
l
-
e
n
e
rg
y
e
sti
m
a
ti
o
n
to
m
a
k
e
o
p
ti
m
ize
d
ro
u
ti
n
g
d
e
c
isio
n
s
in
re
a
l
ti
m
e
.
Us
in
g
AI
-
d
ri
v
e
n
m
o
d
e
ls,
AI
-
EE
RP
d
y
n
a
m
ica
ll
y
a
d
j
u
sts
ro
u
ti
n
g
p
a
t
h
s
b
a
se
d
o
n
e
n
e
rg
y
p
a
tt
e
rn
s,
li
n
k
q
u
a
li
t
y
,
a
n
d
n
e
tw
o
rk
to
p
o
lo
g
y
c
h
a
n
g
e
s.
Th
e
sim
u
latio
n
o
u
tco
m
e
s
c
lea
rly
in
d
ica
te
th
a
t
th
e
p
r
o
p
o
se
d
a
p
p
ro
a
c
h
a
c
h
iev
e
s
n
o
tab
le
g
a
i
n
s
in
e
n
e
rg
y
e
fficie
n
c
y
,
p
a
c
k
e
t
d
e
li
v
e
ry
re
li
a
b
il
it
y
,
a
n
d
n
e
two
rk
l
ifetime
wh
e
n
c
o
m
p
a
re
d
with
trad
i
ti
o
n
a
l
ro
u
ti
n
g
p
r
o
to
c
o
ls,
in
c
lu
d
in
g
L
EACH,
P
EG
ASIS
,
a
n
d
HEED
.
Th
e
p
r
o
p
o
se
d
a
p
p
r
o
a
c
h
e
sta
b
li
s
h
e
s
a
ro
b
u
st
a
n
d
sc
a
lab
le
fra
m
e
wo
rk
f
o
r
fu
tu
re
in
telli
g
e
n
t
WS
N
d
e
p
lo
y
m
e
n
ts
a
c
ro
ss
a
p
p
l
ica
ti
o
n
s
in
c
l
u
d
in
g
sm
a
rt
c
it
ies
,
p
re
c
isio
n
a
g
ric
u
lt
u
re
,
e
n
v
ir
o
n
m
e
n
t
-
f
o
c
u
se
d
a
p
p
li
c
a
ti
o
n
s
a
n
d
a
u
to
m
a
ted
i
n
d
u
str
ial
o
p
e
ra
ti
o
n
s
.
K
ey
w
o
r
d
s
:
Ad
ap
tiv
e
clu
s
ter
in
g
AI
-
en
ab
led
r
o
u
tin
g
E
n
er
g
y
ef
f
icien
t p
r
o
to
co
l
I
n
tellig
en
t c
o
m
m
u
n
icatio
n
Ma
ch
in
e
lear
n
in
g
Nex
t
-
g
en
I
n
ter
n
et
o
f
T
h
in
g
s
n
etwo
r
k
s
Sm
ar
t sen
s
in
g
s
y
s
tem
s
W
ir
eles
s
s
en
s
o
r
n
etwo
r
k
s
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Sh
am
s
h
er
Sin
g
h
Dep
ar
t
m
en
t
o
f
C
o
m
p
u
ter
Scie
n
ce
,
SR
PA A
d
ar
s
h
B
h
ar
tiy
a
C
o
lleg
e
J
alan
d
h
ar
-
Dalh
o
u
s
ie,
B
y
-
Pas
s
R
o
ad
,
Path
an
k
o
t.
Pu
n
jab
,
I
n
d
i
a
E
m
ail: d
r
s
am
s
in
g
h
1
6
1
2
@
g
m
a
il.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
W
ir
eles
s
s
en
s
o
r
n
etwo
r
k
s
(
W
SNs
)
ar
e
in
cr
ea
s
in
g
ly
u
s
ed
as
a
co
r
e
tech
n
o
lo
g
y
in
m
o
d
er
n
in
tellig
en
t
s
y
s
tem
s
,
en
ab
lin
g
c
o
n
tin
u
o
u
s
m
o
n
ito
r
in
g
a
n
d
d
ata
ac
q
u
is
itio
n
ac
r
o
s
s
d
iv
er
s
e
e
n
v
ir
o
n
m
en
ts
.
T
h
e
r
ap
i
d
g
r
o
wth
o
f
ap
p
licatio
n
s
in
clu
d
in
g
s
m
ar
t
cities,
en
v
ir
o
n
m
en
tal
s
en
s
in
g
,
in
d
u
s
tr
ial
led
a
u
to
m
ati
o
n
,
h
ea
lth
ca
r
e,
an
d
p
r
ec
is
io
n
ag
r
icu
ltu
r
e
h
as
in
cr
ea
s
ed
th
e
d
em
an
d
f
o
r
W
SNs
ca
p
ab
le
o
f
lo
n
g
-
ter
m
au
to
n
o
m
o
u
s
an
d
r
eliab
le
o
p
er
atio
n
[
1
]
.
On
e
o
f
th
e
m
o
s
t
s
ig
n
if
ican
t
ch
allen
g
es
af
f
ec
ti
n
g
s
u
ch
d
ep
lo
y
m
en
ts
is
en
er
g
y
co
n
s
u
m
p
tio
n
,
as
s
en
s
o
r
n
o
d
es
ar
e
g
en
er
all
y
p
o
wer
ed
b
y
co
n
s
tr
ain
e
d
,
n
o
n
-
r
ec
h
ar
g
ea
b
le
b
atter
y
s
o
u
r
ce
s
.
Sin
ce
co
m
m
u
n
icatio
n
—
p
ar
ticu
lar
ly
r
o
u
tin
g
—
co
n
s
u
m
es
th
e
h
ig
h
est
p
r
o
p
o
r
tio
n
o
f
en
er
g
y
,
d
esig
n
in
g
ef
f
icien
t
r
o
u
tin
g
m
ec
h
an
is
m
s
r
em
ain
s
c
en
tr
al
to
im
p
r
o
v
in
g
th
e
o
v
er
all
n
etwo
r
k
p
er
f
o
r
m
an
ce
a
n
d
life
tim
e.
T
r
ad
itio
n
al
r
o
u
tin
g
p
r
o
to
c
o
ls
s
u
ch
as
L
E
A
C
H,
T
E
E
N,
PEGA
SIS,
an
d
HE
E
D
h
av
e
p
r
o
v
id
ed
f
u
n
d
am
e
n
tal
s
o
lu
tio
n
s
f
o
r
d
at
a
f
o
r
war
d
i
n
g
a
n
d
clu
s
ter
in
g
;
h
o
wev
er
,
t
h
ese
ap
p
r
o
ac
h
es
o
f
ten
r
ely
o
n
s
tatic
d
ec
is
io
n
-
m
ak
in
g
,
h
e
u
r
is
tic
th
r
esh
o
ld
s
,
o
r
p
r
ed
eter
m
i
n
ed
en
e
r
g
y
m
o
d
els.
As
W
SNs
ev
o
lv
e
to
war
d
lar
g
e
-
s
ca
le,
h
eter
o
g
en
e
o
u
s
,
an
d
m
is
s
io
n
-
c
r
itical
ap
p
licatio
n
s
,
th
ese
clas
s
ical
m
eth
o
d
s
b
ec
o
m
e
in
s
u
f
f
i
cien
t
to
ef
f
ec
tiv
ely
h
an
d
le
d
y
n
am
ic
n
etwo
r
k
co
n
d
itio
n
s
,
u
n
p
r
e
d
ictab
le
n
o
d
e
f
a
ilu
r
es,
v
ar
y
in
g
tr
af
f
ic
p
atter
n
s
,
an
d
th
e
n
ee
d
f
o
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
5
4
3
-
1
5
6
1
1544
r
ea
l
-
tim
e
o
p
tim
izatio
n
.
T
h
is
in
cr
ea
s
in
g
co
m
p
lex
ity
h
as
led
to
a
s
h
if
t
to
war
d
lev
e
r
ag
in
g
a
r
tific
ial
in
tellig
en
ce
(
AI
)
an
d
m
ac
h
in
e
lea
r
n
in
g
(
M
L
)
m
eth
o
d
s
to
en
h
a
n
ce
r
o
u
tin
g
d
ec
is
io
n
s
with
in
r
o
u
tin
g
p
r
o
to
co
ls
d
esig
n
.
AI
-
en
ab
led
r
o
u
tin
g
i
n
tr
o
d
u
ce
s
th
e
ab
ilit
y
to
lear
n
f
r
o
m
n
et
wo
r
k
s
tates,
p
r
ed
ict
n
o
d
e
e
n
e
r
g
y
lev
els,
id
en
tify
o
p
tim
al
f
o
r
war
d
in
g
p
ath
s
,
an
d
ad
a
p
tiv
ely
r
estru
ct
u
r
e
clu
s
ter
s
ac
co
r
d
in
g
to
en
v
ir
o
n
m
en
tal
ch
a
n
g
es.
B
y
in
teg
r
atin
g
AI
-
d
r
i
v
en
d
ec
is
io
n
-
m
ak
in
g
,
W
SNs
ca
n
s
ig
n
if
i
ca
n
tly
en
h
a
n
ce
e
n
er
g
y
ef
f
icien
cy
,
r
e
d
u
ce
c
o
n
tr
o
l
o
v
er
h
ea
d
,
a
n
d
ac
h
iev
e
s
u
s
tain
ed
n
etwo
r
k
p
er
f
o
r
m
a
n
ce
ev
en
u
n
d
er
u
n
ce
r
tain
o
r
d
y
n
am
i
c
co
n
d
itio
n
s
.
T
h
ese
in
tellig
en
t
s
y
s
tem
s
f
u
r
th
er
s
u
p
p
o
r
t
n
ex
t
-
g
en
er
atio
n
d
ep
lo
y
m
en
ts
,
wh
er
e
s
ca
lab
ilit
y
,
r
o
b
u
s
tn
ess
,
an
d
r
ea
l
-
tim
e
o
p
tim
izatio
n
ar
e
k
ey
r
e
q
u
ir
em
en
ts
.
T
h
is
r
esear
ch
p
r
o
p
o
s
es
an
AI
-
en
ab
led
en
e
r
g
y
-
e
f
f
icien
t
r
o
u
t
in
g
p
r
o
to
c
o
l
(
AI
-
E
E
R
P)
th
at
lev
er
ag
es
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
s
t
o
o
p
tim
ize
r
o
u
tin
g
d
ec
is
io
n
s
b
ased
o
n
r
esid
u
al
e
n
er
g
y
p
r
e
d
ictio
n
,
lin
k
q
u
ality
esti
m
atio
n
,
an
d
ad
ap
tiv
e
cl
u
s
ter
in
g
s
tr
ateg
ies.
Un
lik
e
co
n
v
e
n
tio
n
al
p
r
o
t
o
co
ls
,
AI
-
E
E
R
P
co
n
tin
u
o
u
s
ly
lear
n
s
f
r
o
m
n
etwo
r
k
b
e
h
av
io
r
to
d
y
n
am
ically
ad
ju
s
t
r
o
u
tin
g
p
ath
s
,
th
er
eb
y
m
a
x
im
izin
g
e
n
er
g
y
co
n
s
er
v
atio
n
an
d
ex
ten
d
in
g
n
etwo
r
k
life
tim
e.
T
h
r
o
u
g
h
ex
ten
s
iv
e
s
im
u
latio
n
s
an
d
p
er
f
o
r
m
an
ce
an
a
ly
s
is
,
th
is
s
tu
d
y
d
em
o
n
s
tr
ates
th
at
AI
-
E
E
R
P
o
u
tp
er
f
o
r
m
s
ex
is
tin
g
en
er
g
y
-
ef
f
icien
t
p
r
o
to
c
o
ls
ac
r
o
s
s
p
er
f
o
r
m
an
ce
in
d
icato
r
s
r
ef
lectin
g
en
er
g
y
c
o
n
s
u
m
p
tio
n
,
d
ata
d
eliv
er
y
s
u
cc
ess
,
n
etwo
r
k
s
tab
ilit
y
,
an
d
th
r
o
u
g
h
p
u
t.
Ov
er
all,
th
is
in
teg
r
atio
n
o
f
A
I
tech
n
iq
u
es
in
to
r
o
u
tin
g
d
esig
n
r
e
p
r
esen
ts
a
tr
an
s
f
o
r
m
ativ
e
s
tep
f
o
r
n
ex
t
-
g
en
e
r
atio
n
W
SNs
,
en
ab
l
in
g
m
o
r
e
s
u
s
tain
ab
le,
s
ca
lab
l
e,
an
d
in
tellig
en
t
s
en
s
o
r
n
et
wo
r
k
a
r
ch
itectu
r
es
s
u
itab
le
f
o
r
em
er
g
in
g
I
n
te
r
n
et
o
f
T
h
in
g
s
-
d
r
iv
en
(
I
o
T
-
d
r
iv
e
n
)
en
v
ir
o
n
m
en
ts
.
a.
