TELKOM
NIKA
, Vol. 13, No. 4, Dece
mb
er 201
5, pp. 1180
~1
186
ISSN: 1693-6
930,
accredited
A
by DIKTI, De
cree No: 58/DIK
T
I/Kep/2013
DOI
:
10.12928/TELKOMNIKA.v13i4.3105
1179
Re
cei
v
ed Se
ptem
ber 16, 2015; Revi
se
d No
vem
ber
6, 2015; Acce
pted No
vem
b
er 21, 201
5
Power Characteristics of Wireless Charging for Sensor
Networks in Tunnels
Xiaoming Liu
1,2,3,4
,
Enjie
Ding
1,2,3,4
*, Duan Zhao
1,2,3,4
, Youfang Yang
1,2,3,4
1
School of Infor
m
ation a
nd El
e
c
trical Eng
i
ne
e
r
ing, Ch
ina U
n
i
v
ersit
y
of Min
i
n
g
and T
e
chno
l
o
g
y
, Xuzho
u
,
Jian
gsu, Chi
n
a
2
IOT
Perceptio
n Mine R
e
sear
ch Center
,
C
h
i
na Un
iversit
y
o
f
Mining a
nd T
e
chn
o
lo
g
y
, Xu
zhou, Jia
ngs
u,
Chin
a.
3
State and Loc
al Joi
n
t Engi
ne
erin
g Lab
orator
y of Mini
ng Inte
rnet Appl
ic
atio
n T
e
chnolo
g
y
,
Xuz
h
o
u
, Jian
g
s
u,
Chin
a
4
Jiangs
u Engi
n
eeri
ng La
bor
ator
y
of Perc
epti
on Min
e
Rese
a
r
ch Center,
Xu
zhou, Jia
ngs
u, Chin
a.
e-mail: e
n
ji
ed
@cumt.edu.cn
A
b
st
r
a
ct
In order to ge
nerate w
i
rel
e
s
s
microw
ave p
o
w
e
r c
hargi
ng
technol
ogy in
coal mine tu
n
nels, it is
necess
a
ry to know
the pow
er
charac
teristics
of w
i
reless el
ectromag
netic
w
a
ve alo
ng th
e tunn
el w
a
lls.
In
this p
aper, t
he
exper
i
m
en
tal resu
lts of
narrow
b
and
w
i
reless
ele
c
troma
g
n
e
tic
w
a
ve pro
pag
ati
o
n
me
asur
e
m
ents
are
prese
n
te
d, an
d a st
ati
s
tical
mo
del
o
f
the pow
er c
har
acter
i
stics
at 2.4 GH
z
i
n
rea
l
rectang
ular
mi
ne tun
n
e
l
s is p
r
opos
ed. T
w
o standar
d h
a
lf
-
w
ave dip
o
l
e
an
tennas w
e
r
e
u
s
ed to p
e
rfor
m the
field ex
peri
m
en
ts in tunne
ls w
i
th different w
a
ll
mate
ri
als. A 1
0
-meter w
i
rele
ss chargi
ng d
i
s
t
ance be
lo
ngs
to
the free-sp
ace
prop
agati
on
z
o
ne. T
he
pat
h
lo
ss expo
ne
nts rely h
eavi
l
y o
n
t
he l
o
cati
on
of the rec
e
iv
er an
d
anten
na p
o
lar
i
z
a
ti
ons. T
o
obt
ain
mor
e
pow
e
r
, the locatio
n
s of the receiv
er shou
ld
match t
he corres
p
o
ndi
ng
pol
ari
z
a
t
i
ons.
Ke
y
w
ords
: W
i
reless C
harg
i
n
g
, Pow
e
r Char
acteristic, T
unn
el, Sensor N
e
tw
orks
Copy
right
©
2015 Un
ive
r
sita
s Ah
mad
Dah
l
an
. All rig
h
t
s r
ese
rved
.
1. Introduc
tion
The Wi
rele
ss Senso
r
Network
(WS
N) pl
ays a si
gnifi
cant role in mi
ne safety pro
ductio
n
.
Conve
n
tional
battery p
o
wer fo
r the
WSN
affect
s its sta
b
le
o
peratio
n time
. The resea
r
ch
prog
re
ss of
microwave p
o
we
r tr
an
sfer [1-4] all
o
ws
the battery to
be
cha
r
g
ed
usin
g mi
cro
w
ave
wirel
e
ss
ch
arging te
chn
o
lo
gy, thus p
r
ol
ongin
g
t
he lif
e cy
cle of th
e WS
N. In o
r
der to
dete
r
m
i
ne
wave tra
n
sfe
r
efficiency be
tween the
se
nding a
nd re
ceiving device
s
, the cha
r
a
c
t
e
risti
cs
of wa
ve
prop
agatio
n in mine tunnel
s mu
st be stu
d
ied.
Several theo
retical and p
r
actical studie
s
have bee
n perfo
rmed o
n
wave pro
p
a
gation in
tunnel
s [5
-8]
and
can
be
su
mma
rized
as an
alytica
l
and
nu
meri
cal
model
s.
