Intern
ati
o
n
a
l
Jo
u
r
n
a
l
of
P
o
we
r El
ec
tr
on
i
c
s
an
d D
r
i
v
e
S
y
stem
(I
JPE
D
S)
Vol
.
1
0
, N
o
. 4
,
De
ce
mbe
r
20
1
9
,
p
p
.
2
130~
21
37
I
SSN
:
208
8-8
6
9
4
, D
O
I:
10.
115
91
/i
jp
e
d
s.v
1
0
.i4
.
p
p21
30-
2137
2
1
30
Jo
urn
a
l
h
o
me
pa
ge
: ht
tp://
i
a
escore
.c
o
m
/
j
ou
rnals/
i
n
d
e
x.
p
hp/
IJ
PE
D
S
User selection protocols in F
D
PSP EH
cooperative network
over rayleigh fading channel:
outage and
interc
ept probabil
ity
P
h
u Tra
n
Tin
1
, Min
h
Tran
2
, Va
n
-
Duc Ph
a
n
3
,
Ho
an
g-Na
m Ng
uye
n
4
, Th
anh
-
L
o
ng
N
g
uy
en
5
1
Wire
les
s
Co
mmu
nica
tion
s Res
e
a
r
ch
Gro
u
p
,
F
acul
t
y
of
Electr
i
ca
l a
n
d
El
ectron
i
cs
En
gineer
in
g, To
n D
u
c
Than
g
Universi
ty, Ho
Chi
Mi
nh
Cit
y
,
Vie
t
na
m
2
Op
to
ele
c
tr
o
n
i
c
s R
e
s
ear
ch
G
r
oup
, F
acu
l
t
y
o
f
El
ec
t
r
ic
al
and
El
e
c
tr
o
n
ics
E
n
g
i
nee
r
i
n
g,
To
n
D
u
c
T
h
a
n
g U
n
i
v
er
si
ty
, H
o
C
h
i
Minh
City
,
Vi
etnam
3
Cente
r
of Exce
llenc
e
fo
r
Au
to
m
a
tio
n
and
Precisi
on
M
echan
ic
al E
n
g
i
n
eering
,
N
g
u
y
en Tat Th
a
n
h
U
n
ivers
i
ty
,
H
o
Ch
i
Minh
City
,
Vi
etnam
4
M
o
delin
g
Evo
l
utio
nary
Algo
rith
ms S
i
mula
tion
and
Artif
i
ci
al In
tellig
enc
e
,
F
acult
y
of E
l
e
c
tri
cal &
Electron
i
cs
E
n
gi
nee
r
i
n
g
,
Ton Duc
T
h
a
n
g Un
i
v
e
r
si
ty,
Ho Chi
Mi
nh
Cit
y
,
Vie
t
na
m
5
Cent
er
for Infor
m
at
ion
Technology,
Ho
Ch
i Mi
nh Cit
y
Uni
v
e
r
si
ty
of Fo
od
Ind
u
stry
,
Ho
Chi
M
i
n
h
Ci
ty, Vie
t
na
m
A
r
ticle In
fo
A
B
S
T
RAC
T
A
r
tic
le
h
i
st
o
r
y:
Rec
e
i
v
ed
D
e
c
1
2
, 2
018
Re
vise
d Ma
r 1, 2
0
1
9
A
c
ce
p
t
ed
M
a
y
2
8
,
2
019
In
th
is
p
a
per
,
we
in
ves
t
ig
ate
th
e s
y
s
t
em
p
e
rf
ormanc
e
in
t
e
r
m
o
f
o
u
t
ag
e
probabi
lit
y
(OP) and intercept
probabil
ity (IP) user
selection
prot
ocol
s
in
full
-duplex (FD) power
spli
tting prot
ocol (PS
P
) energy harvesti
ng (E
H)
coo
p
era
t
ive
n
e
t
w
o
r
k ov
er
the
Rayleig
h
f
a
ding
ch
annel
.
In
th
is
netw
ork
,
secu
rity
and
p
r
iv
acy
is
sues
are s
i
g
n
ific
ant
d
u
e
to
th
e p
o
s
s
ib
le
e
a
v
e
s
d
ro
pping
b
y
su
rro
un
di
ng
use
r
s.
In t
h
i
s
ca
se
, t
h
e sec
u
rity
pe
rform
a
n
ce a
n
d
re
lia
b
l
e
performance are
repr
esented by
out
a
ge
probabi
lity (OP) and i
n
t
e
r
cept
p
r
ob
a
b
i
lit
y (IP),
re
spe
c
t
i
v
e
l
y
.
The
p
o
w
e
r-split
ting e
n
erg
y
ha
rve
s
tin
g pro
t
oc
o
l
i
s
a
p
pl
ie
d i
n
ou
r a
n
a
l
y
s
i
s
.
We
rigo
ro
usly
de
rive
th
e
cl
ose
d
-form e
x
p
r
e
s
sio
n
s
of b
o
th
OP
and
IP
o
f
th
e
s
y
s
t
e
m
and
stu
d
y
th
e e
ffect
o
f
various
param
e
ters
.
F
i
n
a
lly, th
e M
o
n
t
e Carlo
sim
u
lat
i
o
n
res
u
lts
ar
e als
o
perfo
r
m
ed
to
co
nfirm th
e
correc
t
ness
o
f
all th
eoretic
al
an
a
l
y
s
is d
e
rived
.
Ke
yw
ords:
Co
op
er
a
tiv
e n
e
tw
or
k
Energ
y
harve
s
t
i
ng
(E
H)
F
u
ll dup
le
x
(
F
D
)
P
o
w
e
r
s
p
litt
in
g
pro
t
o
c
o
l
(
P
SP)
U
s
e
r
se
le
ct
io
n
Co
pyrigh
t ©
2
0
1
9
In
stitute
o
f
Ad
vanced
En
gin
eer
ing
an
d
S
c
ien
c
e
.
A
l
l
ri
g
h
t
s
re
se
r
v
e
d
.
Corres
p
o
n
din
g
A
u
t
h
or:
Van
-
D
u
c
P
h
an
C
e
n
t
er
o
f
Ex
c
e
ll
en
c
e
fo
r Au
tom
a
t
i
o
n
and
Pr
eci
sio
n
M
ech
a
n
i
c
al
Eng
i
n
eeri
n
g
,
N
guy
e
n
Ta
t
Tha
n
h Un
iv
er
sity
,
Ho
Ch
i
M
i
nh
Cit
y
,
Vi
et
n
a
m
Emai
l:
pvd
uc
@ntt
.e
du
.v
n
1.
IN
TR
O
DUCTION
R
a
di
o fre
q
ue
nc
y (RF
)
e
n
erg
y
tra
n
s
f
er
a
n
d
harve
s
ti
n
g
ha
s
re
ce
nt
l
y
e
m
e
r
ged
as
a
prom
i
s
ing
wa
y t
o
ext
e
n
d
t
h
e li
fe
t
i
m
e of e
n
e
r
g
y
-con
st
ra
ine
d
wi
re
l
e
ss
ne
t
w
ork
s
, espec
i
a
l
l
y
w
h
en c
o
nve
nt
i
o
nal
ene
r
gy
har
v
est
i
n
g
te
c
h
n
i
qu
e
s
f
r
om r
e
n
e
w
a
b
l
e e
n
ergy
sour
ce
s
a
r
e no
t ap
p
lic
a
b
l
e
.