C
o
n
tr
ib
u
tio
n
s
o
f
t
h
is
wo
r
k
Mo
s
t
ex
is
tin
g
en
er
g
y
-
awa
r
e
r
o
u
tin
g
p
r
o
t
o
co
ls
in
W
SNs
r
ely
o
n
f
i
x
ed
h
eu
r
is
tics
an
d
in
s
tan
tan
eo
u
s
n
etwo
r
k
p
ar
am
ete
r
s
,
wh
ich
lim
its
th
eir
ab
ilit
y
to
ad
ap
t to
c
h
an
g
in
g
co
n
d
itio
n
s
.
T
o
o
v
er
c
o
m
e
th
ese
lim
itatio
n
s
,
th
is
wo
r
k
p
r
esen
ts
a
lear
n
in
g
-
ass
is
ted
r
o
u
tin
g
ap
p
r
o
ac
h
th
at
en
ab
les
p
r
o
ac
tiv
e
an
d
ad
a
p
tiv
e
d
ec
is
io
n
-
m
ak
in
g
.
T
h
e
m
ain
co
n
tr
ib
u
tio
n
s
o
f
th
e
p
r
o
p
o
s
ed
s
tu
d
y
ar
e
s
u
m
m
ar
ize
d
as f
o
llo
ws:
b.
L
ea
r
n
in
g
-
ass
is
ted
p
r
ed
ictiv
e
r
o
u
tin
g
d
esig
n
T
h
is
s
tu
d
y
in
tr
o
d
u
ce
s
a
n
ew
AI
-
E
E
R
P
th
at
in
co
r
p
o
r
ates
m
ac
h
in
e
lear
n
i
n
g
tech
n
iq
u
es
to
p
r
ed
ict
r
esid
u
al
n
o
d
e
en
er
g
y
an
d
ass
ess
lin
k
q
u
ality
.
B
y
em
b
ed
d
in
g
p
r
ed
ictio
n
ca
p
a
b
ilit
ies
in
to
r
o
u
tin
g
d
ec
is
io
n
s
,
th
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
s
h
if
ts
f
r
o
m
r
ea
ctiv
e
b
eh
a
v
io
r
t
o
f
o
r
war
d
-
lo
o
k
i
n
g
r
o
u
tin
g
m
an
ag
e
m
en
t.
c.
Mu
lti
-
cr
iter
ia
ad
ap
tiv
e
clu
s
ter
-
h
ea
d
s
elec
tio
n
An
ad
ap
tiv
e
clu
s
ter
-
h
ea
d
s
elec
tio
n
m
ec
h
a
n
is
m
is
d
ev
elo
p
e
d
th
at
jo
in
tly
co
n
s
id
er
s
p
r
ed
ic
ted
en
er
g
y
av
ailab
ilit
y
,
co
m
m
u
n
icatio
n
r
eliab
ilit
y
,
an
d
n
o
d
e
p
r
o
x
im
i
ty
.
T
h
is
m
u
lti
-
f
ac
to
r
s
tr
ateg
y
r
ed
u
ce
s
ar
b
itra
r
y
clu
s
ter
-
h
ea
d
ass
ig
n
m
en
t,
m
in
i
m
izes
p
r
em
atu
r
e
n
o
d
e
d
ep
leti
o
n
,
an
d
e
n
h
an
ce
s
clu
s
ter
s
tab
ilit
y
o
v
er
s
u
cc
ess
iv
e
co
m
m
u
n
icatio
n
r
o
u
n
d
s
.
d.
E
n
er
g
y
-
a
n
d
lin
k
-
awa
r
e
m
u
lti
-
h
o
p
r
o
u
tin
g
s
tr
ateg
y
AI
-
E
E
R
P
em
p
lo
y
s
an
in
telli
g
en
t
m
u
lti
-
h
o
p
r
o
u
tin
g
s
ch
e
m
e
th
at
d
y
n
am
ically
s
elec
ts
f
o
r
war
d
in
g
p
ath
s
b
ased
o
n
an
ticip
ated
e
n
e
r
g
y
lev
els
an
d
lin
k
co
n
d
itio
n
s
.
T
h
is
ap
p
r
o
ac
h
p
r
ev
e
n
ts
o
v
er
u
s
e
o
f
wea
k
o
r
lo
w
-
en
er
g
y
n
o
d
es a
n
d
i
m
p
r
o
v
es th
e
r
o
b
u
s
tn
ess
o
f
d
ata
tr
an
s
m
is
s
io
n
to
war
d
th
e
b
ase
s
tatio
n
.
e.
L
ig
h
tweig
h
t m
ac
h
in
e
lear
n
in
g
f
o
r
p
r
ac
tical
d
ep
lo
y
m
en
t
T
o
en
s
u
r
e
f
ea
s
ib
ilit
y
o
n
r
eso
u
r
ce
-
lim
ited
s
en
s
o
r
p
latf
o
r
m
s
,
th
e
p
r
o
p
o
s
ed
p
r
o
t
o
co
l u
tili
ze
s
lig
h
tweig
h
t
lear
n
in
g
m
o
d
els
with
lo
w
co
m
p
u
tatio
n
al
o
v
er
h
ea
d
.
Mo
d
el
tr
ain
in
g
is
p
er
f
o
r
m
ed
ce
n
tr
al
ly
,
wh
ile
n
o
d
e
-
lev
el
in
f
er
en
ce
is
k
ep
t e
f
f
icie
n
t,
en
a
b
lin
g
p
r
ac
tical
d
ep
lo
y
m
en
t w
ith
o
u
t e
x
ce
s
s
iv
e
en
er
g
y
o
r
p
r
o
c
ess
in
g
co
s
ts
.
f.
E
x
ten
s
iv
e
s
im
u
latio
n
-
b
ased
p
e
r
f
o
r
m
a
n
ce
an
aly
s
is
T
h
e
ef
f
ec
tiv
en
ess
o
f
AI
-
E
E
R
P
is
v
alid
ated
th
r
o
u
g
h
co
m
p
r
eh
en
s
iv
e
NS
-
3
s
im
u
latio
n
s
u
n
d
e
r
d
if
f
er
e
n
t
n
etwo
r
k
d
e
n
s
ities
,
b
ase
s
tatio
n
lo
ca
tio
n
s
,
an
d
tr
af
f
ic
s
ce
n
ar
i
o
s
.
Per
f
o
r
m
a
n
ce
is
ass
ess
ed
u
s
in
g
m
etr
ics
s
u
ch
as
n
etwo
r
k
life
tim
e,
r
esid
u
al
en
er
g
y
b
alan
ce
,
p
ac
k
et
d
eliv
e
r
y
r
atio
,
th
r
o
u
g
h
p
u
t,
d
elay
,
r
o
u
tin
g
o
v
er
h
ea
d
,
an
d
en
er
g
y
f
air
n
ess
.
g.
Scalab
ilit
y
an
d
lo
n
g
-
ter
m
n
et
wo
r
k
s
u
s
tain
ab
ilit
y
Simu
latio
n
r
esu
lts
d
em
o
n
s
tr
ate
th
at
AI
-
E
E
R
P
co
n
s
is
ten
tly
o
u
tp
e
r
f
o
r
m
s
b
en
c
h
m
ar
k
p
r
o
t
o
co
ls
,
in
clu
d
in
g
L
E
AC
H,
HE
E
D,
a
n
d
PEGA
SIS,
ac
r
o
s
s
b
o
th
s
p
ar
s
e
an
d
d
e
n
s
e
n
etwo
r
k
c
o
n
f
ig
u
r
atio
n
s
.
T
h
ese
r
esu
lts
h
ig
h
lig
h
t
th
e
p
r
o
t
o
co
l’
s
s
ca
lab
ilit
y
an
d
s
u
itab
ilit
y
f
o
r
lo
n
g
-
ter
m
o
p
er
atio
n
in
lar
g
e
-
s
ca
le
W
SN
-
en
ab
led
I
n
ter
n
et
o
f
T
h
in
g
s
(
I
o
T
)
a
p
p
lic
atio
n
s
.
Nex
t
-
g
en
er
atio
n
W
SNs
ar
e
r
eq
u
ir
ed
to
o
p
er
ate
i
n
m
ass
-
s
ca
le,
h
eter
o
g
en
eo
u
s
,
an
d
tim
e
-
v
ar
y
in
g
en
v
ir
o
n
m
en
ts
,
wh
er
e
s
en
s
o
r
-
e
q
u
ip
p
e
d
u
n
its
s
en
s
e
th
e
en
v
ir
o
n
m
en
t
o
n
a
n
o
n
g
o
in
g
b
asis
,
p
r
o
ce
s
s
,
an
d
tr
an
s
m
it
d
ata
u
s
in
g
lim
ited
b
atter
y
p
o
wer
.
T
h
e
p
r
im
ar
y
ch
allen
g
e
in
s
u
ch
n
etwo
r
k
s
is
en
er
g
y
-
ef
f
icien
t
r
o
u
tin
g
,
as
co
m
m
u
n
icatio
n
co
n
s
u
m
es
th
e
m
ajo
r
ity
o
f
a
n
o
d
e’
s
en
er
g
y
.
E
x
is
tin
g
m
eth
o
d
s
f
o
r
d
ata
r
o
u
tin
g
—
n
o
tab
ly
L
E
AC
H,
PEGA
SIS,
T
E
E
N,
an
d
HE
E
D
—
u
s
e
s
tatic
th
r
esh
o
ld
s
,
h
eu
r
is
tic
clu
s
ter
f
o
r
m
ati
o
n
,
o
r
f
ix
ed
r
o
u
tin
g
p
ath
s
,
m
ak
i
n
g
t
h
em
i
n
s
u
f
f
icien
t
f
o
r
r
ea
l
-
tim
e
ad
a
p
tatio
n
t
o
ch
an
g
in
g
n
etwo
r
k
co
n
d
itio
n
s
.
T
h
ese
p
r
o
t
o
co
ls
ar
e
u
n
ab
le
to
ac
cu
r
ately
p
r
ed
ict
n
o
d
e
en
er
g
y
d
e
p
letio
n
,
b
al
an
ce
lo
ad
d
is
tr
ib
u
tio
n
,
o
r
d
y
n
am
ically
o
p
tim
ize
r
o
u
tin
g
b
ased
o
n
lin
k
q
u
ality
a
n
d
to
p
o
lo
g
y
v
ar
iatio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
AI
-
en
a
b
led
en
erg
y
-
a
w
a
r
e
r
o
u
tin
g
a
p
p
r
o
a
ch
fo
r
fu
tu
r
e
-
w
ir
eless
s
en
s
o
r
n
et
w
o
r
ks
(
S
h
a
msh
er S
in
g
h
)
1545
As a
r
esu
lt,
cu
r
r
en
t
W
SN
s
f
r
eq
u
en
tly
ex
p
er
ien
ce
p
r
em
atu
r
e
n
o
d
e
f
ailu
r
es,
u
n
ev
e
n
en
er
g
y
d
is
s
ip
atio
n
,
r
ed
u
ce
d
n
etwo
r
k
s
tab
ilit
y
,
an
d
s
h
o
r
ten
e
d
o
p
er
atio
n
al
life
ti
m
e.
W
ith
in
cr
ea
s
in
g
d
e
m
an
d
s
p
lace
d
o
n
W
SNs
in
s
m
ar
t
city
in
f
r
astru
ctu
r
es,
i
n
d
u
s
tr
ial
I
o
T
s
y
s
tem
s
,
en
v
ir
o
n
m
en
tal
m
o
n
ito
r
i
n
g
,
a
n
d
o
t
h
er
a
p
p
licatio
n
s
,
th
er
e
is
an
u
r
g
en
t
n
ee
d
f
o
r
an
in
tellig
en
t
r
o
u
tin
g
ap
p
r
o
ac
h
ca
p
ab
le
o
f
lear
n
in
g
n
etwo
r
k
b
e
h
av
io
r
,
f
o
r
ec
asti
n
g
en
er
g
y
co
n
s
u
m
p
tio
n
p
atter
n
s
,
an
d
s
elec
tin
g
o
p
tim
al
co
m
m
u
n
icatio
n
p
ath
s
au
to
n
o
m
o
u
s
ly
.