Zhi Sun
et
al.
prop
osed the multimode
-waveguid
e
mo
del [8], which is
cap
able of accurately ch
ara
c
teri
zin
g
fast
fluctuation
s
o
f
the ch
ann
el
and
provide
s
a
n
an
al
ytical expressio
n
of t
he
re
ceived po
we
r at
any
locatio
n
in a tunnel. Rui
s
i
He [9] repo
rt
ed mea
s
u
r
e
m
ents taken i
n
a sub
w
ay tunnel at 2.4 GHz.
Acco
rdi
ng to
the pra
c
tical
application, this
autho
r d
e
co
rated the
antenna o
n
the tunnel side
wall
s, providi
ng u
s
a reference valu
e. Y.P. Zhang
et
al. [10] pla
c
e
d
the tran
smit
anten
na
at th
ree
locatio
n
s: mi
ddle, upp
er
middle, and
uppe
r-le
ft corne
r
of the
tunnel cro
s
s se
ction. T
he
con
c
lu
sio
n
s show
that
the attenuat
ion i
s
almost the
same for the t
h
ree
ca
se
s, but the tran
smit
antenn
a in
sertion po
wer lo
ss i
s
different
. Chah
é Nerg
uizia
n
[11] pl
ace
d
the tran
smitting a
n
te
nna
at different l
o
catio
n
s
duri
ng the n
a
rro
w
ba
nd
me
asurem
ent p
r
o
c
edure an
d a
nalyze
d
the
path
loss. Ca
rlo
s
Rizzo [1
2] prese
n
ted a
co
mplete
study
about t
r
an
sversal fadi
ng i
n
the Som
p
o
r
t
tunnel
at 2.4
GHz.
The
fa
ding
beh
avior wa
s simil
a
r i
n
ea
ch
tra
n
sversal
h
a
lf of
the tu
nnel
b
u
t
different in
th
e center. Stu
d
ies [13] dev
elope
d a fu
si
on mo
del
co
mbining
the
modal th
eory
and
ray theo
ry in
the re
ctang
ul
ar tunn
el. Th
ese
aut
ho
rs
modified the
ray mod
e
l by
the modal t
h
eory
and em
ulate
d
the wave
p
r
opa
gation i
n
a re
ctang
ula
r
tunnel. Sun
Ji-pi
ng et al.
[14] intensiv
ely
studie
d
the
cha
r
a
c
teri
stics of wave p
r
opa
gat
ion i
n
tunnel
s an
d
pre
s
ente
d
a mathem
atical
expre
ssi
on t
hat con
s
ide
r
s the
facto
r
s of cro
s
s-se
ctional dim
e
n
s
ion a
nd
con
ductivity, among
others.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
9
30
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 117
9 – 1186
1180
Ho
wever,
m
o
st of
the
s
e
studi
es focuse
d o
n
th
e p
r
op
agatio
n
cha
r
a
c
teri
stics of
electroma
gne
tic (EM
)
wav
e
s
alon
g the
t
unnel. In
a
c
tu
al un
dergrou
nd mi
ne tu
nn
els,
sen
s
o
r
s
are
usu
a
lly arran
ged on
both
the side
s a
n
d
the top of
the tunnel
walls (as lo
cati
ons A, B an
d C
sho
w
n i
n
Fig
u
re
1),
which
formed
zona
l distri
but
io
n
c
h
ar
ac
te
r
i
s
t
ics
o
f
s
e
ns
or
no
d
e
s
in
tu
nne
ls
.
The
po
wer transmitte
r i
s
p
l
ace
d
in
the
middle
of th
e
tunnel
(lo
c
ati
on O
in
Figu
re 1
)
; the
r
efore, we
need to d
e
termine the EM
prop
agatio
n chara
c
te
risti
c
s along the tu
nnel walls in t
h
e nea
r
regio
n
.
To date, the
EM propa
g
a
tion ch
aract
e
risti
cs
al
on
g
the recta
n
g
u
lar tun
nel
wall
s at a short
distan
ce h
a
ve been p
oorly
mentioned a
nd not studi
e
d
in detail.
The m
e
a
s
urement-ba
s
ed
ch
ara
c
te
riza
tion of
u
nde
rgroun
d min
e
s
provid
es reali
s
ti
c
results that
are
of imm
e
diate u
s
e
to
desi
gne
rs an
d devel
ope
r
s and
ha
s th
erefo
r
e
attra
c
ted
increa
sing
at
tention in th
e
last d
e
cade.
This
pa
per
descri
be
s th
e re
sult
s of
a mea
s
u
r
em
ent
camp
aign th
a
t
was p
e
rfo
r
med at freq
u
enci
e
s of
2.4
GHz i
n
a re
al re
ct
ang
ular small-scale
mine
tunnel.