Th
i
s
t
e
c
hniq
u
e
e
n
ab
le
s
wi
re
le
ss te
rm
in
a
l
s to
scave
nge e
n
e
r
gy from
R
F
s
i
gnal
s
br
oadca
s
t by am
bie
n
t
/
d
e
d
ic
at
e
d
wi
re
l
e
ss t
r
ansmi
t
t
e
rs
t
o
s
upp
ort
t
h
ei
r
o
p
e
r
a
ti
on
and in
fo
rmat
io
n tr
an
smi
ssi
o
n
.
Th
i
s
n
e
w co
mmun
i
c
a
t
i
o
n for
m
at h
a
s
b
e
en
term
ed
w
i
re
le
ss-po
w
e
r
e
d
comm
uni
ca
t
i
o
n
(WP
C
)
i
n
t
h
e
l
i
t
e
rat
u
re
,
whi
c
h
a
d
voc
at
es the
d
u
a
l
fu
nct
i
o
n
o
f
R
F
si
gnal
s
fo
r bot
h
inf
o
r
m
a
t
ion
del
i
v
ery a
n
d
e
n
e
r
g
y
t
r
a
n
sfer.
I
n
WP
C,
wi
rel
e
ss te
rmin
a
l
s can
av
oi
d
be
i
n
g
i
n
t
e
rru
p
t
e
d
b
y
t
h
ei
r ba
t
t
e
r
i
e
s
’
depl
et
i
o
n,
whi
c
h
ca
n
th
us
be
depl
oye
d
m
o
re
fle
x
ibl
y
a
n
d
mai
n
t
a
i
n
e
d
at
a l
o
wer
c
o
st.
In t
h
is
sens
e,
WP
C
ha
s
gre
a
t
e
r p
o
t
e
nt
i
a
l t
o
sust
ai
n the
ne
tw
ork op
e
r
at
i
on t
h
a
n
i
t
s c
onve
nti
o
nal
bat
t
e
ry
-po
w
e
r
ed c
o
u
n
t
e
r
p
art
in t
h
e
l
ong
ru
n. T
h
a
nks
t
o
the
s
e i
nhe
re
n
t
me
rit
s
,
WPC
has
be
en
re
gar
d
ed
as
an
i
n
di
spe
n
sa
bl
e a
n
d
i
rrepl
a
c
e
a
bl
e
bui
ldi
n
g
bl
ock
i
n
a
wide
ra
n
g
e
of a
p
pl
i
cat
i
ons
, e.
g., R
F
ID
,
wi
rel
e
ss s
e
nso
r
net
w
orks
, machi
n
e-t
o
-ma
c
hi
ne
comm
uni
ca
t
i
o
n
s, l
o
w
-
p
o
w
er
wi
de-a
re
a
ne
tw
ork
s
, a
nd t
h
e
Int
e
rne
t
o
f
Th
i
ngs
,
an
d so
o
n
[1-5]
.
The
r
e
are so
man
y
pa
pe
rs
focu
s on
t
h
e
WPC
di
rec
t
i
o
n
.
S
u
ch
a
s
so
m
e
p
a
pers prese
n
t
e
d
t
h
e p
r
oce
ss
e
n
e
r
gy
h
a
r
v
est
i
ng
th
ro
ugh
t
h
e RF si
gn
al
s i
n
c
oop
er
a
tiv
e
w
i
r
e
less
n
e
two
r
k
s
b
y
a MIM
O
r
e
lay
sy
st
em th
e d
i
ff
er
en
ce b
e
t
w
e
e
n
t
h
e
ene
r
g
y
tra
n
sfe
r
an
d t
h
e i
n
fo
rmat
i
o
n
rat
e
s t
o
p
r
ovi
de
the
opti
m
al
s
o
u
r
ce
and
rel
a
y p
r
e
c
odi
ng
[6-8]
.
In t
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J Po
w El
ec
&
Dr
i
S
y
st
IS
SN:
208
8-8
6
9
4
User
se
le
ct
i
on
prot
oc
ols in
F
D
P
S
P
EH
c
o
o
p
er
ati
v
e
ne
tw
o
r
k over
ray
l
e
i
gh f
a
din
g
… (V
a
n
-Du
c
P
h
a
n
)
2
131
li
te
ra
tu
r
e
of
t
h
e
o
t
h
e
rs, th
e a
u
tho
r
s inv
e
sti
g
ate
d
mu
l
ti-
u
s
er an
d mul
ti-
h
o
p
sy
st
ems for
simu
lta
n
e
ou
s
in
fo
r
m
a
tio
n a
n
d
p
o
w
e
r
tr
a
n
sfe
r
w
i
th
a du
a
l
-
hop
ch
ann
e
l w
ith
a
n
e
n
e
r
gy
h
a
rv
est
i
ng
re
la
y
;
t
h
e
t
r
an
smissi
o
n
strat
e
gy de
pe
n
d
s
o
n
the
q
u
al
it
y
of t
h
e
se
co
n
d
l
i
n
k
[9
-12
]
. Ph
ysi
cal
l
a
ye
r sec
u
ri
ty ha
s re
ce
nt
ly
a
t
t
r
a
c
t
e
d
a
l
o
t
of at
t
e
nt
i
on t
o
wi
rel
e
ss c
o
m
m
u
n
ic
at
ion
c
o
mmu
nit
y
,
es
pe
ci
al
l
y
w
h
e
n
t
h
e t
r
a
d
i
t
i
ona
l c
r
ypt
og
ra
ph
y
me
tho
d
s
se
e
m
s to
b
e
vu
ln
er
ab
l
e
to ad
v
a
n
ced
a
t
t
acks.
Wy
n
e
r
in
trod
u
ced
th
e id
e
a
of phy
sic
a
l
lay
e
r se
c
u
r
i
t
y
in
197
5
[1
3], b
u
t
t
h
i
s
t
opi
c
ha
s
onl
y b
e
e
n
reco
nsi
d
ere
d
re
ce
ntl
y
[
1
4
,
15
]. The
a
u
t
hors i
n
[1
4]
p
r
o
pos
ed
an
al
g
o
ri
thm t
o
maxi
mi
ze
t
h
e
sec
r
ec
y
su
m
ra
t
e
o
f
a
tw
o
-
wa
y rel
a
y s
y
st
em
by
o
p
ti
mi
zi
ng
the
re
la
y
bea
m
formi
n
g
ve
ct
or a
n
d
t
h
e tra
n
sce
i
ve
rs
p
o
w
e
r
s
u
n
de
r tot
a
l
po
wer
c
o
n
s
t
r
a
i
nt
.
A
p
h
y
s
ic
al
la
yer-sec
uri
t
y
sc
heme
for an
un
de
rl
a
y
re
l
a
y-
base
d c
o
g
n
it
i
v
e ra
di
o
net
w
o
r
k (C
RN
) t
h
at
use
s
ort
h
o
g
o
n
a
l
fre
q
u
e
n
cy
-d
i
v
i
s
ion
mul
t
i
p
l
e
xi
n
g
(OF
D
M
)
as t
h
e
medi
u
m
a
cce
s
s
te
ch
ni
q
u
e i
s
p
r
o
p
o
sed
i
n
[1
5], i
n
w
h
ic
h t
h
e
sec
r
ec
y
rat
e
is
maxi
miz
e
d s
u
bj
ect
t
o
t
h
e
con
s
t
r
ai
nt
s o
n
t
r
a
n
s
m
i
ssio
n
po
wer an
d i
n
t
e
rfe
r
ence
l
e
ve
l at
t
h
e
pri
m
a
r
y
u
s
e
r
s.