Desp
ite
s
ig
n
if
ican
t
ad
v
an
ce
m
en
ts
,
ex
is
tin
g
r
o
u
tin
g
p
r
o
t
o
co
ls
s
till
s
tr
u
g
g
le
to
b
alan
ce
en
er
g
y
co
n
s
u
m
p
tio
n
,
ad
a
p
t
to
d
y
n
am
i
c
co
n
d
itio
n
s
,
an
d
p
r
o
v
i
d
e
s
ca
lab
le
p
er
f
o
r
m
a
n
ce
.
I
n
p
ar
ticu
la
r
,
cu
r
r
en
t
AI
-
b
ased
ap
p
r
o
ac
h
es
o
f
ten
lack
in
teg
r
a
tio
n
o
f
p
r
ed
ictio
n
,
clu
s
ter
in
g
,
an
d
r
o
u
tin
g
with
in
a
u
n
if
ie
d
f
r
am
ewo
r
k
.
T
h
is
m
o
tiv
ates
th
e
d
esig
n
o
f
an
in
t
ellig
en
t
r
o
u
tin
g
p
r
o
to
co
l
th
at
co
m
b
in
es
p
r
e
d
ictiv
e
en
er
g
y
e
s
tim
atio
n
,
ad
ap
tiv
e
clu
s
ter
in
g
,
an
d
ef
f
icien
t p
ath
s
elec
tio
n
to
en
h
a
n
ce
n
etwo
r
k
li
f
etim
e
an
d
r
eliab
ilit
y
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
E
n
er
g
y
-
awa
r
e
r
o
u
tin
g
is
wid
el
y
r
ec
o
g
n
ized
as
a
f
u
n
d
a
m
en
ta
l
r
esear
ch
ch
allen
g
e
in
W
SNs
,
o
win
g
to
th
e
lim
ited
e
n
er
g
y
r
eso
u
r
ce
s
o
f
s
en
s
o
r
n
o
d
es,
b
ec
a
u
s
e
s
en
s
in
g
u
n
its
o
p
er
ate
with
r
estricte
d
b
atter
y
ca
p
ac
ity
an
d
lim
ited
co
m
p
u
tatio
n
al
r
es
o
u
r
ce
s
.
Sin
ce
co
m
m
u
n
icatio
n
co
n
s
u
m
es
a
m
ajo
r
p
o
r
tio
n
o
f
n
o
d
e
en
er
g
y
,
ea
r
ly
r
o
u
tin
g
s
o
lu
tio
n
s
lar
g
ely
co
n
ce
n
tr
ated
o
n
clu
s
ter
in
g
tech
n
i
q
u
es,
co
m
b
in
i
n
g
d
ata
ag
g
r
eg
atio
n
with
r
ed
u
ce
d
r
ed
u
n
d
an
t c
o
m
m
u
n
icatio
n
to
e
x
ten
d
n
etwo
r
k
life
tim
e.
2
.
1
.
Cla
s
s
ica
l e
nerg
y
-
ef
f
icient
r
o
uting
pr
o
t
o
co
ls
L
E
AC
H
ad
o
p
ts
a
r
an
d
o
m
ize
d
m
eth
o
d
f
o
r
r
o
tatin
g
clu
s
ter
h
ea
d
s
s
o
th
at
en
er
g
y
co
n
s
u
m
p
tio
n
is
s
h
ar
ed
m
o
r
e
e
v
en
ly
am
o
n
g
t
h
e
n
o
d
es
[
1
]
.
Alth
o
u
g
h
L
E
A
C
H
ef
f
ec
tiv
ely
r
ed
u
ce
s
co
m
m
u
n
icatio
n
o
v
e
r
h
ea
d
an
d
s
im
p
lifie
s
clu
s
ter
f
o
r
m
atio
n
,
it
o
f
ten
lead
s
to
u
n
s
tab
le
clu
s
ter
s
an
d
ea
r
ly
ex
h
au
s
tio
n
o
f
clu
s
ter
h
ea
d
(
C
H)
n
o
d
es,
p
ar
tic
u
lar
ly
in
h
eter
o
g
e
n
eo
u
s
n
etwo
r
k
en
v
i
r
o
n
m
e
n
ts
.
PEGA
SIS
f
o
llo
ws
a
ch
ain
-
b
ased
co
m
m
u
n
icatio
n
m
o
d
el
wh
er
e
n
o
d
es
co
m
m
u
n
icate
with
n
ea
r
b
y
n
eig
h
b
o
r
s
,
an
d
o
n
e
n
o
d
e
s
en
d
s
th
e
co
llected
d
ata
to
th
e
b
as
e
s
tatio
n
[
2
]
.
W
h
ile
th
is
s
tr
ateg
y
m
in
im
izes
lo
n
g
-
d
is
tan
ce
tr
an
s
m
is
s
io
n
s
,
it
s
u
f
f
er
s
f
r
o
m
i
n
cr
ea
s
ed
c
o
m
m
u
n
icatio
n
d
elay
an
d
lim
ited
s
ca
lab
ilit
y
in
lar
g
e
-
s
ca
le
d
ep
lo
y
m
e
n
ts
.
Hy
b
r
id
e
n
er
g
y
-
ef
f
icien
t
d
is
tr
i
b
u
ted
cl
u
s
ter
in
g
(
HE
E
D
)
en
h
an
ce
d
clu
s
ter
-
h
ea
d
elec
tio
n
s
t
r
ateg
y
b
y
in
co
r
p
o
r
atin
g
av
ailab
le
b
atter
y
p
o
wer
a
n
d
c
o
m
m
u
n
icatio
n
ef
f
o
r
t
as
d
ec
is
io
n
p
a
r
am
eter
s
,
r
esu
ltin
g
in
m
o
r
e
co
n
s
is
ten
t
clu
s
ter
f
o
r
m
atio
n
a
n
d
im
p
r
o
v
ed
n
etwo
r
k
s
tab
ilit
y
[
3
]
.
T
h
ese
ea
r
l
y
p
r
o
to
co
ls
laid
th
e
f
o
u
n
d
ati
o
n
f
o
r
h
ier
ar
ch
ical
r
o
u
ti
n
g
in
W
SN
s
an
d
co
n
tin
u
e
to
s
er
v
e
as
r
ef
er
en
ce
b
e
n
ch
m
ar
k
s
f
o
r
th
e
d
ev
elo
p
m
e
n
t
o
f
ad
v
an
ce
d
en
er
g
y
-
awa
r
e
r
o
u
tin
g
s
tr
ateg
ies.
2
.
2
.
E
nh
a
nced
a
nd
met
a
heu
ristic
-
ba
s
ed
r
o
uti
ng
a
pp
r
o
a
ches
Nu
m
er
o
u
s
ex
ten
s
io
n
s
to
L
E
AC
H,
PEGA
SIS,
an
d
HE
E
D
h
a
v
e
s
o
u
g
h
t
to
a
d
d
r
ess
th
eir
lim
i
tatio
n
s
b
y
in
co
r
p
o
r
atin
g
h
eter
o
g
e
n
eity
,
m
u
lti
-
h
o
p
r
o
u
tin
g
,
m
o
b
ilit
y
h
an
d
lin
g
,
an
d
im
p
r
o
v
ed
C
H
s
elec
tio
n
h
eu
r
is
tics
.
V
ar
ian
ts
s
u
ch
as
M
-
L
E
AC
H,
T
E
E
N,
E
-
PEGA
SIS,
an
d
H
-
HE
E
D
d
em
o
n
s
tr
ate
in
cr
em
e
n
tal
im
p
r
o
v
e
m
en
ts
u
n
d
er
s
p
ec
if
ic
s
ce
n
a
r
io
s
[
4
]
.
R
esear
ch
er
s
h
av
e
also
a
p
p
l
ied
m
etah
e
u
r
is
tic
tech
n
iq
u
es,
s
u
ch
as
g
en
etic
alg
o
r
ith
m
(
GA)
,
p
ar
ticle
s
war
m
o
p
tim
izatio
n
(
PS
O)
,
an
d
a
n
t
co
lo
n
y
o
p
tim
izatio
n
(
AC
O)
,
to
r
e
f
in
e
clu
s
ter
-
h
ea
d
s
elec
tio
n
an
d
r
o
u
tin
g
p
a
th
f
o
r
m
atio
n
[
5
]
.
W
h
ile
th
ese
m
eth
o
d
s
ac
h
iev
e
b
etter
e
n
er
g
y
b
alan
cin
g
,
th
eir
co
m
p
u
tatio
n
al
o
v
er
h
ea
d
lim
its
p
r
ac
tical
d
ep
lo
y
m
en
t
o
n
s
en
s
o
r
n
o
d
es
with
lim
ited
en
er
g
y
an
d
p
r
o
ce
s
s
in
g
ca
p
ac
ity
.
2
.
3
.
AI a
nd
ma
chine le
a
rnin
g
-
ba
s
ed
r
o
uting
t
ec
hn
iq
ues
R
ec
en
t r
esear
ch
tr
en
d
s
h
ig
h
lig
h
t th
e
in
cr
ea
s
in
g
r
elian
ce
o
n
AI
an
d
ML
tech
n
iq
u
es to
en
h
an
ce
r
o
u
tin
g
ad
ap
tab
ilit
y
in
W
SN
s
.
ML
tech
n
iq
u
es
h
av
e
b
ee
n
ap
p
lied
to
r
esid
u
al
en
er
g
y
p
r
e
d
ictio
n
,
lin
k
-
q
u
ality
esti
m
atio
n
,
an
o
m
aly
d
etec
tio
n
,
an
d
d
y
n
am
ic
r
o
u
tin
g
d
ec
i
s
io
n
s
[
6
]
.
R
ein
f
o
r
ce
m
en
t
lear
n
in
g
-
b
ased
r
o
u
tin
g
,
s
u
ch
as
Q
-
lear
n
in
g
an
d
DR
L
,
h
elp
n
o
d
es
i
d
en
tify
ef
f
icien
t
f
o
r
war
d
in
g
s
tr
ateg
ies
o
v
er
tim
e
i
n
d
y
n
am
i
c
en
v
ir
o
n
m
en
ts
,
im
p
r
o
v
in
g
th
r
o
u
g
h
p
u
t
an
d
r
ed
u
ci
n
g
e
n
er
g
y
d
ep
letio
n
h
o
ts
p
o
ts
.
R
ec
en
t
s
tu
d
ies
h
av
e
e
x
p
lo
r
e
d
v
ar
io
u
s
in
tellig
en
t
an
d
en
er
g
y
-
awa
r
e
r
o
u
tin
g
s
tr
ateg
ies
f
o
r
wir
eless
s
en
s
o
r
n
etwo
r
k
s
.
Me
tah
eu
r
is
tic
-
b
ased
clu
s
ter
in
g
an
d
r
o
u
tin
g
ap
p
r
o
a
ch
es
h
av
e
b
ee
n
in
v
esti
g
ated
t
o
im
p
r
o
v
e
n
etwo
r
k
life
tim
e
a
n
d
co
m
m
u
n
icatio
n
ef
f
icien
cy
[
7
]
,
[
8
]
.
Sev
er
al
s
u
r
v
ey
s
tu
d
ies
h
av
e
also
an
aly
z
ed
r
o
u
tin
g
tech
n
iq
u
es,
en
er
g
y
-
ef
f
icien
t
p
r
o
to
co
ls
,
an
d
o
p
tim
izatio
n
s
tr
ateg
ies
f
o
r
W
SNs
an
d
I
o
T
-
e
n
ab
led
en
v
i
r
o
n
m
en
ts
[
9
]
–
[
1
4
]
.
I
n
a
d
d
itio
n
,
m
ac
h
in
e
lear
n
in
g
an
d
ar
tific
ial
in
tellig
en
ce
t
ec
h
n
iq
u
es
h
a
v
e
in
cr
ea
s
in
g
ly
b
ee
n
in
co
r
p
o
r
ated
in
to
w
ir
eless
n
etwo
r
k
in
g
ap
p
licatio
n
s
,
in
clu
d
in
g
an
o
m
a
ly
d
etec
tio
n
,
ad
a
p
tiv
e
r
o
u
tin
g
,
an
d
p
r
e
d
ictiv
e
r
eso
u
r
ce
m
an
a
g
em
en
t
[
1
5
]
–
[
1
9
]
.