2. Wireless
Chargin
g
Sy
stem for Se
n
s
or Ne
t
w
o
r
k
s
In the actu
al
coal
mine tu
nnel, sen
s
or
node
s a
r
e
arrang
ed o
n
tu
nnel
wall
s orderly (as
sho
w
n in Fi
g
u
re 1
-
a
)
, whi
c
h p
r
e
s
ent zo
nal dist
ributio
n. In orde
r to wirel
e
ssly ch
arge th
e se
n
s
or
node
s, a mi
crowave t
r
an
smitter device
(as sho
w
n in
Figure 1
-
b) should
be d
e
velope
d to
sup
p
ly
the RF (radio
freque
ncy
)
en
ergy. The wh
ole ch
argi
ng
scheme i
s
sh
own in Fig
u
re
2.
(a) Sen
s
o
r
no
de in the tunn
el
(b) Mi
crowave transmitter
Figure 1. Sensor n
ode a
nd
microw
ave transmitte
r in the tunnel
M
i
c
r
o
w
av
e
Tr
a
n
s
m
i
t
t
e
r
S
e
ns
or
N
ode
s
Mi
n
e
T
u
nne
l
T
r
ans
m
i
t
t
i
ng
A
n
t
e
nna
Figure 2. Microwave
wirel
e
ss
cha
r
gin
g
schem
e in min
e
tunnel
Figure 2 sh
ows the microwave wirel
e
ss
ch
argi
ng
sch
eme in a mine tunn
el. The
microwave transmitte
r is
p
l
aned to b
e
i
n
stalle
d on
p
e
riodi
c m
obil
e
devices,
su
ch a
s
lo
com
o
tive
for tunnel tra
n
sp
ortation,
scrap
e
r conv
eyor and
mo
nkey vehicl
e
et al. These
mobile device
s
cover almo
st
the area
s wh
ere th
e
sen
s
or n
e
two
r
k n
ode
s existin
g
,
so m
o
st of t
he no
de
s
can
be
cha
r
ge
d
with
the m
o
veme
nt of the
microwave tran
smitter. Before b
r
ing
the
wirel
e
ss
ch
arging
techn
o
logy in
tunnel
s, it is
nec
essa
ry to study the p
o
w
er
ch
arac
te
ristic b
e
twe
e
n
the tran
smitter
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
9
30
Powe
r Ch
ara
c
teri
stics of Wirel
e
ss Cha
r
ging fo
r Sen
s
or
Networks in Tunnel
s
(E
njie Ding)
1181
and the
sen
s
or no
de
s. Th
us to an
alyze
the power
transmi
s
sion
efficien
cy and f
u
rthe
r theo
ret
i
cal
basi
s
for the
device d
e
si
gn
is provide
d
.
3. Descrip
tio
n
of the Me
a
s
uremen
t
s
3.1. The Und
e
rgroun
d Mining En
v
i
ronment
Two
typical
rectan
gula
r
tu
nnel
s in
the
Jin
D
a
Mine, l
o
cate
d in
the
city of
Ten
g
Zho
u
,
Shando
ng Province, were
cho
s
en fo
r the expe
rime
nts. The tun
n
els in th
e e
x
perime
nts h
a
ve
different dime
nsio
ns a
nd wall material
s. The wi
dth,
he
ight and wall
material
s of e
a
ch tun
nel a
r
e
sho
w
n in Ta
b
l
e 1.
(a) T
unnel
with rock roof an
d con
c
rete si
de
wall
s
(b) T
unnel
with rock roof an
d coal
side
walls
Figure 3. Cro
ss
se
ction of the tunnel
Table 1. Information of t
he tunnel
s ado
pted in experi
m
ents
Width(m)
Height(m)
Roof Material
Side-w
a
ll Materia
l
Experiment
1
2.6 2.2 rock concrete
Experiment
2
3.5 2.4 rock
coal
As
sho
w
n
in
Figure 3,
the
tunnel
se
ctio
n is re
cta
n
gul
ar. Th
e thi
c
kn
ess of
the
co
ncrete
at
the tunnel lini
ng is a
p
p
r
oxi
m
ately 10 cm, and that
on the tunn
el
floor is
app
roximately 25
cm.
Outsid
e of th
e tunn
el i
s
rock mate
rial.
The i
r
o
n
wire enta
nglem
ents
are
pla
c
ed b
e
twe
en t
h
e
con
c
rete a
n
d
the sand
ro
ck
with the
distribution
of sq
uare
gri
d
(ora
nge lin
e
sho
w
n in
Figu
re
2),
whe
r
e the dia
m
eter of the iron wi
re i
s
0.2 mm and
th
e width of the
squa
re g
r
id i
s
10 cm. Some
tunnel
s use rock for the ro
of and floor,
while oth
e
rs use
coal a
s
si
de wall
s.