In
t
h
i
s
pa
pe
r,
we
i
nvest
ig
a
t
e
t
h
e
s
y
st
em
pe
rfo
r
ma
nce
(i
n t
e
rm
o
f
o
u
t
a
g
e
pro
b
abi
l
i
t
y
(OP
)
a
n
d
in
t
e
r
c
ep
t
p
r
o
b
a
b
i
lity
(
I
P))
u
s
er s
e
l
e
c
tio
n
p
r
o
t
o
c
o
l
s i
n
fu
ll
-du
p
l
e
x
(F
D)
p
o
w
e
r
sp
lit
tin
g
p
r
o
t
o
c
o
l
(P
SP)
en
e
r
gy
h
a
rv
est
i
ng
(
E
H)
co
op
e
r
a
t
iv
e
ne
tw
or
k ov
er
t
h
e
Ra
y
l
e
i
gh
fadin
g
ch
ann
e
l.
In th
i
s
n
e
twor
k,
s
e
c
u
r
i
ty an
d p
r
i
v
ac
y
is
su
es
are
sign
if
ic
a
n
t
du
e
to
t
h
e
po
ss
ib
le e
a
v
e
s
d
r
opp
i
n
g
b
y
sur
r
o
undin
g
u
s
e
r
s. I
n
t
h
i
s
c
a
s
e
,
th
e se
c
u
rit
y
perfo
rma
n
ce
a
nd re
li
abl
e
per
f
o
r
ma
nce
a
r
e represe
n
t
e
d
b
y
out
a
g
e
proba
bi
l
i
t
y
(OP
)
a
nd i
n
te
rce
p
t p
r
o
b
a
b
i
l
i
t
y
(IP
),
re
s
p
ect
i
v
el
y.
The
p
o
w
e
r-s
pl
it
t
i
ng e
n
e
r
gy
ha
rvest
i
n
g
prot
oc
ol
i
s
ap
p
l
i
e
d i
n
o
u
r
a
n
a
l
ysi
s
.
We
ri
g
o
r
ousl
y
deri
ve
the
cl
o
s
ed-form
e
x
pressi
ons
o
f
b
o
t
h
O
P
a
n
d IP
of the
s
y
ste
m
a
n
d stu
dy
t
h
e e
ffe
ct
of
vari
ous
p
a
rame
te
r
s
. F
i
n
a
lly
,
M
o
n
t
e
C
a
rl
o
si
mu
la
tio
n
r
e
su
lts
a
r
e
a
l
so
p
e
rfo
rmed
t
o
conf
ir
m t
h
e
c
o
rr
e
c
tn
ess
of al
l
t
h
eo
re
ti
ca
l ana
l
ysi
s
de
ri
v
e
d
.
T
h
e ma
i
n
c
ont
ri
buti
o
ns
o
f
t
h
e
pa
pe
r a
r
e
s
u
m
m
a
r
i
z
e
d
as
fol
l
ows:
1
)
Th
e
u
s
er selectio
n
pro
t
o
c
o
l
s in
fu
ll-du
p
l
e
x
(FD)
pow
er sp
littin
g
pro
t
o
c
o
l
(
PSP)
en
er
gy
h
a
rvestin
g
(EH
)
c
o
ope
rat
i
v
e
ne
t
w
ork
o
v
e
r R
a
yl
ei
g
h
fa
di
n
g
c
h
a
n
nel
i
s
p
r
o
p
o
sed
an
d
i
nve
st
i
g
at
ed.
2) The cl
ose
d
-fo
rm
e
x
pressi
o
n
s
o
f
OP
a
n
d IP
a
r
e de
ri
ve
d.
F
u
rt
he
rm
ore
,
t
h
e
i
n
fl
uen
c
e
of t
h
e
ma
i
n
para
met
e
rs
on
t
h
e
s
y
st
em pe
rformanc
e
is de
mo
n
s
t
r
a
t
e
d
e
n
ti
re
ly
by
t
h
e
Mont
e Ca
rl
o si
mul
a
ti
on.
The
st
ruct
ure
of
t
h
i
s
pa
per
is
p
r
o
p
o
se
d
a
s
foll
o
w
s. Sec
t
i
ons
II present
s
t
h
e syste
m
mo
del
of
t
h
e
r
e
l
a
y
i
ng n
e
tw
ork.
Se
c
tio
ns
III
d
e
r
i
v
e
s th
e sy
st
em p
e
r
f
orman
c
e of
t
h
e
mo
d
e
l sy
ste
m
. S
e
c
tio
n IV pr
ov
id
es
nu
me
ri
ca
l
resu
l
t
s
and
some
di
sc
ussio
n
s
.
Fi
na
l
l
y
, Se
ct
io
n
V c
oncl
ude
s
t
h
e
pape
r.
2.
SY
ST
E
M
MO
DE
L
In
th
is p
a
p
e
r
,
th
e sy
st
em
mod
e
l
is t
h
e
fu
l
l
-d
up
le
x
(
F
D
)
p
o
w
e
r sp
litti
n
g
p
r
o
t
o
c
o
l
(
P
SP)
en
erg
y
harve
s
ti
n
g
(E
H
)
co
ope
rat
i
v
e
n
e
t
w
o
r
k
ove
r R
a
yle
i
g
h
fa
din
g
cha
nne
l
(F
i
g
ure 1). I
n
t
h
i
s
m
odel
,
t
h
e in
for
m
a
t
i
on
is
tr
an
s
f
er
re
d
fr
om
th
e
sou
r
ce (
S
) to
th
e
m
u
l
t
i
-
d
e
sti
n
at
ion
(D
i
), t
h
ro
ug
h
e
n
erg
y
c
onst
r
a
i
ne
d
int
e
r
m
e
d
ia
t
e
rel
a
y
(
R
)
.
In
th
is
n
e
tw
or
k, t
h
e
r
e
is
an
eav
e
s
d
r
opp
er (
E
)
w
h
o
tries
to
l
i
ste
n
i
lle
g
a
l
l
y
t
o
th
e
so
u
r
c
e
d
a
t
a
.
We
al
so
assume
t
h
at
al
l
o
f
t
h
e
n
o
de
s
a
r
e
e
qui
pp
ed
w
i
t
h
a
si
n
g
l
e
a
n
te
nna
,
a
n
d t
h
e
da
ta
rel
a
yin
g
i
s
pe
rfo
rme
d
vi
a t
w
o
o
r
t
h
ogo
n
a
l
t
i
me
slo
t
s. Th
e ene
r
gy
h
a
r
v
e
s
ting
and
i
n
fo
rmat
io
n pr
oce
ssi
ng of th
e sy
st
em
mod
e
l
w
ith
FD
P
SP
are
pro
pose
d
i
n
Fi
g
u
re
2
.
In
t
h
i
s
sc
he
me
,
T is t
h
e
bloc
k
t
i
m
e
in
w
h
i
c
h t
h
e so
urce
ful
l
y
t
r
a
n
s
m
i
t
s
t
h
e
in
fo
r
m
a
tio
n
d
a
ta
to
th
e mu
lt
i-
d
e
st
in
a
t
ion
.