T
h
ese
s
tu
d
ies
d
em
o
n
s
tr
ate
t
h
e
g
r
o
win
g
im
p
o
r
tan
ce
o
f
i
n
tellig
en
t
d
ec
is
io
n
-
m
a
k
in
g
m
ec
h
an
is
m
s
in
n
ex
t
-
g
en
er
atio
n
W
SN
ar
ch
itectu
r
es
.
R
ec
en
t
s
u
r
v
ey
s
h
ig
h
lig
h
t
th
e
g
r
o
win
g
ad
o
p
tio
n
o
f
en
er
g
y
-
a
war
e
an
d
lear
n
in
g
-
b
ased
r
o
u
tin
g
p
r
o
to
co
ls
as
ef
f
ec
tiv
e
s
o
lu
tio
n
s
f
o
r
ex
ten
d
in
g
n
etwo
r
k
life
tim
e
[
2
0
]
.
Ma
ch
in
e
-
lear
n
in
g
-
b
ased
r
o
u
tin
g
s
ch
em
es
h
av
e
b
ee
n
s
h
o
wn
to
o
u
tp
er
f
o
r
m
h
eu
r
is
tic
m
eth
o
d
s
b
y
en
ab
lin
g
ad
a
p
tiv
e
an
d
p
r
ed
ictio
n
-
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
5
4
3
-
1
5
6
1
1546
d
r
iv
en
e
n
er
g
y
m
a
n
ag
em
e
n
t
[
2
1
]
.
Hy
b
r
id
AI
a
n
d
m
etah
e
u
r
is
tic
ap
p
r
o
ac
h
es
h
av
e
also
b
ee
n
ex
p
lo
r
ed
to
o
p
tim
ize
r
o
u
tin
g
p
at
h
s
an
d
clu
s
ter
f
o
r
m
atio
n
u
n
d
er
d
y
n
am
i
c
n
etwo
r
k
co
n
d
itio
n
s
[
2
2
]
.
AI
-
en
ab
led
clu
s
ter
in
g
an
d
r
o
u
tin
g
m
ec
h
a
n
is
m
s
h
av
e
r
ec
en
tly
b
ee
n
p
r
o
p
o
s
ed
to
en
h
an
ce
ad
a
p
tab
ilit
y
an
d
en
er
g
y
b
alan
cin
g
in
I
o
T
-
b
ased
W
SN
s
[
2
3
]
.
C
o
m
p
r
eh
en
s
iv
e
s
u
r
v
ey
s
f
r
o
m
2
0
2
2
–
2
0
2
4
r
e
p
o
r
t
th
at
ML
-
b
ased
r
o
u
tin
g
p
r
o
to
co
ls
o
u
tp
e
r
f
o
r
m
class
ical
m
eth
o
d
s
in
ter
m
s
o
f
s
ca
lab
ilit
y
,
ad
ap
tab
ilit
y
,
an
d
n
etwo
r
k
life
tim
e
u
n
d
e
r
v
ar
y
i
n
g
tr
af
f
ic
an
d
to
p
o
lo
g
y
co
n
d
itio
n
s
[
4
]
,
[
6
]
.
Ho
wev
er
,
ch
allen
g
es
r
em
ain
in
ter
m
s
o
f
co
m
p
u
tatio
n
al
co
s
t,
d
ata
av
ail
ab
ilit
y
f
o
r
tr
ain
in
g
,
an
d
th
e
o
v
er
h
ea
d
r
e
q
u
ir
ed
f
o
r
m
o
d
el
d
is
tr
ib
u
tio
n
an
d
u
p
d
ates w
ith
in
th
e
n
etwo
r
k
.
2
.
4
.
Resea
r
ch
g
a
p a
nd
mo
t
i
v
a
t
io
n
T
h
e
liter
atu
r
e
r
ev
ea
ls
th
at
wh
ile
AI
-
en
ab
led
r
o
u
tin
g
o
f
f
er
s
s
ig
n
if
ican
t
p
er
f
o
r
m
an
ce
im
p
r
o
v
em
en
ts
,
ex
is
tin
g
ap
p
r
o
ac
h
es
o
f
ten
n
e
g
lect
r
ea
l
-
tim
e
c
o
n
s
tr
ain
ts
an
d
im
p
o
s
e
h
ea
v
y
co
m
p
u
tatio
n
al
b
u
r
d
e
n
s
o
n
s
en
s
o
r
n
o
d
es.
Mo
r
eo
v
er
,
f
ew
s
tu
d
ies
h
o
lis
tically
in
teg
r
ate
en
er
g
y
p
r
ed
ictio
n
,
ad
ap
tiv
e
clu
s
ter
in
g
,
an
d
in
tellig
en
t
p
ath
s
elec
tio
n
in
to
a
u
n
if
ie
d
f
r
am
ewo
r
k
with
m
in
im
al
c
o
m
m
u
n
icatio
n
o
v
e
r
h
ea
d
.
T
h
ese
o
b
s
er
v
atio
n
s
led
to
t
h
e
d
ev
elo
p
m
e
n
t
o
f
AI
-
E
E
R
P
,
a
r
o
u
tin
g
p
r
o
to
co
l
th
at
b
len
d
s
lig
h
tweig
h
t
ML
m
o
d
els
with
en
er
g
y
-
awa
r
e
r
o
u
tin
g
s
tr
ateg
ies
to
im
p
r
o
v
e
b
o
th
n
etwo
r
k
s
tab
ilit
y
an
d
o
p
e
r
atio
n
al
life
tim
e.
L
ea
r
n
in
g
-
ass
is
te
d
r
o
u
tin
g
s
tr
ateg
ies
h
av
e
d
em
o
n
s
tr
ated
im
p
r
o
v
e
d
lo
n
g
-
ter
m
s
u
s
tain
ab
ilit
y
b
y
p
r
ed
ictin
g
n
o
d
e
b
e
h
av
io
r
an
d
a
v
o
id
in
g
p
r
em
at
u
r
e
en
er
g
y
d
ep
letio
n
[
2
4
]
.
R
ec
en
t
s
tu
d
ies
em
p
h
asize
th
e
im
p
o
r
tan
ce
o
f
lig
h
tweig
h
t
m
ac
h
i
n
e
-
l
ea
r
n
in
g
m
o
d
els
to
en
s
u
r
e
f
ea
s
ib
ilit
y
o
n
r
eso
u
r
ce
-
co
n
s
tr
ain
ed
s
en
s
o
r
n
o
d
es
[
2
5
]
.
3.
P
RO
P
O
SE
D
M
E
T
H
O
DO
L
O
G
Y
T
h
e
p
r
o
p
o
s
ed
AI
-
E
E
R
P
in
tr
o
d
u
ce
s
an
in
tellig
en
t,
ad
ap
ti
v
e
r
o
u
tin
g
f
r
am
ewo
r
k
th
at
in
teg
r
ates
m
ac
h
in
e
lear
n
in
g
-
b
ased
d
ec
is
io
n
-
m
ak
in
g
with
h
ier
ar
ch
ical
clu
s
ter
in
g
to
o
p
tim
ize
en
er
g
y
co
n
s
u
m
p
tio
n
an
d
ex
ten
d
th
e
o
p
e
r
atio
n
al
d
u
r
atio
n
o
f
W
SNs
.
T
h
e
m
eth
o
d
o
lo
g
y
co
n
s
is
ts
o
f
s
ix
p
r
im
ar
y
co
m
p
o
n
e
n
ts
:
n
etwo
r
k
m
o
d
el
d
ef
in
itio
n
,
en
er
g
y
m
o
d
el
f
o
r
m
u
latio
n
,
d
ata
co
llectio
n
,
m
ac
h
in
e
lear
n
in
g
–
b
ased
p
r
ed
ictio
n
,
ad
ap
tiv
e
clu
s
ter
in
g
,
an
d
in
tellig
en
t r
o
u
t
in
g
p
ath
s
elec
tio
n
.
3
.
1
.
Net
w
o
rk
mo
del
I
n
th
e
p
r
o
p
o
s
ed
m
o
d
el,
th
e
W
SN
co
m
p
r
is
es
N
s
en
s
o
r
n
o
d
es
r
an
d
o
m
ly
s
ca
tter
ed
o
v
er
a
two
-
d
im
en
s
io
n
al
p
lan
e
.
Ass
u
m
p
tio
n
s
o
f
th
e
s
im
u
latio
n
m
o
d
el
in
clu
d
e:
i)
On
ce
d
ep
lo
y
ed
,
t
h
e
s
en
s
o
r
n
o
d
es
ar
e
s
tatio
n
ar
y
an
d
h
av
e
ac
ce
s
s
to
th
eir
lo
ca
tio
n
in
f
o
r
m
atio
n
.
ii)
T
h
e
b
ase
s
ta
tio
n
d
o
es
n
o
t
m
o
v
e
an
d
is
p
lace
d
eith
er
in
s
id
e
o
r
o
u
ts
id
e
th
e
s
e
n
s
in
g
f
ield
.
iii)
No
d
es
g
en
er
at
e
d
ata
p
er
io
d
ically
an
d
c
o
m
m
u
n
icate
v
ia
s
in
g
le
-
h
o
p
o
r
m
u
lti
-
h
o
p
h
ier
ar
c
h
ical
r
o
u
tin
g
.
iv
)
No
d
es
s
tar
t
with
u
n
if
o
r
m
i
n
itial
en
er
g
y
an
d
f
o
llo
w
th
e
f
ir
s
t
-
o
r
d
er
r
ad
io
en
er
g
y
m
o
d
el.
an
d
v
)
T
h
e
b
ase
s
tatio
n
(
B
S)
p
o
s
s
es
s
es
h
ig
h
er
co
m
p
u
tatio
n
al
p
o
wer
an
d
s
to
r
es
h
is
to
r
ical
n
etwo
r
k
d
ata
f
o
r
t
r
ain
in
g
ML
m
o
d
els.
3
.
2
.
E
nerg
y
mo
del
AI
-
E
E
R
P
ad
o
p
ts
th
e
class
ica
l
f
ir
s
t
-
o
r
d
er
r
ad
i
o
to
ac
co
u
n
t
f
o
r
th
e
en
er
g
y
r
e
q
u
ir
ed
to
tr
a
n
s
m
it
an
d
r
ec
eiv
e
k
-
b
it d
ata
is
:
a.
T
r
an
s
m
is
s
io
n
en
er
g
y
:
(
,
)
=
⋅
+
⋅
⋅
b.
R
ec
ep
tio
n
en
er
g
y
:
(
)
=
⋅
w
h
er
e
=
elec
tr
o
n
ic
cir
cu
itry
co
s
t
,
=
am
p
lifie
r
co
s
t,
in
d
icate
s
th
e
p
ath
-
lo
s
s
ex
p
o
n
en
t
(
2
f
o
r
f
r
ee
-
s
p
ac
e
co
m
m
u
n
icatio
n
a
n
d
4
f
o
r
m
u
ltip
at
h
p
r
o
p
a
g
atio
n
)
,
d
=
d
is
tan
ce
b
etwe
en
s
en
d
er
an
d
r
e
ce
iv
er
.
T
h
e
m
o
d
el
h
elp
s
esti
m
ate
th
e
en
er
g
y
co
n
s
u
m
ed
d
u
r
in
g
clu
s
ter
in
g
an
d
r
o
u
tin
g
ac
tiv
ities
.
3
.
3
.
Da
t
a
c
o
llect
io
n a
nd
f
ea
t
ur
e
ex
t
ra
ct
io
n
E
ac
h
s
en
s
o
r
n
o
d
e
p
er
io
d
ically
ca
p
tu
r
es
lo
ca
l
p
ar
am
eter
s
an
d
f
o
r
war
d
s
ess
en
tial
f
ea
tu
r
es
to
th
e
B
S
o
r
clu
s
ter
h
ea
d
f
o
r
ML
p
r
o
ce
s
s
in
g
.
E
x
tr
ac
te
d
f
ea
tu
r
es
in
cl
u
d
e:
r
esid
u
al
en
er
g
y
,
n
o
d
e
d
e
g
r
ee
/n
eig
h
b
o
r
h
o
o
d
d
en
s
ity
,
lin
k
q
u
ality
(
R
SS
I
/
L
QI
)
,
d
is
tan
ce
to
C
H
an
d
B
S
,
h
is
to
r
ical
en
er
g
y
co
n
s
u
m
p
tio
n
tr
en
d
,
b
u
f
f
er
o
cc
u
p
an
cy
,
a
n
d
p
ac
k
et
s
u
c
ce
s
s
/f
ailu
r
e
r
atio
.