One
side
of t
he tunn
el h
a
s 4 cable
s
, ea
ch
wi
th a
dia
m
eter of
10
cm. The volta
ge of the
cabl
es i
s
60
0
volt, and the freque
ncy of
the alte
rn
ating cu
rrent is
60Hz. The
di
stan
ce bet
we
en
cabl
es i
s
8 cm, with a
5-cm dista
n
ce to
the t
unn
el wall and
an
ap
proximately 1
.
4-m di
stan
ce
to
the tunnel
flo
o
r. On
e water pipe
with a
5
-
cm
diam
et
er is o
n
the
oth
e
r
side
of the
tunnel, a
nd t
h
e
distan
ce to th
e tunnel floo
r is app
roxima
tely 1.2
cm. In addition, th
ere i
s
a drain
age chan
nel
on
the pipe
sid
e
, with a
width
of 30 cm an
d
a depth
of
20
cm. Two rails are
distri
bute
d
on the
tunn
el
floor a
pproximately 10
cm
from th
e cen
t
er. Anothe
r
cable i
s
o
n
the
roof
of the tu
nnel
with a
1
0
-
cm dist
ance from the top for the floodlig
ht.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 117
9 – 1186
1182
3.2. Experimental Sy
stem
Two sta
nda
rd half-wave dipole ante
n
nas o
per
atin
g at 2.4 GHz we
re u
s
ed
in these
experim
ents.
As sho
w
n in
Figure 3, the trans
mitting
antenn
a wa
s placed in th
e tunnel ce
nter
(marke
d a
s
O). When th
e anten
na i
s
locate
d in th
e ce
nter
of the tunn
el, th
e insertio
n lo
ss i
s
least. Th
e tra
n
smitting
ant
enna
was co
nne
cted to
th
e micro
w
ave
sign
al
sou
r
ce
(R&S
-SMB1
00A)
by a 2-m-lo
n
g
low-l
o
ss RF cable (go
r
e
X
N344
9,
0.4 dB/m at 2.4
GHz). The o
u
t
put powe
r
of the
sign
al so
urce
was
set to 16 dBm. The receivin
g ante
nna was pl
aced at three lo
cation
s A, B
and
C, whe
r
e th
e
sen
s
o
r
s
were
usually lai
d
o
u
t. The
re
ceiv
ing a
n
tenn
a
wa
s
con
n
e
c
te
d to
a
spe
c
tru
m
analyzer (GS
P
-830E
) by a 5-m-lo
ng ca
ble (go
r
eX
N3
449), an
d a PC wa
s also
con
n
e
c
ted to the
spe
c
tru
m
a
n
a
lyzer to record th
e d
a
ta. Multi-p
a
th ef
fects
existed
in the tun
nel,
and th
e
sign
als
cha
nge
d dra
m
atically du
ri
ng the expe
ri
ments. In
ord
e
r to co
nveni
ently record t
he data, we
set
the spe
c
tru
m
analyze
r
to show the test
data aver
a
g
e
d
over 20 times. The whol
e test system
is
s
h
ow
n
in
F
i
gu
r
e
4
.
Figure 4. The
whole te
st system
The re
ceivin
g
anten
na was arrang
ed at
three
lo
cat
i
ons A, B an
d C, a
s
sh
o
w
n in
the
Figure 3. At
the begin
n
ing
of the test,
the
re
ceiving
antenn
a and
transmitting antenna were on
the sam
e
cro
ss
se
ction. T
he tran
smittin
g
antenn
a was move
d al
ong the tun
n
e
l axis. The
data
were record
e
d
every 0.2
m
,
and the tota
l tested
di
sta
n
ce
wa
s 1
0
m. Powe
r at l
o
catio
n
s A, B
and
C was m
e
a
s
ured
und
er X
,
Y and Z ant
enna
pola
r
iza
t
ions. Each lo
cation
obtain
ed thre
e sets of
data, totaling nine sets.
4. Results a
nd Analy
s
is
Several me
a
s
ureme
n
t ca
mpaign
s h
a
ve been
co
nd
ucted
as di
scussed a
bove.
Prior to
descri
b
ing th
e experi
m
ent
al re
sults, th
e co
ordi
nate
system
of a
typical tunn
el co
mplex
wa
s
defined for re
feren
c
e in th
e sub
s
e
quent
data analys
i
s
. The Z axis is defined a
s
the longitudi
nal
dire
ction
of t
he tun
nel, th
e X axi
s
a
s
t
he
width,
a
n
d
the Y
axis
as th
e h
e
ight
. The
co
ordi
n
a
te
origin i
s
defin
ed in the tun
nel floor
cent
er. The initial
Z coo
r
din
a
te
s of the tran
smitting anten
na
wa
s set to 0 i
n
every experiment. The experim
ental
re
sults a
r
e di
scussed u
nde
r three
con
d
itio
ns
as
follows
.