I
n
t
h
is
i
n
te
rva
l
tim
e
,
R h
a
r
v
e
s
ts
ρ
P
s
ener
g
y
fro
m t
h
e s
o
u
r
ce a
nd
use
(1
-
ρ
) P
s
ene
r
g
y
t
o
t
r
ans
f
er
i
n
fo
rmat
io
n
fro
m
S
t
o
t
h
e
mu
lt
i
-
de
st
i
n
at
i
on
node
s
,
w
h
i
c
h
ρ
is th
e
p
o
w
e
r sp
l
itti
n
g
fa
ct
or
ρ
∈
(0,
1). Al
l t
h
e
fa
di
n
g
cha
n
ne
ls
from S t
o
R
a
nd R
t
o
D
a
r
e
pro
pose
d
as t
h
e R
a
yl
ei
g
h
fa
d
i
n
g
cha
nne
l
s
.
En
er
g
y
H
a
r
v
est
i
n
g
(EH
)
I
n
for
m
a
t
io
n tr
a
n
s
m
is
s
i
on
(I
T
)
SR
h
S
R
D
1
D
n
D
k
n
R
D
h
R
R
h
E
SE
h
R
E
h
EH
at
R
(
ρ
P
S
)
IT
S
R
D
(1-
ρ
)P
S
T
F
i
gu
re 1.
S
y
ste
m
m
odel
F
i
gu
re 2.
The
po
we
r
s
p
l
i
t
t
i
n
g
p
r
ot
ocol
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
: 2
088
-8
6
94
In
t
J
P
o
w
Ele
c
& D
r
i
S
y
st Vol.
10
,
No
.
4
,
De
c
20
19
: 2130
–
2
137
2
132
3.
THE
SYST
EM PE
RFO
R
MANCE
Du
ri
ng
th
e
first t
r
a
n
smi
ssi
o
n
p
h
a
se
,
t
h
e
r
eceiv
e
d
s
i
gn
a
l
a
t
th
e
r
e
lay
c
a
n
b
e
g
i
v
e
n
by
:
(1
)
rs
r
s
r
r
r
r
yh
x
h
x
n
(1)
whe
r
e
h
sr
i
s
t
h
e
so
urc
e
to
rel
a
y cha
nnel
gai
n
,
h
rr
i
s
th
e
loopb
a
c
k
in
terf
er
enc
e
ch
an
n
e
l
,
s
x
is
th
e tr
ansm
i
t
t
e
d
sign
a
l
at
th
e sou
r
ce,
n
r
is
th
e
a
d
d
i
tiv
e
wh
i
t
e
G
a
ussia
n
no
ise
(
A
WGN)
wi
th
v
a
r
i
an
c
e
N
0
,
01
is PS
r
a
tio at
t
h
e
re
la
y
,
r
x
is the
t
r
an
smi
t
t
e
d sig
n
a
l
at
t
h
e
re
la
y
,
2
s
s
x
P
,
2
,
rr
x
P
: expec
t
at
io
n
ope
ra
to
r,
P
s
is a
v
era
g
e t
r
ans
m
it
p
o
we
r
at
t
h
e s
o
u
r
ce
,
P
r
is a
v
e
r
a
g
e
tra
n
sm
it po
w
e
r
at
th
e
re
lay
.
The
ha
rvest
e
d po
we
r at
t
h
e
re
l
a
y ca
n be
o
b
ta
ine
d
a
s
2
2
ss
r
h
rs
s
r
Ph
T
E
PP
h
TT
(2)
whe
r
e
01
is en
er
gy
co
nv
er
sion
e
f
f
i
c
i
e
n
c
y
.
The
re
ce
ive
d
si
gnal
at
t
h
e
n
rd
dest
ina
t
io
n ca
n
be e
x
pre
sse
d a
s
the
fol
l
o
wi
ng
ii
i
dr
d
r
d
yh
x
n
(3)
whe
r
e,
h
rdi
is th
e
re
lay
to
t
h
e
i
rd
dest
i
n
at
ion
c
h
a
nnel
ga
in
,
n
di
i
s
t
h
e a
ddit
i
ve w
h
i
t
e
G
a
u
ssia
n
noise
(A
WG
N)
wi
th varia
n
ce N
0
.
In
thi
s
m
o
del
,
w
e
c
onsi
d
e
r
t
h
e
dec
o
de
-a
n
d
-forwa
r
d
p
r
ot
o
c
ol
.
F
r
om
(1
), t
h
e
si
gnal
t
o
noi
se
ra
t
i
o
(S
NR
) a
t
t
h
e
re
l
a
y c
a
n
be
gi
ve
n
by
2
1
2
0
(1
)
s
rs
rr
r
hP
hP
N
(4)
S
u
bsti
t
u
ti
ng
(2
)
i
n
t
o
(
4
)
a
nd u
s
ing
t
h
e
fact
th
a
t
0
s
NP
, in
(4
)
can
be rew
r
itte
n
as
1
2
1
rr
h
(5)
Fr
om (2
), t
h
e SN
R
a
t
th
e d
e
stin
ati
o
n c
a
n
be
ca
lc
u
l
a
t
ed
a
s
22
2
2
2
20
00
ii
i
rr
d
s
s
r
r
d
sr
r
d
Ph
P
h
h
hh
NN
(6)
whe
r
e w
e
de
no
te
0
0
s
P
N
.
The
fi
na
l DF S
N
R
c
a
n
be
gi
v
e
n
as
12
mi
n
,
DF
(7)
3.1. Use
r
sel
e
cti
o
n
m
e
th
o
d
.
We
p
r
o
p
o
se
t
h
e
o
p
t
i
mal
u
s
e
r
sel
ect
i
o
n p
r
otoc
o
l
i
n
whi
c
h t
h
e
be
st
se
l
ect
i
on use
r
i
s
se
l
ect
ed
as
f
o
llo
ws
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J Po
w El
ec
&
Dr
i
S
y
st
IS
SN:
208
8-8
6
9
4
User
se
le
ct
i
on
prot
oc
ols in
F
D
P
S
P
EH
c
o
o
p
er
ati
v
e
ne
tw
o
r
k over
ray
l
e
i
gh f
a
din
g
… (V
a
n
-Du
c
P
h
a
n
)
2
133
2
2
1
,
2
,
..
.,
ma
x
(
)
i
rd
rd
iK
hh
(8)
In
[16
]
,
t
h
e c
u
mul
a
t
i
ve
de
nsi
t
y
fu
nc
ti
o
n
(C
D
F
)
of
2
rd
h
ca
n be
gi
ven
b
y
2
/
0
()
(
1
)
rd
rd
K
py
pp
K
h
p
Fy
C
e
(9)
whe
r
e
!
!(
)
!
p
K
K
C
p
Kp
,
rd
i
s
the
mean
o
f
the
ra
nd
om va
ri
abl
e
(R
V)
2
rd
h
.
The
n
, t
h
e
c
o
r
r
e
s
po
n
d
i
n
g
pro
b
a
bi
l
i
t
y
de
nsit
y
fu
nct
i
o
n
(P
D
F
)
ca
n be
o
b
t
a
i
n
e
d
by
2
1
(1
)
/
1
0
1
()
1
rd
rd
K
p
py
p
K
h
p
rd
fy
C
K
e
(10)
3
.
2
.
O
u
ta
g
e
pro
b
a
b
ilit
y
(OP).
2
2
12
0
2
1
Pr
Pr
m
i
n
,
Pr
m
i
n
,
m
a
x
i
DF
sr
rd
rr
OP
z
z
h
h
z
h
(11)
By
d
e
no
t
e
d
a
s
2
22
2
12
3
1
,
2
,
..