T
h
ese
f
ea
tu
r
es f
o
r
m
th
e
in
p
u
t d
ataset
f
o
r
ML
-
b
ased
p
r
ed
ictio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
AI
-
en
a
b
led
en
erg
y
-
a
w
a
r
e
r
o
u
tin
g
a
p
p
r
o
a
ch
fo
r
fu
tu
r
e
-
w
ir
eless
s
en
s
o
r
n
et
w
o
r
ks
(
S
h
a
msh
er S
in
g
h
)
1547
3
.
4
.
M
a
chine
lea
rning
-
ba
s
ed
r
esi
du
a
l e
nerg
y
a
nd
lin
k q
ua
lity
pr
edict
io
n
AI
-
E
E
R
P
in
co
r
p
o
r
ates lig
h
twe
ig
h
t M
L
m
o
d
els to
p
er
f
o
r
m
tw
o
cr
itical
p
r
ed
ictio
n
s
:
a.
R
esid
u
al
en
er
g
y
p
r
ed
ictio
n
A
r
eg
r
ess
io
n
-
b
ased
p
r
ed
ictio
n
m
o
d
el
u
s
in
g
s
im
p
le
o
r
en
s
em
b
le
lear
n
in
g
tech
n
iq
u
es
p
r
ed
icts
th
e
f
u
tu
r
e
e
n
er
g
y
lev
el
o
f
ea
ch
n
o
d
e:
=
(
,
ℎ
,
,
,
)
T
h
e
B
S tr
ain
s
th
e
m
o
d
el
o
f
f
lin
e,
wh
ile
s
en
s
o
r
n
o
d
es e
x
ec
u
te
lig
h
tweig
h
t in
f
er
e
n
ce
.
b.
L
in
k
q
u
ality
p
r
ed
ictio
n
A
class
if
ier
p
r
ed
icts
lin
k
r
eliab
ilit
y
u
s
in
g
m
etr
ics s
u
ch
as R
SS
I
,
n
o
is
e,
an
d
p
ac
k
et
lo
s
s
:
=
(
,
,
)
T
h
ese
p
r
ed
ictio
n
s
g
u
id
e
C
H
s
elec
tio
n
an
d
r
o
u
te
f
o
r
m
atio
n
.
3
.
5
.
Ada
ptiv
e
clus
t
er
-
hea
d select
io
n us
ing
AI
AI
-
E
E
R
P
p
er
f
o
r
m
s
clu
s
ter
-
h
e
ad
(
C
H)
s
elec
tio
n
u
s
in
g
a
h
y
b
r
id
d
ec
is
io
n
m
etr
ic:
=
⋅
+
⋅
+
⋅
1
w
h
er
e
α
,
β
,
γ
ar
e
s
y
s
tem
-
d
ef
i
n
ed
weig
h
ts
.
Hig
h
er
s
co
r
e
i
n
d
icate
s
b
etter
s
u
itab
ilit
y
as
C
H.
No
d
es
with
to
p
-
r
an
k
ed
s
co
r
es
b
ec
o
m
e
cl
u
s
ter
-
h
ea
d
s
,
e
n
s
u
r
in
g
:
i)
b
alan
ce
d
en
er
g
y
co
n
s
u
m
p
tio
n
,
ii)
b
e
tter
r
eliab
ilit
y
,
an
d
iii)
r
ed
u
ce
d
r
e
-
clu
s
ter
in
g
f
r
eq
u
en
cy
.
C
o
m
p
ar
ed
to
L
E
AC
H/HE
E
D,
th
is
r
ed
u
ce
s
r
an
d
o
m
n
ess
an
d
en
h
an
ce
s
s
tab
ilit
y
.
3
.
6
.
I
nte
llig
ent
r
o
uting
pa
t
h
f
o
rma
t
io
n
On
ce
C
Hs ar
e
s
elec
ted
,
AI
-
E
E
R
P
co
n
s
tr
u
cts an
o
p
tim
ized
m
u
lti
-
h
o
p
r
o
u
te
f
r
o
m
C
Hs to
th
e
B
S.
Step
s
:
a.
C
Hs b
r
o
ad
ca
s
t th
eir
p
r
ed
icted
en
er
g
y
an
d
lin
k
s
co
r
es.
b.
Neig
h
b
o
r
in
g
C
Hs ev
alu
ate
p
o
t
en
tial f
o
r
war
d
in
g
ca
n
d
id
ates u
s
in
g
:
_
=
⋅
+
⋅
−
⋅
c.
C
H
s
elec
t
s
th
e
n
ex
t h
o
p
th
at
m
ax
im
izes
_
.
d.
Data
is
ag
g
r
eg
ated
at
C
Hs an
d
f
o
r
war
d
e
d
in
a
n
en
er
g
y
-
awa
r
e
m
u
lti
-
h
o
p
f
ash
io
n
.
T
h
is
en
s
u
r
es m
in
im
al
en
er
g
y
wastag
e
an
d
av
o
id
s
lo
w
-
e
n
er
g
y
n
o
d
es.
3
.
7
.
O
pera
t
io
na
l
ph
a
s
es
o
f
AI
-
E
E
RP
T
h
e
p
r
o
to
co
l o
p
er
ates in
f
o
u
r
p
h
ases
:
a.
Setu
p
p
h
ase
-
ML
m
o
d
els in
itialized
at
B
S
-
C
lu
s
ter
in
g
p
ar
am
eter
s
d
ef
in
e
d
-
I
n
itial f
ea
tu
r
e
co
llectio
n
p
er
f
o
r
m
ed
b.
ML
-
B
ased
p
r
ed
ictio
n
p
h
ase
-
No
d
es c
o
m
p
u
te
o
r
r
ec
ei
v
e
u
p
d
ated
en
er
g
y
/lin
k
p
r
ed
ictio
n
s
-
C
H
s
elec
tio
n
p
r
o
b
ab
ilit
y
co
m
p
u
ted
c.
Ad
ap
tiv
e
clu
s
ter
in
g
p
h
ase
-
No
d
es jo
in
th
e
n
ea
r
est C
H
b
ased
o
n
m
u
lti
-
cr
iter
ia
s
co
r
in
g
-
C
lu
s
ter
tab
les u
p
d
ated
d.
Data
tr
an
s
m
is
s
io
n
p
h
ase
-
C
H
ag
g
r
eg
ates d
ata
-
Mu
lti
-
h
o
p
r
o
u
tin
g
to
B
S u
s
in
g
AI
-
o
p
tim
ized
p
ath
s
-
R
o
u
n
d
co
m
p
letes a
n
d
en
er
g
y
t
ab
les u
p
d
ate
3
.
8
.
Adv
a
nta
g
es o
f
AI
-
E
E
R
P
T
h
e
ad
v
an
ta
g
es
o
f
AI
-
E
E
R
P
in
clu
d
e:
i)
Hig
h
er
en
er
g
y
ef
f
ic
ien
cy
v
ia
p
r
ed
ictiv
e
r
o
u
tin
g
,
ii)
r
ed
u
ce
d
p
ac
k
et
lo
s
s
d
u
e
to
lin
k
-
q
u
ality
p
r
ed
ictio
n
,
iii)
lo
n
g
er
n
et
wo
r
k
life
tim
e
f
r
o
m
b
alan
ce
d
C
H
s
elec
tio
n
,
iv
)
f
ewe
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
5
4
3
-
1
5
6
1
1548
re
-
clu
s
ter
in
g
o
p
er
atio
n
s
co
m
p
ar
ed
t
o
L
E
AC
H/HE
E
D
,
an
d
v
)
s
ca
lab
le
a
n
d
a
d
ap
tab
le
to
d
iv
er
s
e
W
SN
d
ep
lo
y
m
e
n
ts
.
3
.
9
.
M
a
chine
lea
rning
a
nd
mo
del v
a
lid
a
t
io
n
T
o
v
alid
ate
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
m
ac
h
in
e
lear
n
in
g
co
m
p
o
n
en
t,
th
e
p
r
ed
ictio
n
p
e
r
f
o
r
m
an
ce
o
f
th
e
r
esid
u
al
en
er
g
y
m
o
d
el
is
ev
alu
ated
in
d
ep
e
n
d
en
tly
o
f
th
e
r
o
u
tin
g
m
ec
h
an
is
m
.
T
h
e
r
e
g
r
ess
io
n
m
o
d
el
is
ass
es
s
ed
u
s
in
g
m
ea
n
ab
s
o
lu
te
er
r
o
r
(
MA
E
)
,
r
o
o
t
m
ea
n
s
q
u
ar
e
er
r
o
r
(
R
MSE
)
,
an
d
co
ef
f
icien
t
o
f
d
eter
m
in
atio
n
(
R
²)
.
T
h
e
r
esu
lts
in
d
icate
th
at
th
e
m
o
d
el
ac
h
iev
es
lo
w
p
r
ed
ict
io
n
er
r
o
r
,
c
o
n
f
ir
m
in
g
its
ca
p
ab
ilit
y
to
ac
cu
r
ately
esti
m
ate
f
u
tu
r
e
n
o
d
e
en
er
g
y
lev
els
u
n
d
er
d
y
n
a
m
ic
n
etwo
r
k
co
n
d
itio
n
s
.
Simi
lar
ly
,
th
e
lin
k
q
u
ality
p
r
ed
ictio
n
m
o
d
el
is
ev
alu
ated
u
s
in
g
class
if
icatio
n
m
etr
ics
s
u
ch
as
ac
cu
r
ac
y
an
d
F1
-
s
co
r
e,
d
em
o
n
s
tr
atin
g
r
eliab
le
id
en
tific
atio
n
o
f
s
tab
le
co
m
m
u
n
icatio
n
lin
k
s
.
T
h
e
s
e
r
esu
lts
co
n
f
ir
m
th
at
th
e
l
ea
r
n
in
g
c
o
m
p
o
n
en
t
p
r
o
v
id
es m
ea
n
in
g
f
u
l p
r
e
d
ictiv
e
ca
p
ab
ilit
y
an
d
is
n
o
t m
er
ely
a
h
eu
r
is
tic
ap
p
r
o
x
im
atio
n
.
3
.
1
0
.
Co
mp
uta
t
io
na
l
co
mp
lex
it
y
a
na
ly
s
is
T
h
e
co
m
p
u
tatio
n
al
co
m
p
lex
i
ty
o
f
th
e
p
r
o
p
o
s
ed
AI
-
E
E
R
P
p
r
o
to
co
l
is
an
aly
ze
d
to
e
v
alu
ate
its
f
ea
s
ib
ilit
y
f
o
r
r
eso
u
r
ce
-
co
n
s
tr
ain
ed
s
en
s
o
r
n
o
d
es.
T
h
e
c
lu
s
ter
-
h
ea
d
s
elec
tio
n
p
r
o
ce
s
s
o
p
e
r
ates
with
a
co
m
p
lex
ity
o
f
O(
N)
,
wh
e
r
e
N
r
ep
r
esen
ts
th
e
n
u
m
b
er
o
f
n
o
d
es.
T
h
e
r
o
u
tin
g
d
ec
is
io
n
p
h
ase
in
v
o
lv
es
n
eig
h
b
o
r
ev
alu
atio
n
,
r
esu
ltin
g
in
a
co
m
p
lex
ity
o
f
O(
k
)
,
wh
er
e
k
is
th
e
n
u
m
b
er
o
f
n
eig
h
b
o
r
in
g
n
o
d
es.
T
h
e
m
a
ch
in
e
lear
n
in
g
in
f
er
e
n
ce
p
r
o
ce
s
s
is
d
esig
n
ed
to
b
e
lig
h
tweig
h
t,
with
co
n
s
tan
t
-
tim
e
p
r
ed
ictio
n
c
o
m
p
lex
ity
O(
1
)
p
e
r
n
o
d
e
wh
en
u
s
in
g
p
r
e
-
tr
ain
ed
m
o
d
els.
Sin
ce
m
o
d
el
tr
ain
in
g
is
p
er
f
o
r
m
e
d
at
th
e
b
ase
s
tatio
n
,
th
e
c
o
m
p
u
tatio
n
al
b
u
r
d
e
n
o
n
s
en
s
o
r
n
o
d
es
r
em
a
in
s
m
in
im
al.
T
h
er
ef
o
r
e,
th
e
o
v
er
all
co
m
p
lex
ity
o
f
th
e
p
r
o
t
o
co
l
r
em
ain
s
lin
ea
r
an
d
s
u
itab
le
f
o
r
lar
g
e
-
s
ca
le
d
e
p
lo
y
m
en
ts
with
o
u
t sig
n
if
ican
t
co
m
p
u
tatio
n
al
o
v
er
h
ea
d
.
4.