4.1. Po
w
e
r
Charac
teris
t
ic
s in the Tun
nel
w
i
th Ro
c
k
Roo
f
and
Concr
e
te Side Walls
Figure 5 illu
st
rates the received po
we
r
of lo
catio
n
s A
,
B and C u
n
der th
ree
pol
arization
con
d
ition
s
. Studie
s
[8] and [13] present a
comp
lete m
u
ltimode-wave
guide mo
del to cal
c
ulate th
e
received p
o
wer at any lo
ca
tion in a tunn
el, while t
he
real re
sult
s in
the re
ctang
ul
ar tunn
el are
not
con
s
i
s
tent wit
h
the
simulati
on results b
a
s
ed
on
th
e m
u
ltimode
-waveguid
e
mo
del
. In studie
s
[1
5]
and [16], an
adja
c
ent regi
on of the tra
n
smitte
r
wa
s named the f
r
ee
-spa
ce p
r
opag
ation zo
ne
,
whe
r
e th
e lo
ss follo
ws the f
r
ee
-spa
ce m
o
del. In compa
r
iso
n
, the m
o
del de
scri
bed
in [16] i
s
mo
re
pre
c
ise and
can be re
presented a
s
0
0
(
)
(
)
10
l
o
g
10(
)
dB
dB
d
PL
d
P
L
d
n
X
d
(1)
whe
r
e
d
i
s
th
e dista
n
ce
be
tween th
e se
nding a
nd
re
ceiving d
e
vices, an
d
PL
dB
(
d
) is th
e
path loss on
distan
ce
d
.
d
0
is the reference dista
n
ce, which is
set to the initial distan
ce
when
signal
sour
ce
sp
ectrum
an
alyz
er
pc
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
9
30
Powe
r Ch
ara
c
teri
stics of Wirel
e
ss Cha
r
ging fo
r Sen
s
or
Networks in Tunnel
s
(E
njie Ding)
1183
sen
d
ing an
d receivin
g device
s
are in the
same cr
o
s
s se
ction at the
beginnin
g
of the experim
e
n
t.
PL
dB
(
d
0
) i
s
th
e re
ceive
d
po
wer at the
ref
e
ren
c
e
di
stan
ce
d
0
.
n
is th
e
path lo
ss ex
pone
nt, and
σ
is
a zero-m
ean
Gau
ssi
an di
stributed
ran
d
o
m
variable in
dB, whose value ran
g
e
s
fro
m
0.3 to 3.
Figure 6 sh
o
w
s the fitting
curve
s
of the me
a
s
ured
data of experiment
1. In the actual
wirel
e
ss charging appli
c
ati
on, t
he average po
we
r over a 10-m
eter
dista
n
ce is our
con
c
e
r
n.
Thus, th
e val
ue of X
σ
i
s
i
g
nore
d
. Th
e p
a
th lo
ss
expo
nents n at th
ree lo
cation
s
A, B and
C u
nder
three pol
ari
z
a
t
ion mode
s are sho
w
n in T
able 2.
(a) lo
catio
n
A
(b) lo
catio
n
B
(c
) loc
a
tion C
Figure 5. Re
ceived po
wer
at three locations
a
nd un
de
r three a
n
ten
n
a pola
r
izatio
ns
(a) lo
catio
n
A
(b) lo
catio
n
B
(c
) loc
a
tion C
Figure 6. Fitted cu
rve of t
he received po
wer at three l
o
catio
n
s a
nd
unde
r thre
e a
n
tenna
polari
z
atio
ns
Table 2. Sum
m
ary of path loss expon
ent
s n at
three lo
cation
s an
d u
n
der th
ree p
o
l
arization
s
X Polarization
Z Polari
zation Y
Polarization
A -1.31
-1.577
-0.3143
B -1.279
-1.434
-0.2618
C -0.9488
-1.323
0.3138
As sho
w
n in
Figures 5 a
n
d
6, at locatio
n
A,
the cond
ition of Y polarization ha
s
not only
the highe
st receive
d
power, but also t
he small
e
st
attenuation,
with a path l
o
ss expon
ent
n=
-
0.3143. Th
e
con
d
ition
s
of X and Z pol
a
r
izatio
n are
si
milar, but the
path loss ex
pone
nt |n| of Z
polari
z
atio
n i
s
slightly la
rger t
han
that
of
X p
o
lari
zation. Thu
s
,
the receive
d
po
wer un
de
r Z
polari
z
ation will
decr
ease com
pared t
o
that und
er X polari
z
ation with in
creasing distance.
In
total, the con
d
ition of Y polarization is
fa
r better than t
he other p
o
la
rizatio
n
s.
At the locatio
n
B, the cond
itions of X an
d
Z pola
r
ization are simil
a
r at a sho
r
t di
stan
ce.
Be
c
a
us
e
th
e
PL
dB
(
d
0
) of
X polarizatio
n is large
r
t
han that of
Z polari
zatio
n
, and |
n
| of
X
polari
z
atio
n i
s
slig
htly sm
aller, the
conditi
on of X
pol
ari
z
ation
is sli
g
h
t
ly supe
rio
r
to
Z p
o
lari
zatio
n
.
In addition, Y polari
z
atio
n is optimal after 2 m.