.,
,,
m
a
x
(
)
i
rr
s
r
rd
rd
iK
hh
h
h
.
In (1
0
)
ca
n be
re
writ
t
e
n
as
02
3
1
1
Pr
m
i
n
,
OP
z
(12)
whe
r
e
2
21
R
z
i
s
t
h
res
h
ol
d
a
n
d R:
the
sou
r
ce
ra
t
e
.
In
(1
2)
c
a
n
be
re
writ
t
e
n as
02
3
1
1
1P
r
P
r
OP
z
z
(13)
we den
o
t
e
(1
)
11
1
1
Pr
P
r
1
1
rr
z
z
Pz
e
(14)
whe
r
e
rr
is the
m
e
a
n
of RV
2
rr
h
and
03
23
3
20
2
3
0
2
3
2
03
33
3
3
3
03
00
Pr
1
P
r
1
Pr
1
|
()
1
1
()
sr
z
z
Pz
z
z
Ff
d
e
f
d
(15)
S
u
bsti
t
u
ti
ng
(9
) i
n
t
o
(
15)
,
we
have
03
3
03
3
1
(1
)
/
21
3
0
0
1
(1
)
/
13
0
0
1
1
1
1
sr
rd
sr
r
d
z
K
p
p
p
K
p
rd
z
K
p
p
p
K
p
rd
Pe
C
K
e
d
CK
e
e
d
(16)
U
s
i
n
g Ta
ble
of
Int
e
gral
Eq
[3.32
4
,
1
] i
n
[
17],
the
(1
6) c
a
n
reform
ul
at
e
as
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
: 2
0
88
-8
6
94
In
t
J
P
o
w
Ele
c
& D
r
i
S
y
st Vol.
10
,
No
.
4
,
De
c
20
19
: 2130
–
2
137
2
134
1
21
1
0
00
(1
)
21
2
(1
)
K
p
p
K
p
r
r
rd
rr
rd
zz
p
PC
K
K
p
(17)
S
u
bsti
t
u
ti
ng
(1
4), (1
7) int
o
(1
3), w
e
have
1
(1
)
11
0
00
(1
)
12
1
1
2
(1
)
rr
z
K
p
p
K
p
r
r
rd
rr
r
d
zz
p
OP
e
C
K
K
p
(18)
whe
r
e
()
v
K
is the
m
odi
fie
d
B
e
ssel
fu
nct
i
o
n
of
t
h
e
seco
nd
ki
nd
an
d v
th
ord
e
r.
3
.
2
.
Int
e
r
c
e
pt p
r
ob
abi
lit
y (
I
P)
.
The
no
de
s
S
a
n
d R
c
a
n
ran
d
o
m
l
y
ge
ne
rat
e
c
o
de
b
o
o
k
s t
o
c
o
n
f
use E fro
m c
o
m
b
i
n
in
g t
h
e rec
e
i
v
ed
d
a
ta w
ith
M
R
C
(see r
a
ndo
m
ize-
and
-for
w
a
r
d
(
R
F
)
met
hod
[1
8,
19
]). The
r
efo
r
e
,
t
h
e SNR
ob
tain
ed at E
fo
r
dec
odi
n
g
t
h
e
s
o
u
r
ce
da
ta
c
a
n
be
fo
rm
ul
at
ed
by
ma
x
,
E
SE
RE
(19)
whe
r
e
22
22
2
00
00
,
ss
e
r
r
e
S
E
s
e
R
E
sr
re
Ph
P
h
hh
h
NN
F
r
o
m
(19
)
,
IP
ca
n be ca
lc
ula
t
e
d
pre
c
i
s
el
y
a
s
(see
[20]
)
22
2
00
22
2
00
Pr
1
P
r
m
a
x
,
1P
r
P
r
Es
e
s
r
r
e
se
sr
r
e
IP
z
h
h
h
z
hz
h
h
z
(20)
0
2
30
Pr
1
se
z
se
Ph
z
e
(21)
whe
r
e
s
e
is th
e
mea
n
o
f
RV
2
s
e
h
. T
h
en
we ha
ve
22
40
0
00
Pr
Pr
Pr
sr
re
X
zz
Ph
h
z
X
z
X
F
(22)
whe
r
e w
e
de
no
te
22
s
rr
e
X
hh
.
U
t
i
l
i
z
i
ng t
h
e
re
sul
t
i
n
[2
1
]
,
t
h
e
CD
F o
f
X ca
n
be sh
ow
n as
th
e
bel
o
w e
q
uat
i
on:
1
()
1
2
2
X
sr
re
s
r
r
e
xx
Fx
K
(23)
whe
r
e
()
v
K
is the
m
odi
fie
d
B
e
ssel
fu
nct
i
o
n
of
t
h
e
seco
nd
ki
nd
an
d v
th
ord
e
r.
Fr
om (2
3)
an
d
(
2
2
)
c
a
n
b
e
c
o
mpu
t
ed
as
41
00
12
2
sr
r
e
sr
re
zz
PK
(24)
whe
r
e
,
s
rr
e
is
th
e
m
e
an
o
f
RV
22
,
s
rr
e
hh
, res
p
ect
i
v
el
y
.
S
u
bsti
t
u
ti
ng
(2
1), (2
4) int
o
(2
0), t
h
e
IP
can
b
e
ref
o
rm
ul
at
e
d
as
0
1
00
11
1
2
2
se
z
sr
r
e
sr
r
e
zz
IP
e
K
(25)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J Po
w El
ec
&
Dr
i
S
y
st
IS
SN:
208
8-8
6
9
4
User
se
le
ct
i
on
prot
oc
ols in
F
D
P
S
P
EH
c
o
o
p
er
ati
v
e
ne
tw
o
r
k over
ray
l
e
i
gh f
a
din
g
… (V
a
n
-Du
c
P
h
a
n
)
2
135
4.
NU
ME
R
I
C
A
L
RE
S
U
LT
S
AN
D
D
I
S
C
US
SION
In
t
h
i
s
se
ct
io
n,
t
h
e
Mo
nt
e Ca
rl
o
si
m
u
l
a
ti
on
is use
d
for val
i
dat
i
ng t
h
e a
n
a
l
yt
i
cal
exp
r
e
ssi
on i
n
t
h
e
abo
v
e se
ct
i
on.
We
co
nsi
d
e
r
a
ne
tw
ork wi
t
h
one
so
urce
, o
n
e
re
la
y, a
n
d
m
u
l
t
i
-dest
i
na
ti
o
n
,
w
h
e
r
e
so
urc
e
-
rel
a
y
and
rel
a
y
-
dest
i
n
a
t
i
o
n
di
st
ance
s
are
bot
h
n
o
r
m
a
l
iz
ed
t
o
u
n
i
t
va
l
u
e
[
22-
25
].
In
Fi
g
u
r
e 3
a
n
d
4, OP
a
n
d
IP
versus
p
o
w
e
r sp
lit
tin
g
f
a
cto
r
ρ
of
th
e
c
o
n
s
id
er
ed
pro
t
o
c
o
l
s is pr
ese
n
ted
.
H
e
re
,
we se
t t
h
e m
a
in
p
a
r
a
mete
rs a
s
R=0
.
5
bps
, P
s
/
N
o
=
10
dB
, K=
1,
2,4.