CO
M
P
UT
AT
I
O
NAL CO
M
P
L
E
X
I
T
Y
A
NAL
YSI
S
T
h
e
co
m
p
u
tatio
n
al
co
m
p
lex
ity
o
f
A
I
-
E
E
R
P
is
an
aly
ze
d
t
o
ev
alu
ate
its
s
u
itab
ilit
y
f
o
r
r
eso
u
r
ce
-
co
n
s
tr
ain
ed
s
en
s
o
r
n
o
d
es.
T
h
e
clu
s
ter
-
h
ea
d
s
elec
tio
n
p
r
o
ce
s
s
o
p
er
ates
with
a
tim
e
co
m
p
lex
ity
o
f
O(
N)
,
wh
er
e
N
is
th
e
n
u
m
b
e
r
o
f
n
o
d
es.
T
h
e
r
o
u
tin
g
d
ec
is
io
n
p
h
ase
in
v
o
l
v
es
ev
alu
atin
g
n
eig
h
b
o
r
in
g
n
o
d
es,
r
esu
ltin
g
in
a
co
m
p
lex
ity
o
f
O(
k
)
,
wh
er
e
k
r
ep
r
esen
ts
th
e
n
u
m
b
e
r
o
f
n
eig
h
b
o
r
s
.
T
h
e
m
ac
h
in
e
lear
n
in
g
i
n
f
er
en
ce
p
r
o
ce
s
s
is
d
esig
n
ed
to
b
e
lig
h
tweig
h
t,
with
co
n
s
tan
t
-
tim
e
co
m
p
le
x
ity
O(
1
)
p
er
n
o
d
e
u
s
in
g
p
r
e
-
tr
ain
ed
m
o
d
els.
Sin
ce
m
o
d
el
tr
ai
n
in
g
is
p
er
f
o
r
m
e
d
at
th
e
b
ase
s
tatio
n
,
t
h
e
c
o
m
p
u
tat
io
n
al
b
u
r
d
e
n
o
n
i
n
d
iv
id
u
al
s
en
s
o
r
n
o
d
es
r
em
ai
n
s
m
in
im
al.
Ov
er
all,
th
e
p
r
o
p
o
s
ed
p
r
o
to
c
o
l
m
ai
n
tain
s
lin
e
ar
co
m
p
lex
ity
a
n
d
is
s
ca
la
b
le
f
o
r
lar
g
e
-
s
ca
le
d
ep
lo
y
m
e
n
ts
with
o
u
t
in
tr
o
d
u
cin
g
s
ig
n
if
ican
t
co
m
p
u
tatio
n
a
l
o
v
er
h
ea
d
.
Alg
o
r
ith
m
1
o
u
tl
in
es
th
e
AI
-
E
E
R
P
-
b
ased
ad
ap
tiv
e
clu
s
ter
-
h
ea
d
s
e
lectio
n
an
d
e
n
er
g
y
-
awa
r
e
r
o
u
ti
n
g
p
r
o
ce
d
u
r
e.
Alg
o
r
ith
m
1
.
AI
-
E
E
R
P
-
b
ased
ad
ap
tiv
e
clu
s
ter
-
h
ea
d
s
elec
tio
n
an
d
e
n
er
g
y
-
awa
r
e
r
o
u
tin
g
p
r
o
ce
d
u
r
e
INPUT:
N
// Number of sensing units
BS
// Base Station (high compute)
R_max
// Maximum number of rounds
k_bits
// Packet size (bits)
E_initial[i]
// energy available at node initialization i
E_elecs, E_amp
// Radio model parameters
alpha, beta, gamma
// Weights for CH_score
lambda, mu, delta
// Weights for Route_score
ML_model_residual
// Trained residual
-
energy predictor (at BS / edge)
ML_model_link
// Trained link
-
quality predictor (at BS / edge)
GLOBAL DATA STRUCTURES:
Node[i].E
// Current residual energy
Node[i].pos
// Position of node i
Node[i].neighbours
// Neighbor list (IDs, distances)
ClusterHeads
// Set of elected CH ids each round
ClusterTable
// Cluster membership mapping
RoutingTable
// Next
-
hop info for CHs
PROCEDURE MAIN()
Initialize nodes 1..N with E_initial, positions, neighbors
BS.train_models(historical_data)
// offline / periodic training
for round=1 to R_max do
// Phase 1: Feature Collection
for each node i in 1..N do
features_i=collect_local_features(i)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
AI
-
en
a
b
led
en
erg
y
-
a
w
a
r
e
r
o
u
tin
g
a
p
p
r
o
a
ch
fo
r
fu
tu
r
e
-
w
ir
eless
s
en
s
o
r
n
et
w
o
r
ks
(
S
h
a
msh
er S
in
g
h
)
1549
// optionally: nodes send features to BS/edge (compressed)
end for
// Phase 2: Prediction (at BS / Edge or lightweight on
-
node)
for each node i in 1..N do
E_pred[i]=
infer_residual_energy(ML_model_residual, features_i)
LQ_score[i]=infer_link_quality(ML_model_link, features_i)
end for
// Phase 3: Cluster
-
Head (CH) Selection
ClusterHeads={}
for each node i in 1..N do
CH_score[i]=alpha * normalize(E_pred[i])
+ beta * normalize(LQ_score[i])
+ gamma * normalize(1 / distance(i, BS))
end for
ClusterHeads=select_top_k(CH_score, strategy=”threshold/percentile”)
broadcast_CH_announcement(ClusterHeads)
// Phase 4: Cluster Formation
ClusterTable={}
for every non
-
CH node j in 1..N do
choose CH c in ClusterHeads that maximizes:
Join_score=w1 * CH_score[c]
–
w2 * distance(j, c)
ClusterTable[c].append(j)
send_join_request(j, c)
end for
// Phase 5: Intra
-
cluster Scheduling (TDMA)
for each CH c in ClusterHeads do
generate_TDMA_schedule(ClusterTable[c])
inform_members(ClusterTable[c])
end for
// Phase 6: Data Transmission (Sensing
-
> CH
-
> BS via multi
-
hop)
for each time
-
slot in TDMA do
for each member node m transmitting to its CH c do
transmit_packet(m, c, k_bits)
Node[m].E
-
= energy_tx(k_bits, distance(m,c))
Node[c].E
-
= energy_rx(k_bits)
end for
// CH aggregates data
for every CH c do
aggregate_energy=energy_aggregate(k_bits, num_packets)
Node[c].E
-
= aggregate_energy
// CH selects next hop toward BS
n
e
x
t
_
h
op
=s
e
l
e
c
t
_
ne
x
t
_
h
op
(
c
,
C
l
u
s
t
er
H
e
a
d
s
,
E_
p
r
e
d
,
L
Q
_
sc
o
r
e
,
l
a
m
b
d
a
,
mu
,
d
e
lt
a
)
if next_hop == BS then
transmit_packet(c, BS, k_bits_agg)
Node[c].E
-
= energy_tx(k_bits_agg, distance(c, BS))
else
transmit_packet(c, next_hop, k_bits_agg)
Node[c].E
-
= energy_tx(k_bits_agg, distance(c, next_hop))
end if
end for
end for
// Phase 7: Energy Update & Node Death Handling
for each node i do
if Node[i].E <= 0 then
mark_node_dead(i)
end if
end for
// Phase 8: Model Update (Optional)
if (round % model_update_period == 0) then
gather_labeled_samples()
BS.retrain_models(new_data)
distribute_model_updates_if_needed()
end if
end for
// rounds
END PROCEDURE
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
5
4
3
-
1
5
6
1
1550
//
----------------
Helper Functions
----------------
FUNCTION collect_local_features(i):
features={
residual_energy: Node[i].E,
energy_history: get_energy_history(i),
neighbor_count: len(Node[i].neighbours),
avg_RSSI: measure_avg_RSSI(i),
PDR: measure_packet_delivery_ratio(i),
dist_to_BS: distance(i, BS),
buffer_occupancy: Node[i].buffer_size
}
return features
END FUNCTION
FUNCTION infer_residual_energy(model, features):
/
/
I
f
o
n
-
n
o
de
i
n
fe
r
e
n
ce
i
s
h
e
av
y
,
u
se
c
o
mp
r
e
s
s
ed
m
o
d
e
l
or
r
e
q
ue
s
t
pr
e
d
i
c
ti
o
n
f
r
o
m
B
S
/
e
d
ge
return model.predict(features)
END FUNCTION
FUNCTION infer_link_quality(model, features):
return model.predict(features)
END FUNCTION
FUNCTION energy_tx(k, d):
// First
-
order radio model (use threshold if required)
if d < d0 then
return E_elecs * k + E_amp_fs * k * d^2
else
return E_elecs * k + E_amp_mp * k * d^4
end if
END FUNCTION
FUNCTION energy_rx(k):
return E_elecs * k
END FUNCTION
FUNCTION select_next_hop(c, CH_set, E_pred, LQ_score, lambda, mu, delta):
best_score=
-
inf
best_node=BS // default to BS if direct is best
for each candidate h in CH_set U {BS} where h != c do
dist=distance(c, h)
score=lambda * normalize(E_pred[h]) + mu * normalize(LQ_score[h])
–
delta *
normalize(dist)
if score > best_score and Node[h] is alive then
best_score=score
best_node=h
end if
end for
return best_node
END FUNCTION
FUNCTION select_top_k(CH_score, strategy):
// Strategy may be: fixed percentage, score threshold, or dynamic number based on N
// Example: choose top P% of nodes by CH_score
sorted_nodes=sort_descending(CH_score)
k=
ceil(P_percent * N)
return sorted_nodes[1..k]
END FUNCTION
FUNCTION normalize(x):
// map x to [0,1] using min
-
max normalization across available values
return (x
–
min_val)/(max_val
–
min_val + epsilon)
END FUNCTION
5.
SI
M
UL
A
T
I
O
N
SE
T
UP
5
.
1
.
Sim
ula
t
io
n o
bje
ct
iv
es
T
h
e
m
ain
g
o
al
o
f
th
e
s
im
u
latio
n
s
tu
d
y
is
to
ev
alu
ate
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
p
r
o
p
o
s
ed
AI
-
E
n
ab
led
m
eth
o
d
u
n
d
e
r
v
a
r
y
in
g
n
etwo
r
k
co
n
d
itio
n
s
in
co
m
p
a
r
is
o
n
wi
th
p
r
o
m
in
en
t
b
aselin
e
s
ch
em
e
s
,
n
am
ely
L
E
AC
H,
HE
E
D,
an
d
PEGA
SIS.
T
h
e
p
er
f
o
r
m
a
n
ce
ass
ess
m
en
t
is
ca
r
r
ied
o
u
t
co
n
s
id
er
i
n
g
t
h
e
f
o
llo
win
g
k
e
y
ev
al
u
atio
n
m
etr
ics:
i)
s
y
s
tem
life
tim
e,
m
e
asu
r
ed
u
s
in
g
th
e
f
ir
s
t
n
o
d
e
d
e
ath
(
FND)
,
h
al
f
n
o
d
e
d
ea
th
(
H
ND)
,
an
d
last
n
o
d
e
d
ea
th
(
L
ND)
,
ii)
a
v
er
ag
e
r
esid
u
al
en
er
g
y
v
a
r
iatio
n
ac
r
o
s
s
s
im
u
latio
n
r
o
u
n
d
s
,
iii)
p
ac
k
et
d
eliv
er
y
r
atio
(
PDR
)
,
iv
)
t
h
r
o
u
g
h
p
u
t
a
n
d
av
er
ag
e
en
d
-
to
-
en
d
d
elay
,
v
)
r
o
u
tin
g
a
n
d
co
n
tr
o
l
p
ac
k
et
o
v
er
h
ea
d
,
a
n
d
v
i)
e
n
er
g
y
f
air
n
ess
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
AI
-
en
a
b
led
en
erg
y
-
a
w
a
r
e
r
o
u
tin
g
a
p
p
r
o
a
ch
fo
r
fu
tu
r
e
-
w
ir
eless
s
en
s
o
r
n
et
w
o
r
ks
(
S
h
a
msh
er S
in
g
h
)
1551
ev
alu
ated
u
s
in
g
th
e
u
n
e
v
en
n
e
s
s
in
r
esid
u
al
en
er
g
y
d
is
tr
ib
u
tio
n
.
C
o
llectiv
ely
,
th
ese
m
etr
ic
s
p
r
o
v
id
e
a
h
o
lis
tic
ev
alu
atio
n
o
f
en
er
g
y
ef
f
icien
c
y
,
r
o
u
tin
g
r
eliab
ilit
y
,
an
d
th
e
o
v
er
all
p
er
f
o
r
m
an
ce
o
f
th
e
n
et
wo
r
k
.