At location
C, the path lo
ss expo
nent n
of Y pol
a
r
iza
t
ion is
positiv
e. In the initi
a
l sh
ort
distan
ce, the
tenden
cy of the re
ceive
d
p
o
we
r in
crea
sed gradu
ally
. The whole
re
ceived
po
wer is
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
9
30
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 117
9 – 1186
1184
grad
ual
whe
n
the value
|
n
| is sm
all. Durin
g
the
initial pha
s
e, the re
cei
v
ing po
wer
of Z
polari
z
atio
n is high
er, but
with increa
sing dista
n
ce, the powe
r i
s
mu
ch lowe
r than that o
f
Y
polari
z
atio
n.
Comp
ari
ng t
he
symmetri
c
al l
o
cation
s A and
C, t
he atten
u
atio
n tren
ds of
X and
Z
polari
z
atio
ns are simila
r. For
the
Y p
o
lari
zation
s, the calculated
powers a
r
e
com
parable.
In
addition, thei
r path lo
ss
expone
nts |
n
| a
r
e
clo
s
e, but
the differe
nt
PL
dB
(
d
0
) resul
t
in the different
sign
s of
n
. In
summ
ary, for locatio
n
s A a
nd C, Y
pola
r
i
z
ation i
s
opti
m
al. Ho
weve
r, for location
B,
X polari
z
ation outperform
s
others before approxim
ately 2 m,
and Y polari
zation is still opti
m
al
after approximately 2 m.
4.2. Po
w
e
r
Charac
teris
t
ic
s in the Tun
nel
w
i
th Rock Roof and
Coal Side Walls
The l
o
cation
of expe
riment
2 i
s
a
se
gm
ent of
a
strai
g
ht tunn
el
wit
h
a
greate
r
width an
d
coal
si
de
wa
lls. Fig
u
re
s
7 an
d 8
sho
w
the
me
asured
curve
s
and th
e
co
rresp
o
ndi
ng fit
t
ing
curves, respectively. The param
et
ers that were
used in the
cu
rve-fitting are illustrated in
Table 3.
(a) lo
catio
n
A
(b) lo
catio
n
B
(c
) loc
a
tion B
Figure 7. Re
ceived po
wer
at three locations
a
nd un
de
r three a
n
ten
n
a pola
r
izatio
ns
(a) lo
catio
n
A
(b) lo
catio
n
B
(c
) loc
a
tion B
Figure 8. Fitted cu
rve of t
he received po
wer at three l
o
catio
n
s a
nd
unde
r thre
e a
n
tenna
polari
z
atio
ns
Table 3. Sum
m
ary of path loss expon
ent
s n at three lo
cation
s an
d u
nder th
ree
polari
z
atio
ns
X Polarization
Z Polari
zation Y
Polarization
A -1.104
-1.225
1.3
B -1.392
-1.74
-0.1012
C -0.627
-2.245
0.9475
At location A, the result
s of X and Z polarizatio
ns a
r
e
simila
r. The initial received
powe
r
s
and th
e p
a
th
loss
expon
e
n
ts u
nde
r
bo
th pola
r
izat
io
ns
are very
clo
s
e, b
u
t X
pola
r
ization
is
sup
e
rio
r
to Z
polari
z
atio
n. The received
powers of
Y
polari
z
atio
n a
r
e si
gnificantl
y
different fro
m
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Powe
r Ch
ara
c
teri
stics of Wirel
e
ss Cha
r
ging fo
r Sen
s
or
Networks in Tunnel
s
(E
njie Ding)
1185
those of X an
d Z polari
z
ati
ons. Th
e initial re
ce
ived p
o
we
r und
er Y
polari
z
ation i
s
far bel
ow t
he
values of X a
nd Z pola
r
i
z
a
t
ions, and th
e
path loss
ex
pone
nt n is pl
us 1.3. Unde
r Y polari
z
atio
n,
with in
crea
si
ng di
stan
ce, t
he received
power
doe
s
not de
cay b
u
t
grad
ually ri
ses. Th
e recei
v
e
d
power of Y polari
z
atio
n surpa
s
ses th
e values
of
X and Z pol
arization
s
at a distan
ce
of
approximatel
y 3 m an
d
excee
d
s th
e
initial value
s
of th
ree
p
o
lari
zation
s
at a di
stan
ce of
approximatel
y 6 m.
At location B, the path loss exp
onent
s n of
Y polarizatio
n are
clo
s
e to 0, and the
cha
nge
s
alon
g the
dista
n
ce a
r
e
relativel
y
stable.
The
initial received po
we
r und
e
r
Y pola
r
ization
is smalle
r th
an that un
de
r X and
Z po
larization
s. T
he conditio
n
s of X and Z
polari
z
atio
ns
are
simila
r; they
have cl
ose in
itial
re
ceived
powers, but t
he ab
sol
u
te
value
of n
of X
polari
z
atio
n
is
slightly less than that of Z polar
i
z
ation.