W
e
al
so
se
e
from
Fi
g
u
re 3
tha
t
t
h
e OP
of
t
h
e
p
r
op
ose
d
s
y
ste
m
has a
hu
ge
dec
r
ea
se
in t
h
e i
n
te
rval
o
f
ρ
from
0
t
o
0.5
an
d
ha
s a
m
a
ssi
ve i
n
c
r
e
a
se
whi
l
e
ρ
v
a
r
i
e
s
f
r
o
m
0.
5 to
1.
Th
e
m
i
ni
m
u
m
val
u
e
of
the
OP c
a
n
be fo
un
d at
ρ
=
0
.
5
.
On
a
n
ot
he
r
wa
y,
IP
ha
s a
sli
g
ht
inc
r
e
a
s
e w
h
en
ρ
in
cr
ea
se
s
from
0 t
o
1.
In
the
re
s
u
lt
s, t
h
e
si
m
u
l
a
ti
on
a
n
d
t
h
eo
re
ti
c
a
l
re
s
u
lt
s
a
r
e i
n
go
o
d
a
g
ree
m
e
n
t.
Fi
g
u
r
e 3.
O
P
ve
rsus
ρ
.
Fi
g
u
r
e
4. IP
ve
rsus
ρ
.
Fig
u
r
e 5
a
n
d F
i
g
u
re
6
sh
ow
s
t
h
e ou
ta
g
e
prob
ab
i
lity
and
in
t
e
r
c
e
p
t
pro
b
a
b
i
lity
o
f
th
e
p
r
o
p
o
s
ed
sy
s
t
e
m
versus
t
h
e
ra
ti
o P
s
/N
0
. Th
e
da
ta
r
a
t
e
i
s
se
t
to
R =
0.
5
b
p
s,
K
=
2,
ρ
=0
.2
,
0
.
5
,
0.
85 fo
r t
h
e
O
P
c
a
s
e
a
nd R =
0
.
5
bps
,
ρ
=0
.5
η
=
0
.3,
0.6, 0.
85,
re
spec
ti
vel
y
. It
c
a
n
be
o
b
se
rve
d
tha
t
t
h
e si
m
u
l
a
t
i
on c
u
rve
a
nd t
h
e
a
n
al
yt
ic
al
c
u
rve
f
o
r
ea
c
h
c
a
s
e
o
v
e
r
l
a
p
tog
e
the
r
. Th
a
t
co
nfi
r
m
s
th
e
c
o
rr
e
c
tn
ess
o
f
ou
r anal
y
s
is.
It
’s als
o
ea
sil
y
seen
tha
t
wh
e
n
th
e
r
a
t
i
o
P
s
/No
in
cr
ease
s,
t
h
e
o
u
t
a
g
e
pr
ob
ab
ili
ty
d
ecr
ea
se
s b
u
t
th
e i
n
te
rcep
t p
r
ob
ab
ilit
y
i
n
cre
a
s
e
s
.
Mo
reo
v
e
r,
t
h
e
i
m
pa
ct
of K on t
h
e
outa
g
e
pro
b
a
b
il
i
t
y a
nd
η
on
in
ter
c
e
p
t pr
ob
ab
ili
ty
is sho
w
n
in
Fig
u
r
e
7
and
F
i
g
u
r
e
8
.
H
e
re,
th
e r
a
t
i
o
Ps/No
i
s
set
to
1
0
d
B
.
Ag
a
i
n
,
th
e a
n
aly
tic
al
so
lu
t
i
o
n
s
ar
e in
e
x
a
c
t agree
m
en
t
w
i
t
h
th
e
si
mu
la
tio
n
r
e
su
l
t
s.
Th
e
o
u
t
age
pro
b
a
b
ili
ty
t
e
n
s
to d
ecr
ea
se
as
K
in
cr
e
a
se.
I
n
co
n
t
rast,
t
h
e i
n
ter
c
e
p
t pr
oba
b
ili
ty
in
c
r
ea
se
s wi
th
η
in
c
r
ea
se
s.
Fi
g
ur
e 5
.
OP ve
rsu
s
P
s
/N
0
.
Fi
g
u
r
e 6.
IP
v
e
rsus
P
s
/N
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
: 2
088
-8
6
94
In
t
J
P
o
w
Ele
c
& D
r
i
S
y
st Vol.
10
,
No
.
4
,
De
c
20
19
: 2130
–
2
137
2
136
Fi
g
ur
e
7
.
OP
ve
r
s
u
s
K
Fi
g
u
r
e 8.
IP
ve
rsus
η
.
5.
CO
NCL
U
S
I
O
N
In
t
h
i
s
pa
pe
r,
we
i
nvest
ig
a
t
e
t
h
e
s
y
st
em
pe
rfo
r
ma
nce
(i
n t
e
rm
o
f
o
u
t
a
g
e
pro
b
abi
l
i
t
y
(OP
)
a
n
d
in
t
e
r
c
ep
t
p
r
o
b
a
b
i
lity
(
I
P))
u
s
er s
e
l
e
c
tio
n
pro
t
o
c
o
l
s i
n
fu
ll
-du
p
l
e
x
(F
D)
p
o
w
e
r
sp
lit
tin
g
p
r
o
t
o
c
o
l
(P
SP)
en
ergy
harve
s
ti
n
g
(E
H
)
c
o
ope
ra
t
i
ve
n
e
t
w
o
r
k
ove
r R
a
yl
ei
gh
fa
di
n
g
chan
ne
l.
I
n
t
h
i
s
net
w
ork,
t
h
e
se
curit
y
pe
rform
anc
e
and
rel
i
a
b
le
per
f
o
r
m
a
nce
a
r
e re
prese
n
te
d
by
o
u
ta
ge
p
r
o
b
a
b
il
i
t
y
(O
P
)
an
d i
n
t
e
rc
e
p
t
p
r
o
b
a
b
i
l
i
t
y
(IP
)
,
re
spe
c
t
i
v
el
y. W
e
ri
g
o
rou
s
l
y
deri
ve the
c
l
os
ed-
f
o
r
m
e
x
pre
s
si
ons o
f
bot
h O
P
an
d IP o
f
t
h
e
syst
em
an
d
st
u
d
y
t
h
e
effec
t
of
va
ri
o
u
s pa
ram
e
t
e
rs
.
F
i
na
ll
y
,
M
o
n
t
e
C
a
rl
o si
m
u
l
a
t
i
on res
u
l
t
s
a
r
e
a
l
so perfo
rm
ed
t
o
c
o
nfi
r
m
t
h
e
correc
t
ne
ss
o
f
a
l
l
the
o
ret
i
c
a
l
anal
ysi
s
de
rive
d.
RE
FERE
NC
E
S
[1]
Chen
, He,
Cha
o
Zh
ai,
Yon
g
h
u
i Li, an
d
Br
ank
a
Vu
ce
tic
, "
C
oop
erativ
e
S
t
rategies
for
Wire
les
s
-P
o
w
ered
Co
m
m
u
ni
ca
ti
ons: An
Ove
r
v
i
ew,
"
IEEE Wi
reless Communicat
ions,
vol. 25,
n
o
. 4
,
p
p
. 1
1
2
-
19,
2
0
18.
do
i:10
.11
0
9
/
mw
c.2
0
1
7
.1
70
02
45
.
[2]
Bi,
S.
, Ho
, C
.
K
., &
Zhan
g
,
R., "
W
ire
l
e
s
s
p
o
w
e
red co
mm
u
n
icatio
n: Op
po
rtun
ities
and
ch
all
e
ng
es,"
IEEE
Co
mmu
n
i
c
at
io
ns Ma
ga
zin
e
,
vo
l.