5
.
2
.
NS
-
3
mo
du
les a
nd
f
un
ct
io
na
l c
o
mp
o
nents
T
h
e
s
im
u
latio
n
e
n
v
ir
o
n
m
en
t
is
d
ev
elo
p
e
d
u
s
in
g
th
e
NS
-
3
n
etwo
r
k
s
im
u
lato
r
,
in
c
o
r
p
o
r
atin
g
th
e
f
o
llo
win
g
co
r
e
m
o
d
u
les an
d
f
e
atu
r
es:
a.
T
h
e
co
r
e,
n
etwo
r
k
,
I
n
ter
n
et,
an
d
Wi
-
Fi
m
o
d
u
les
ar
e
em
p
lo
y
ed
f
o
r
n
o
d
e
cr
ea
tio
n
,
p
r
o
to
co
l
s
tack
co
n
f
ig
u
r
atio
n
,
wir
eless
co
m
m
u
n
icatio
n
,
a
n
d
UDP/TCP
tr
af
f
ic
g
en
er
atio
n
.
b.
T
h
e
NS
-
3
en
er
g
y
f
r
am
ewo
r
k
,
in
clu
d
in
g
th
e
B
a
s
icE
n
er
g
yS
o
u
r
ce
an
d
Dev
iceE
n
er
g
yMo
d
el
,
is
u
tili
ze
d
to
ac
cu
r
ately
m
o
d
el
en
er
g
y
co
n
s
u
m
p
tio
n
d
u
r
i
n
g
tr
a
n
s
m
is
s
io
n
,
r
ec
ep
tio
n
,
i
d
le
lis
ten
in
g
,
an
d
d
ata
ag
g
r
eg
atio
n
p
r
o
ce
s
s
es.
c.
T
h
e
m
o
b
ilit
y
m
o
d
u
le
is
i
n
cl
u
d
ed
t
o
s
u
p
p
o
r
t
s
e
m
i
-
d
y
n
a
m
i
c
o
r
m
o
b
il
e
cl
u
s
te
r
-
h
ea
d
s
ce
n
ar
io
s
wh
e
n
r
e
q
u
ir
ed
.
d.
T
h
e
tr
ac
in
g
an
d
f
lo
w
m
o
n
ito
r
u
tili
ties
ar
e
u
s
ed
to
r
ec
o
r
d
an
d
an
aly
ze
k
e
y
m
ea
s
u
r
es
o
f
p
ac
k
et
d
eliv
er
y
r
atio
,
th
r
o
u
g
h
p
u
t,
a
n
d
en
d
-
to
-
e
n
d
d
elay
.
e.
Ad
d
itio
n
ally
,
o
p
tio
n
al
m
o
d
el
-
u
p
d
ate
r
o
u
tin
es
ar
e
in
co
r
p
o
r
a
ted
with
in
th
e
s
im
u
latio
n
s
cr
ip
t
to
em
u
late
b
ase
-
s
tatio
n
-
ass
is
ted
tr
ain
in
g
an
d
p
er
i
o
d
ic
d
is
tr
ib
u
tio
n
o
f
AI
p
r
ed
ictio
n
p
ar
am
eter
s
.
5
.
3
.
Sim
ula
t
io
n pa
ra
met
er
s
T
h
e
s
im
u
latio
n
en
v
i
r
o
n
m
e
n
t
a
n
d
ex
p
e
r
im
en
tal
co
n
f
ig
u
r
atio
n
u
s
ed
to
ev
alu
ate
th
e
p
r
o
p
o
s
ed
AI
-
E
E
R
P
p
r
o
to
co
l
ar
e
s
u
m
m
ar
ized
in
T
ab
le
1
.
T
h
e
lis
ted
p
ar
a
m
eter
s
d
ef
in
e
t
h
e
n
etwo
r
k
s
ettin
g
s
,
co
m
m
u
n
icatio
n
m
o
d
el,
tr
af
f
ic
co
n
d
itio
n
s
,
a
n
d
en
er
g
y
-
r
elate
d
c
o
n
f
i
g
u
r
atio
n
s
em
p
lo
y
ed
d
u
r
i
n
g
th
e
NS
-
3
s
im
u
latio
n
s
.
T
ab
le
1
.
Simu
latio
n
p
ar
am
eter
s
u
s
ed
in
NS
-
3
ex
p
e
r
im
en
ts
P
a
r
a
me
t
e
r
V
a
l
u
e
N
o
t
e
s
S
i
mu
l
a
t
o
r
NS
-
3
(
l
a
t
e
s
t
s
t
a
b
l
e
/
d
e
v
)
B
u
i
l
d
w
i
t
h
.
/
w
a
f
c
o
n
f
i
g
u
r
e
--
e
n
a
b
l
e
-
e
x
a
m
p
l
e
s
--
e
n
a
b
l
e
-
t
e
st
s.
A
r
e
a
(
m²
)
1
0
0
×
1
0
0
A
d
j
u
st
f
o
r
d
e
n
si
t
y
e
x
p
e
r
i
m
e
n
t
s
N
o
d
e
c
o
u
n
t
s
5
0
,
1
0
0
,
2
0
0
R
u
n
e
x
p
e
r
i
me
n
t
s
f
o
r
m
u
l
t
i
p
l
e
d
e
n
s
i
t
i
e
s
B
a
se
S
t
a
t
i
o
n
l
o
c
a
t
i
o
n
C
e
n
t
e
r
o
r
o
u
t
s
i
d
e
(
e
.
g
.
,
(
1
5
0
,
1
5
0
)
)
T
e
st
b
o
t
h
i
n
t
e
r
i
o
r
a
n
d
e
x
t
e
r
i
o
r
B
S
c
a
s
e
s
I
n
i
t
i
a
l
e
n
e
r
g
y
p
e
r
n
o
d
e
2
J (
t
y
p
i
c
a
l
)
T
u
n
e
p
e
r
e
x
p
e
r
i
m
e
n
t
R
a
d
i
o
m
o
d
e
l
8
0
2
.
1
1
(
o
r
s
i
mp
l
i
f
i
e
d
r
a
d
i
o
)
U
se
Wi
-
Fi
P
H
Y
/
M
A
C
o
r
a
s
i
mp
l
e
c
u
st
o
m
M
A
C
i
f
d
e
si
r
e
d
.
Tx
/
R
x
e
n
e
r
g
y
p
a
r
a
ms
E_
e
l
e
c
/
E
_
a
m
p
p
e
r
y
o
u
r
f
i
r
st
-
o
r
d
e
r
m
o
d
e
l
M
a
p
t
o
N
S
-
3
’
s
D
e
v
i
c
e
E
n
e
r
g
y
M
o
d
e
l
p
a
r
a
me
t
e
r
s.
P
a
c
k
e
t
si
z
e
6
4
B
(
sen
si
n
g
)
/
a
g
g
r
e
g
a
t
e
d
si
z
e
v
a
r
i
a
b
l
e
P
a
c
k
e
t
g
e
n
e
r
a
t
i
o
n
P
o
i
ss
o
n
/
p
e
r
i
o
d
i
c
(
e
.
g
.
,
e
v
e
r
y
5
s)
R
u
n
l
i
g
h
t
a
n
d
h
e
a
v
y
t
r
a
f
f
i
c
t
e
s
t
s
S
i
mu
l
a
t
i
o
n
t
i
me
3
6
0
0
s (
o
r
r
o
u
n
d
s
b
a
se
d
:
1
0
0
0
r
o
u
n
d
s
)
C
h
o
o
se
c
o
n
s
i
st
e
n
t
s
t
o
p
p
i
n
g
c
r
i
t
e
r
i
a
C
h
a
n
n
e
l
m
o
d
e
l
Lo
g
-
d
i
st
a
n
c
e
/
F
r
i
i
s
U
se
l
i
n
k
m
o
d
e
l
s a
v
a
i
l
a
b
l
e
i
n
NS
-
3
M
o
b
i
l
i
t
y
S
t
a
t
i
c
(
o
r
r
a
n
d
o
m
w
a
y
p
o
i
n
t
f
o
r
mo
b
i
l
e
t
e
st
)
I
f
t
e
st
i
n
g
sem
i
-
d
y
n
a
mi
c
n
e
t
w
o
r
k
s
N
u
mb
e
r
o
f
r
u
n
s
≥
1
0
s
e
e
d
s
A
v
e
r
a
g
e
r
e
su
l
t
s
w
i
t
h
9
5
%
C
I
6.
E
XP
E
R
I
M
E
N
T
A
L
DE
SI
G
N
T
h
e
ex
p
e
r
im
en
tal
ev
alu
atio
n
o
f
th
e
p
r
o
p
o
s
ed
AI
-
E
E
R
P
is
ca
r
r
ied
o
u
t
u
s
in
g
a
co
n
tr
o
lled
s
im
u
latio
n
en
v
ir
o
n
m
en
t
d
ev
el
o
p
ed
i
n
th
e
NS
-
3
n
etwo
r
k
s
im
u
lato
r
.
T
h
e
ex
p
er
im
en
tal
d
esig
n
is
s
tr
u
ctu
r
ed
to
en
s
u
r
e
a
f
air
,
r
ep
ea
tab
le,
an
d
s
tatis
tically
r
eliab
le
co
m
p
ar
is
o
n
b
etwe
en
AI
-
E
E
R
P
an
d
th
e
b
e
n
ch
m
ar
k
r
o
u
tin
g
p
r
o
to
co
ls
L
E
AC
H,
HE
E
D,
an
d
PEGA
SIS.
A
s
q
u
ar
e
s
en
s
in
g
with
m
o
n
ito
r
in
g
r
eg
i
o
n
o
f
1
0
0
m
×
1
0
0
m
i
s
co
n
s
id
er
ed
f
o
r
all
ex
p
er
im
e
n
ts
.
Sen
s
o
r
n
o
d
es
ar
e
r
an
d
o
m
ly
p
o
s
itio
n
e
d
with
in
th
is
r
eg
io
n
to
s
im
u
late
an
u
n
s
tr
u
ctu
r
ed
W
SN.
Net
wo
r
k
d
en
s
ity
ef
f
ec
ts
ar
e
an
aly
ze
d
b
y
c
o
n
f
ig
u
r
in
g
t
h
e
to
tal
n
u
m
b
e
r
o
f
s
en
s
o
r
n
o
d
es
to
5
0
,
1
0
0
,
an
d
2
0
0
in
d
if
f
e
r
en
t
s
ce
n
ar
io
s
.
T
h
e
b
ase
s
tatio
n
(
B
S)
is
lo
ca
ted
eith
er
at
th
e
ce
n
ter
o
f
th
e
s
en
s
i
n
g
f
ield
o
r
o
u
ts
id
e
th
e
n
etwo
r
k
b
o
u
n
d
ar
y
(
e.
g
.
,
at
co
o
r
d
in
ates (
1
5
0
,
1
5
0
)
)
to
e
v
al
u
ate
b
o
th
s
h
o
r
t
-
r
a
n
g
e
an
d
lo
n
g
-
r
an
g
e
c
o
m
m
u
n
icatio
n
s
ce
n
ar
i
o
s
.
E
ac
h
s
en
s
o
r
n
o
d
e
is
ass
u
m
ed
to
h
a
v
e
an
in
itial
en
er
g
y
ca
p
ac
ity
o
f
2
jo
u
les,
r
ep
r
esen
tin
g
a
ty
p
ical
b
atter
y
-
p
o
wer
ed
s
en
s
in
g
d
ev
i
ce
.
T
h
e
wir
eless
co
m
m
u
n
icatio
n
s
u
b
s
y
s
tem
is
m
o
d
eled
u
s
in
g
an
I
E
E
E
8
0
2
.
1
1
-
b
ased
r
ad
io
m
o
d
el
o
r
a
s
im
p
lifie
d
wir
eless
in
ter
f
ac
e
wh
en
co
m
p
u
tatio
n
al
ef
f
icien
cy
is
r
eq
u
ir
ed
.