Hen
c
e, the result
s of
X polari
z
ation a
r
e better than
th
e
results of Z p
o
lari
zation
within 10 m. Th
e cu
rves
of X
polari
z
ation
and Y pola
r
ization interse
c
t at
a distan
ce of
approximatel
y 2 meters.
The conditio
n
s
at the lo
cat
i
on C a
r
e
si
milar to
tho
s
e at locatio
n
A. Howeve
r, the initial
received p
o
wer u
nde
r Y p
o
lari
zation i
s
slightly large
r
; thus, the p
a
th loss exp
o
nent
n
is
l
o
we
r
than that at locatio
n
A. Mean
while, the
initia
l receiv
ed po
wer a
n
d
the path lo
ss exp
one
nt of
X
polari
z
atio
n decrea
s
e. T
he cu
rves o
f
the thr
ee polari
z
atio
ns interse
c
t at a distance
of
approximatel
y 3 m.
Comp
ari
ng l
o
catio
n
s A
a
nd C, th
e m
easure
m
ent
result
s a
r
e
si
milar. Th
e di
fference
betwe
en the
s
e re
sult
s i
s
that the initial
re
ceived
po
wers
und
er t
h
ree
types of
pola
r
ization
s
are
different. Ove
r
all, X pol
ari
z
ation could
o
b
tain la
rg
e
r
receive
d
po
wers when th
e
distan
ce
is l
e
ss
than 3 m, but whe
n
the dist
ance is far fro
m
t
hat, the condition
s of Y
polari
z
atio
n a
r
e optimal.
Comp
ari
ng E
x
perime
n
ts 1
and 2,
we fi
nd that the t
unnel
si
ze a
nd wall mate
rial h
a
ve
little influence
on th
e path
l
o
ss exp
onent
. The a
b
solut
e
value
s
of
n
obtaine
d in
curve fittings a
r
e
approximatel
y 1, whi
c
h a
r
e far le
ss tha
n
the fre
e
spa
c
e p
a
th lo
ss (
n
=
2). Valu
e
s
of
PL
dB
(
d
0
) v
a
ry
greatly. The reason
s are e
s
timated for t
he ca
bl
e
s
, testers and
so
me other u
n
certain facto
r
s.
5. Conclusio
n
In this
pap
er,
we
explo
r
e t
he p
o
wer
mo
del of
wave
prop
agatio
n
betwe
en th
e
sen
d
ing
and
re
ceivin
g devi
c
e
s
in
mine
tunn
els. Th
e ex
p
e
riments we
re
perfo
rmed
in
two
re
ctan
g
u
lar
tunnel
s with
different wall material
s. Th
e tran
smitting
antenna
wa
s pla
c
ed i
n
the middl
e of the
tunnel and th
e receiving a
n
tenna
wa
s placed on th
e tunnel wall
s. We mea
s
u
r
ed the recei
v
ed
power u
nde
r
three p
o
lari
za
tions X, Y an
d Z in di
ffere
nt tunnels. T
he re
sult
s sh
ow that the p
a
th
loss follows the free
-spa
ce model. In a
ll t
unnels, ant
enna
s workin
g on X or Z
polari
z
atio
n can
gain mo
re p
o
w
er
over
sho
r
t distan
ce
s,
but Y polar
i
z
ation is o
p
tim
a
l after ap
pro
x
imately 2 m. In
addition, u
n
d
e
r Y
pola
r
ization, with th
e i
n
crea
se
d
di
stance, the
re
ceived po
we
rs som
e
time
s d
o
not de
cay but
grad
ually in
crea
se. Th
e tu
nnel
size an
d
wall m
a
terial
s have little i
n
fluen
ce o
n
the
path lo
ss ex
pone
nt. Ca
bl
es, te
sters
a
nd
so
m
e
oth
e
r
un
certai
n
factors im
pa
ct the
PL
dB
(
d
0
)
greatly.
Referen
ces
[1]
Ishjka
w
a
R, Honjo K.
Efficient
supp
ly pow
er control by PW
M
techniq
ue for
micr
ow
ave w
i
reless p
o
w
e
r
transfer system
s
. Sen
dai, Ja
pan. 20
14; 11
0
1
-11
03.
[2]
Nako S, Okuda K, Miy
a
shiro
K, Komurasaki K, Koizumi H. Wire
less Po
w
e
r T
r
ansfer to
a Microa
eria
l
Vehic
l
e
w
i
t
h
a
Micro
w
av
e Active Ph
ased Arr
a
y
.
Inter
natio
n
a
l Jour
na
l of A
n
tenn
as a
nd P
r
opa
gatio
n
.
201
4; 201
4: 1-5.
[3]
Enji
e D, H
u
i
X, Z
h
ifeng S,
D
uan Z
.
Mo
del
i
ng a
nd
an
al
ysi
s
of
w
i
re
less c
harg
i
ng
s
y
ste
m
base
d
o
n
magn
etic co
upl
ing r
e
so
nanc
e.