5
3
,
pp
. 1
17-1
2
5
,
2
015
.
[3]
Niy
a
to
, D., Kim,
D.
I
., Maso
,
M., & Han, Z., "W
irel
es
s Po
we
re
d Com
m
u
nic
a
t
io
n Ne
twork
s
: Re
se
a
r
c
h
Di
re
c
tions
a
n
d Te
c
h
n
o
l
ogi
ca
l App
r
oac
h
e
s
,"
IEEE
Wirel
e
s
s
Co
mm
un
icatio
ns
,
pp
.
2
-
1
1
,
20
17.
[4]
Yu,
H.,
Lee
,
H.
, &
Je
on
, H.
, "
W
ha
t
i
s
5G?
Em
erg
i
ng 5
G
Mo
bi
le
Se
rv
ic
e
s
a
nd
Net
w
o
r
k
Re
q
u
i
r
e
m
e
n
ts
.
Sust
ainabilit
y
,
" vol
.
9,
p
.
1848,
2017.
[5]
C.
R. Va
le
nta
a
nd G.
D.
Du
rgi
n
, "Ha
r
ve
stin
g wir
e
less
po
wer: S
u
r
v
ey of
ene
r
g
y
-h
arv
e
ster
co
nv
ers
i
o
n
eff
i
ci
ency
in
far-fi
eld, w
i
reles
s
power transfer
s
y
s
t
ems,"
IEE
E
M
i
c
r
ow.
Ma
g.
,
v
o
l
.
15
, n
o
.
4
,
p
p
.
108
–1
20
, 2
014
.
[6]
Sharma, V.,
&
Karmakar
,
P
., "A Nove
l
Method of Opport
unistic
W
i
r
e
l
e
ss
E
n
ergy Harvesti
ng
i
n
Cognitive R
a
dio
Ne
twor
ks,"
20
15
7t
h
In
te
rn
at
io
n
a
l
Con
f
e
r
e
n
ce
on
Compu
t
a
t
io
na
l In
tel
l
i
g
e
n
ce
,
Co
mmun
i
c
a
t
io
n S
y
st
e
m
s a
nd
Net
w
orks
,
20
15
.
do
i:10
.11
0
9
/
ci
csyn
.2
015
.2
1
[7]
A.
M. Fo
ul
a
dgar
a
nd O. Sim
e
o
n
e
,
"On
the T
r
an
sfe
r
o
f
In
form
ation
an
d
En
erg
y
in
Mu
lt
i-U
s
er S
y
stems,”
IE
EE
Communi
c
at
ions Letters
,
vo
l.
16,
no
. 1
1
,
pp.
17
33–1
7
3
6
,
2
0
1
2
.
[8]
Z
h
ang,
R
.
,
& H
o
, C.
K.,
"MIMO Broadcasti
ng
for Si
mul
t
aneous Wi
rel
e
s
s
Informat
ion
and Power T
r
ansf
er.
I
E
E
E
T
r
a
n
sa
ct
io
ns on
Wi
rele
ss Co
m
m
u
n
i
c
ati
o
n
s
, v
o
l
.
1
2
(
5
),
19
89
-20
01,
20
13.
do
i:
10
.
1
1
0
9
/
t
w
c
.
2
0
1
3
.0
318
13
.
1
2
0
2
24
[9]
P
a
rk, J
.
,
& C
l
e
r
ckx,
B., "
J
o
i
n
t
Wir
e
les
s
Info
r
m
ation
and
En
e
r
g
y
Tr
an
sfer
in
a
Two-U
s
er M
I
M
O
In
terfer
e
nc
e
Channel,"
I
E
E
E
Tran
sa
c
t
io
ns on
W
i
re
l
e
ss Commu
n
i
c
a
ti
on
s
, vo
l.
1
2
(
8), p
p
.
42
10
-4
22
1,
20
13
.
[10]
Ch
a
l
i
s
e
,
B.
K.
,
Zha
n
g
,
Y.
D., &
Am
i
n
,
M.
G.
,
"En
e
rgy
h
a
rve
s
t
i
n
g
i
n
a
n
OSTB
C
ba
se
d am
pl
if
y-a
nd-f
o
rwa
r
d MIMO
r
e
l
a
y sy
st
e
m
,
"
2
0
1
2
IE
EE
In
ter
n
a
t
iona
l Co
nfer
e
n
ce
on
Acou
st
ic
s
,
Sp
e
e
ch
an
d
S
i
gn
al
Pro
cessing
(ICASS
P
)
,
2
012
.
do
i:10
.11
0
9
/ic
as
sp
.20
1
2
.
62
88
59
6
[11]
Tu
tu
ncuo
glu
,
K.
,
& Y
e
ner
,
A
.
,
"
C
oo
perative
en
e
r
gy
h
a
rv
estin
g
co
mmun
i
c
a
ti
on
s
with r
e
lay
i
ng
an
d en
ergy
s
h
arin
g
,
"
2013 IE
E
E
In
f
o
rmat
i
on Theory
Wor
k
s
h
op (ITW)
,
20
13.
do
i:
10
.
1109
/
i
t
w
.
2
0
1
3
.
6
6912
80
[12]
Wy
ner, A
.
D
.
, "
T
h
e
wire-
t
ap
ch
an
n
e
l,"
T
h
e Bell S
y
stem T
echn
i
ca
l J
o
ur
na
l
, v
o
l
.
5
4
(8
),
pp
. 1
3
5
5–1
38
7,
197
5
.
[13]
W
a
ng, L., W
o
n
g
K.,
Ji
n
,
S.,
Z
h
eng
,
G., Jr.,
R
.
W.
H.,
"A
n
e
w
l
o
o
k
at ph
ysi
cal lay
e
r
s
ecurity,
c
a
c
hing
, and
wir
e
les
s
e
n
e
r
gy ha
rv
e
s
ting
fo
r
he
te
rog
e
n
e
ou
s ul
tra
-
de
nse
ne
twork
s
,
"
I
E
EE Co
mm
unic
a
tio
n
s
Ma
ga
zin
e
,
vol.
56
,
n
o
.
6,
pp
. 4
9
-5
5
,
2
0
1
8
.
https
:
/
/
d
o
i
.
org
/
1
0
.
1
109
/M
CO
M.
20
18.
17
004
39
[14]
Obeed,
M
.
, Mesbah,
W
.
,
"E
ffi
ci
ent al
gori
thms
for
ph
ysi
c
a
l
la
ye
r sec
u
rit
y
in
tw
o-w
a
y re
lay systems,"
Phys
i
c
a
l
Co
mmu
n
i
c
at
io
n,
vo
l.
28
,
p
p
.
7
8
–
8
8
,
201
8.
[15]
Sh
a
h
,
H.
A.
,
Koo
,
I
.
,
"
A
n
o
v
e
l ph
ysic
a
l
l
a
ye
r se
c
u
rit
y
sc
h
e
me
in
o
f
dm
-b
a
s
e
d
co
g
n
i
t
i
v
e
r
a
di
o ne
two
r
ks,
"
IE
E
E
Acc
e
s
s
,
v
o
l
.
6
,
p
p
.