T
h
e
tr
an
s
m
is
s
io
n
an
d
r
ec
ep
tio
n
u
tili
za
tio
n
o
f
en
er
g
y
is
ca
lcu
l
ated
u
s
in
g
th
e
s
tan
d
ar
d
an
d
b
asic
r
ad
io
en
er
g
y
m
o
d
el,
wh
er
e
p
ar
am
eter
s
E
_
elec
an
d
E
_
a
mp
ar
e
m
ap
p
ed
d
ir
ec
tly
to
th
e
NS
-
3
d
e
v
ice
en
er
g
y
m
o
d
el
f
o
r
r
ea
lis
tic
en
er
g
y
d
is
s
ip
atio
n
d
u
r
in
g
d
ata
tr
a
n
s
m
is
s
io
n
,
r
ec
ep
ti
o
n
,
id
le
lis
ten
in
g
,
a
n
d
a
g
g
r
e
g
a
tio
n
.
Data
p
ac
k
ets
o
f
s
ize
6
4
b
y
tes
ar
e
g
en
er
ated
p
e
r
io
d
ically
b
y
s
en
s
o
r
n
o
d
es,
wh
ile
ag
g
r
e
g
ated
p
ac
k
et
s
izes
at
clu
s
ter
h
ea
d
s
v
ar
y
d
e
p
en
d
in
g
o
n
th
e
n
u
m
b
e
r
o
f
m
em
b
er
n
o
d
es.
T
r
af
f
ic
g
en
er
ati
o
n
f
o
llo
ws
eith
er
a
Po
is
s
o
n
d
is
tr
ib
u
tio
n
o
r
a
f
ix
ed
p
er
io
d
ic
r
ate
(
e.
g
.
,
o
n
e
p
a
ck
e
t
ev
er
y
5
s
ec
o
n
d
s
)
to
s
u
p
p
o
r
t
b
o
th
lig
h
t
-
lo
a
d
an
d
h
ea
v
y
-
lo
a
d
n
etwo
r
k
co
n
d
itio
n
s
.
E
ac
h
s
im
u
latio
n
r
u
n
s
f
o
r
a
m
ax
im
u
m
d
u
r
atio
n
o
f
3
,
6
0
0
s
ec
o
n
d
s
o
r
eq
u
iv
alen
tly
1
,
0
0
0
o
p
er
atio
n
al
r
o
u
n
d
s
,
d
ep
e
n
d
in
g
o
n
t
h
e
ev
al
u
atio
n
s
ce
n
ar
io
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
5
4
3
-
1
5
6
1
1552
T
h
e
lo
g
-
d
is
tan
ce
p
ath
lo
s
s
m
o
d
el
is
u
s
ed
to
r
ep
r
esen
t
th
e
wir
eless
ch
an
n
el
co
n
d
itio
n
s
o
r
th
e
Friis
p
r
o
p
a
g
atio
n
m
o
d
el
to
ca
p
tu
r
e
r
ea
lis
tic
s
ig
n
al
atten
u
atio
n
ch
ar
ac
ter
is
tics
.
B
y
d
ef
au
lt,
al
l
s
en
s
o
r
n
o
d
es
a
r
e
ass
u
m
ed
to
b
e
s
tatic,
wh
ile
o
p
tio
n
al
ex
p
er
im
e
n
ts
em
p
lo
y
th
e
R
an
d
o
m
W
ay
p
o
in
t
m
o
b
ilit
y
m
o
d
el
to
ev
alu
ate
s
em
i
-
d
y
n
am
ic
n
etwo
r
k
b
e
h
av
i
o
r
.
T
o
en
s
u
r
e
s
tatis
tical
ac
cu
r
ac
y
an
d
r
e
p
ea
tab
ilit
y
,
ea
ch
ex
p
er
im
en
t
is
e
x
ec
u
ted
u
s
in
g
at
least
1
0
in
d
ep
en
d
en
t
r
an
d
o
m
s
ee
d
s
.
T
h
e
r
ep
o
r
ted
r
esu
lts
r
ep
r
esen
t
th
e
m
ea
n
v
alu
es
with
a
9
5
%
c
o
n
f
id
en
ce
in
ter
v
al.
Per
f
o
r
m
an
ce
cr
iter
ia
s
u
ch
as
n
etwo
r
k
life
tim
e,
p
ac
k
et
d
eliv
e
r
y
r
atio
,
th
r
o
u
g
h
p
u
t,
d
elay
,
r
o
u
tin
g
o
v
er
h
ea
d
,
an
d
en
er
g
y
f
ai
r
n
ess
ar
e
c
o
m
p
u
te
d
u
s
in
g
t
h
e
b
u
ilt
-
in
NS
-
3
f
lo
w
m
o
n
ito
r
an
d
e
n
er
g
y
f
r
am
ew
o
r
k
tr
ac
i
n
g
u
tili
ties
.
T
h
is
ex
p
er
im
en
tal
d
esig
n
e
n
s
u
r
es
a
r
ig
o
r
o
u
s
a
n
d
u
n
b
iased
ass
ess
m
en
t
o
f
AI
-
E
E
R
P
u
n
d
er
d
iv
er
s
e
n
etwo
r
k
co
n
d
itio
n
s
a
n
d
estab
lis
h
es
a
s
tr
o
n
g
f
o
u
n
d
atio
n
f
o
r
p
er
f
o
r
m
a
n
ce
co
m
p
ar
is
o
n
with
c
o
n
v
e
n
tio
n
al
W
SN
r
o
u
tin
g
p
r
o
to
co
ls
.
7.
E
VA
L
UA
T
I
O
N
P
ARA
M
E
T
E
RS
T
o
in
v
esti
g
ate
th
e
p
er
f
o
r
m
a
n
ce
ch
ar
ac
ter
is
tics
o
f
th
e
p
r
o
p
o
s
ed
AI
-
E
E
R
P
,
a
s
et
o
f
well
-
estab
lis
h
ed
ev
alu
atio
n
m
etr
ics
co
m
m
o
n
l
y
u
s
ed
in
W
SN
r
esear
ch
i
s
e
m
p
lo
y
ed
.
T
h
ese
m
etr
ics
co
lle
ctiv
ely
ca
p
tu
r
e
k
ey
asp
ec
ts
o
f
n
etwo
r
k
b
eh
av
io
r
,
i
n
clu
d
in
g
en
er
g
y
u
tili
za
tio
n
,
o
p
er
atio
n
al
life
s
p
an
,
co
m
m
u
n
ica
tio
n
r
eliab
ilit
y
,
an
d
r
o
u
tin
g
ef
f
icien
cy
.
E
ac
h
m
e
tr
ic
is
d
escr
ib
ed
b
elo
w
alo
n
g
with
its
s
ig
n
if
ican
ce
in
ev
alu
atin
g
r
o
u
tin
g
p
er
f
o
r
m
an
ce
.
7
.
1
.
Net
w
o
rk
life
t
ime
Netwo
r
k
life
tim
e
r
ep
r
esen
ts
th
e
p
er
io
d
f
o
r
wh
ic
h
a
W
SN
r
em
ain
s
f
u
n
ctio
n
al
a
n
d
ca
p
ab
le
o
f
p
er
f
o
r
m
in
g
s
en
s
in
g
an
d
d
ata
tr
an
s
m
is
s
io
n
task
s
.
Sin
ce
s
en
s
o
r
n
o
d
es
o
p
e
r
ate
o
n
lim
i
ted
b
atter
y
p
o
wer
,
n
etwo
r
k
life
tim
e
is
a
p
r
im
e
p
ar
am
eter
o
f
r
o
u
tin
g
p
er
f
o
r
m
a
n
ce
.
I
n
th
is
s
tu
d
y
,
n
etwo
r
k
life
tim
e
is
ev
alu
ated
u
s
in
g
th
r
ee
wid
ely
ad
o
p
ted
in
d
icato
r
s
:
a.
First
n
o
d
e
d
ea
th
(
FND)
:
T
h
e
s
im
u
latio
n
lo
o
p
at
wh
ich
th
e
in
itial sen
s
o
r
n
o
d
e
d
e
p
letes its
e
n
er
g
y
s
u
p
p
ly
.
b.
Half
n
o
d
e
d
ea
t
h
(
HND)
:
T
h
e
lo
o
p
at
wh
ich
a
p
p
r
o
x
im
at
ely
5
0
%
o
f
th
e
s
en
s
o
r
n
o
d
e
s
b
ec
o
m
e
n
o
n
-
o
p
er
atio
n
al.
c.
L
ast
n
o
d
e
d
ea
th
(
L
ND)
:
T
h
e
r
o
u
n
d
w
h
en
th
e
last
ac
tiv
e
n
o
d
e
in
th
e
n
etwo
r
k
r
u
n
s
o
u
t
o
f
en
er
g
y
.
T
o
g
eth
er
,
th
ese
m
ea
s
u
r
es
r
ef
l
ec
t
h
o
w
lo
n
g
t
h
e
n
etwo
r
k
r
e
m
ain
s
s
tab
le,
r
o
b
u
s
tn
ess
,
an
d
o
v
er
all
o
p
er
atio
n
al
lo
n
g
ev
ity
,
an
d
ar
e
c
o
m
m
o
n
ly
u
s
ed
to
co
m
p
ar
e
th
e
ef
f
ec
tiv
e
n
ess
o
f
en
er
g
y
-
awa
r
e
r
o
u
tin
g
p
r
o
to
co
ls
.
7
.
2
.
Co
un
t
o
f
a
c
t
iv
e
no
des
per
lo
o
p
T
h
is
p
ar
am
eter
tr
ac
k
s
h
o
w
m
an
y
s
en
s
o
r
n
o
d
es
r
em
ain
ac
tiv
e
d
u
r
in
g
ea
c
h
s
im
u
latio
n
r
o
u
n
d
.
I
t
g
iv
es
in
s
ig
h
t
in
to
lo
ad
b
alan
cin
g
,
r
o
u
tin
g
f
air
n
ess
,
an
d
clu
s
ter
-
h
ea
d
s
elec
tio
n
ef
f
icien
cy
.
A
s
l
o
w
d
ec
lin
e
in
aliv
e
n
o
d
es in
d
icate
s
b
etter
en
er
g
y
d
is
tr
ib
u
tio
n
ac
r
o
s
s
th
e
n
etwo
r
k
.
7
.
3
.
E
nerg
y
c
o
ns
um
ptio
n a
n
d r
esi
du
a
l e
nerg
y
R
esid
u
al
en
er
g
y
m
ea
s
u
r
es h
o
w
m
u
ch
en
er
g
y
r
em
ai
n
s
in
th
e
n
etwo
r
k
at
ea
c
h
r
o
u
n
d
:
(
)
=
∑
=
1
(
)
w
h
er
e
(
)
:
en
er
g
y
r
em
ai
n
in
g
in
n
o
d
e
i
.
T
r
ac
k
i
n
g
r
esid
u
al
e
n
e
r
g
y
h
el
p
s
d
eter
m
i
n
e:
e
f
f
ec
tiv
en
ess
o
f
ML
-
b
ased
p
r
ed
ictio
n
,
e
f
f
icien
cy
o
f
C
H
s
elec
t
io
n
,
an
d
r
ed
u
ctio
n
in
co
m
m
u
n
icatio
n
o
v
er
h
ea
d
.
L
o
wer
en
er
g
y
d
is
s
ip
atio
n
p
er
r
o
u
n
d
s
ig
n
if
ies th
at
AI
-
E
E
R
P
s
u
cc
ess
f
u
lly
co
n
s
er
v
es e
n
er
g
y
.
7
.
4
.
P
a
cket
deliv
er
y
ra
t
io
P
a
c
k
e
t
d
e
li
v
e
r
y
r
a
t
i
o
(
P
DR
)
m
e
a
s
u
r
e
s
t
h
e
d
e
p
e
n
d
a
b
i
l
it
y
o
f
s
en
d
i
n
g
d
a
t
a
p
a
c
k
e
t
s
f
r
o
m
n
o
d
e
s
t
o
t
h
e
B
S
:
=
A
h
ig
h
er
PDR
in
d
icate
s
:
s
ta
b
le
r
o
u
tin
g
,
im
p
r
o
v
e
d
lin
k
-
q
u
ality
p
r
ed
ictio
n
,
an
d
f
ewe
r
p
ac
k
et
d
r
o
p
s
d
u
r
in
g
CH
-
to
-
B
S c
o
m
m
u
n
icatio
n
.
AI
-
E
E
R
P
tar
g
ets im
p
r
o
v
ed
PDR
th
r
o
u
g
h
p
r
e
d
ictiv
e
lin
k
-
q
u
ality
esti
m
atio
n
.
7
.
5
.
T
hr
o
ug
hp
ut
T
h
r
o
u
g
h
p
u
t q
u
a
n
tifie
s
th
e
s
u
cc
ess
f
u
l d
ata
r
ate
r
ec
eiv
ed
at
t
h
e
B
S:
Evaluation Warning : The document was created with Spire.PDF for Python.