29
48
6–
29
4
98,
2
0
1
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J Po
w El
ec
&
Dr
i
S
y
st
IS
SN:
208
8-8
6
9
4
User
se
le
ct
i
on
prot
oc
ols in
F
D
P
S
P
EH
c
o
o
p
er
ati
v
e
ne
tw
o
r
k over
ray
l
e
i
gh f
a
din
g
… (V
a
n
-Du
c
P
h
a
n
)
2
137
[16]
Tan
N
.
N
guy
en,
Tran
Trung
Du
y, G
i
a-Th
ien
Luu
,
P
h
u
o
n
g
T.
Tran
an
d
M
i
ro
slav
Vo
znak
,"
Energ
y
h
a
rv
estin
g
-b
as
ed
S
p
ectrum
A
c
c
e
s
s
With
In
cre
m
en
tal
Coo
p
er
ation
,
R
e
lay
S
e
l
ectio
n an
d
Hardw
a
r
e
No
ises
",
R
a
dio
E
ngine
ering,
Vo
l
2
6
,
No
. 1, pp
.
2
40-2
5
0
,
Apr. 2
017.
[17]
T
a
ble
of In
te
g
r
a
l
s,
Se
rie
s
,
a
n
d
Prod
uc
t
s
,
20
1
5
.
[18]
J.
M
o
, M
.
Tao
,
Y
.
Liu, "
R
e
l
ay
p
l
a
cem
ent
fo
r
p
h
y
s
ical
lay
e
r
s
ecurity: A se
cu
re co
nn
ect
io
n pe
rs
pectiv
e,"
IE
EE
Com
m
un.
Lett
, vo
l.
16 (
6
),
pp
. 8
78–
88
1,
20
1
2
.
[19]
T.T. D
u
y
,
P
.
N
.
S
on, "S
ecrecy
p
e
rforman
ces
of
mu
lticast u
nde
rla
y
c
o
gn
it
iv
e prot
oc
o
l
s wi
th
pa
rtia
l rel
a
y sel
e
c
t
i
o
n
a
n
d wit
h
o
u
t
e
a
v
e
s
d
r
op
pe
r’s in
form
a
t
i
on,
"
KSII
Trans. Internet Infor
m
. Sys
t
.
,
vo
l.9
,
n
o
.
1
1
,
pp
.
46
23
–4
643
, 2
0
1
5
.
[20]
S
.
Jia, J.
Zh
an
g
,
H
.
Zh
ao
, R. Zh
ang
,
"
R
e
l
ay s
e
lec
t
ion
fo
r
imp
r
o
v
ed
secur
i
ty
in
co
gn
it
ive
r
e
l
a
y
net
w
o
r
ks w
i
t
h
jamm
ing,
"
I
E
EE
Wirel
e
ss
Comm
un. Let
t
.
,
vo
l.
6
(
5),
pp
.
66
2–
66
5
,
2
0
1
7
.
[21]
Ng
uy
en, Tan
N., P
hu Tr
an
Tin
,
Ph
uo
ng
T. Tran
, Tran
Ho
ang
Qu
a
n
g
M
i
n
h
, and
M
i
ros
l
av
Vo
zn
ak
., "
P
ow
er-Sp
l
ittin
g
P
r
otocol
in
P
o
w
e
r
B
eaco
n-
ass
i
s
t
ed
Energy
H
a
rv
es
ting F
u
l
l
-D
up
lex R
e
lay
i
n
g
Ne
t
w
ork
s
:
P
e
rfo
rm
an
ce
A
n
alysis
,"
20
18
11
th
I
F
IP
Wirele
s
s
and
Mo
bile
Networ
king
Con
f
er
enc
e
(W
MNC)
,
20
18.
do
i:
10
.
2
3
9
1
9
/
wm
n
c
.
2
01
8.
8
4
8
0
9
0
0
.
[22]
Tran
Ho
ang
Qu
an
g M
i
nh
,
"H
yb
rid
Time
-
P
o
wer
S
w
itch
in
g P
r
o
t
o
c
o
l
of
En
ergy
H
a
rves
ting
Bid
i
r
e
ction
a
l R
e
lay
i
n
g
Ne
twor
k:
Th
ro
u
g
h
p
u
t
a
nd E
r
go
di
c Ca
p
a
c
i
t
y
An
a
l
y
s
i
s
."
T
E
LKO
M
NIKA
(
T
elecomm
uni
c
at
i
o
n, Computi
n
g
,
Ele
c
tr
onics and
Con
t
r
o
l),
vol
.
1
6
,
no.
5
,
p
.
18
9. h
t
t
p
:
/
/
dx.
do
i.
o
r
g/
1
0
.
1
2
9
2
8
/te
lk
om
ni
ka
.
v
1
6
i5
.
9
1
18.
[23]
Tin, P
h
u
Tran
, T
r
an Hoan
g Quan
g M
i
nh
,
Tan N
.
Ng
uyen
,
a
n
d
M
i
ros
l
av
V
o
znak
,
"
S
ys
tem P
e
rfo
rm
ance
A
n
alysis of
Half-Dup
lex
R
e
lay N
e
tw
ork o
v
er Rician F
a
d
i
ng
Chan
nel,
"
T
E
LKO
M
NIKA (T
elecomm
un
ica
t
io
n,
Co
mp
utin
g
,
Ele
c
tr
onics and
Con
t
r
o
l)
,
vol
.
1
6
,
no.
1
,
p
.
18
9, 201
8.
do
i:1
0.
12
9
2
8
/
t
e
lk
om
ni
k
a
.
v
1
6
i1
.
7
4
9
1
.
[24]
Rash
id
, Tariqu
e, S
u
nil K
u
mar
,
A
k
s
h
ay
Verma,
P
r
ateek R
a
j G
a
u
t
am
,
and
Arvind
Ku
mar, "
P
m-
EEM
R
P
:
P
o
s
t
u
r
al
Mo
ve
m
e
nt Ba
sed
Ene
r
g
y
Effic
i
e
n
t Mu
lt
i-
ho
p
Rou
t
in
g
Prot
o
c
o
l
fo
r In
tra
W
i
re
less
Bod
y
Se
nso
r
Ne
twork (
I
nt
ra-
WB
S
N
)
,
"
T
E
L
K
O
M
NIKA (T
elec
o
mmu
nication
,
Com
putin
g,
Elec
t
r
o
n
i
cs
a
n
d
Co
n
t
r
o
l),
vo
l
.
1
6
, n
o
.
1
,
p
.
16
6,
20
18.
do
i:10
.12
9
2
8
/
te
lk
omnika
.
v16
i1
.7
3
1
8
.
[25]
Tin, P
hu Tr
an,
M
i
n
h
Tran,
Tan
N
.
N
guy
en,
an
d Th
a
n
h-Lon
g
N
g
u
y
en,
"S
y
s
tem
P
e
rforman
ce A
n
alysis of
H
ybri
d
T
i
m
e
-p
o
w
e
r
Swit
c
h
i
ng
Prot
o
c
o
l
o
f
E
H
Bidi
re
ct
io
na
l
Rel
a
ying Ne
twork
in
A
m
p
l
ify
-
an
d
-fo
rward M
ode
,
"
In
do
ne
si
an
J
ourna
l o
f
E
l
e
c
t
ric
a
l E
n
g
i
ne
e
r
ing
an
d C
o
mpu
t
e
r
S
c
ie
nc
e,
vol.
1
4
, n
o
.
1, p.
1
1
8
,
201
9
.
do
i:10
.11
5
9
1
/
ije
ecs
.v1
4
.
i
1.
pp1
18
-12
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.