Journ
a
l
of Ch
i
na U
n
iv
ersity o
f
Minin
g
& T
e
c
hno
logy
.
20
14;
(05): 9
27-
932.
[4]
Yoshi
da S,
Noj
i
T
,
F
u
kuda G, Koba
ya
shi
Y,
Ka
w
a
s
a
ki S. E
x
p
e
rime
ntal
De
monstratio
n of
Coe
x
ist
e
n
c
e
of Micro
w
av
e W
i
reless C
o
mmunic
a
tion
an
d Po
w
e
r T
r
ansfer
T
e
chnol
og
ies for Batter
y
-F
ree Sen
s
o
r
Net
w
ork S
y
stems.
Internation
a
l Jour
nal of A
n
tenn
as an
d Propa
gati
o
n
. 20
1
3
; 2013: 1-
10.
[5]
Boutin M, Be
nzako
u
r A, D
e
spi
n
s CL, Af
fes S. Radi
o
W
a
ve Ch
aract
e
rizati
on a
nd
Mode
lin
g i
n
Und
e
rgro
un
d Mine T
unnels.
IEEE Transactions on Ant
enn
as and Pro
p
a
g
a
tion
. 20
08; 56
(2): 540-5
49.
[6]
Dua
n
Z
,
Enjie D, Xin W
.
2.4 GHz Radio W
a
ve Prop
agati
on Ch
aracteris
t
ics in Coal Mi
ne W
o
rkfac
e
T
unnels.
T
E
LKOMNIKA Indon
esia
n Journ
a
l o
f
Electrical Eng
i
ne
erin
g
. 201
4; 12(10): 72
99-
730
3.
[7]
Mahmo
ud
NAL
.
Fuzz
y
Log
ic
PSS Assisted
b
y
Ne
ig
hbor
in
g
Sign
als t
o
Miti
gate th
e El
ectromech
anic
a
l
W
a
ve Propa
ga
tion in Po
w
e
r
S
y
stems.
T
E
L
K
OMNIKA Indones
ian J
ourn
a
l of
Electrica
l
Engin
eeri
n
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 117
9 – 1186
1186
201
5; 14(3): 36
3-37
5.
[8]
Z
h
i S, Ak
y
i
l
d
iz
IF
. Chann
el
mode
lin
g an
d ana
l
y
sis for
w
i
reless n
e
t
w
ork
s
in un
dergr
ou
nd min
e
s an
d
road tun
n
e
l
s.
Communic
a
tions
,
IEEE
Transac
tions on
. 2
010;
58(6): 175
8-1
768.
[9]
Ruisi
H, Z
h
a
n
g
dui Z
,
BO A, K
e
G, Bin
gha
o
C,
Jose IA, C
e
sar B. Pro
pag
a
t
ion c
han
ne
l m
easur
ement
s
and
an
al
ysis
at 2.4 GHz in s
u
b
w
a
y
tu
nne
ls.
Microw
aves, Antenn
as & Pro
pag
atio
n, IET
. 201
3;
7(1
1
)
:
934-
941.
[10]
Z
hang YP, H
w
ang Y. Ch
aract
e
ri
zati
on of UH
F
radio pro
p
a
g
a
tion ch
an
nels
in tunn
el e
n
vir
onme
n
ts for
microcel
lul
a
r a
nd pers
o
n
a
l communic
a
tio
n
s
.
Vehicular Technology, IEEE Transactions on
. 1
9
98;
47(1): 28
3-2
9
6
.
[11]
Nerg
uizi
an C,
Despi
n
s CL, A
ffes S, Djadel
M. Radio-c
h
a
n
nel ch
aracter
i
z
a
tion
of an u
n
dergr
oun
d
mine at 2.4 GH
z.
IEEE
Transactions on Wireless Comm
unic
ations
. 20
05; 4
(
5): 2441-
24
53
.
[12]
Rizzo
C, Ler
a
F, Villarro
el J
L
.
T
r
ansversal
fadi
ng
ana
lysis
in strai
ght tu
nn
els at
2.4 GH
z
.
201
3 1
3
th
Internatio
na
l C
onfere
n
ce o
n
IT
S
T
e
lecommu
nicati
ons (IT
S
T
)
,
T
a
mpere. 20
13: 313-
31
8.
[13]
Yu H, Xu Z
,
Feng
xue L, Z
h
a
o
X. An
gul
ar P
o
w
e
r D
i
stributi
on of W
i
rel
e
ss
Chan
ne
l in Mi
ne T
unnels
.
T
E
LKOMNIKA: Indones
ia
n Jo
urna
l of Electri
c
al Eng
i
ne
eri
n
g
. 2013; 1
1
: 14
22-1
435.
[14]
Jipin
g
S,
Li
ngf
ei
C, Ch
an
gse
n
Z
.
Influ
ence
of
con
ductiv
i
t
y
on
ra
dio
w
a
v
e
s pr
opa
gati
o
n
in
tun
nels
.
Journ
a
l of Li
ao
nin
g
T
e
chnic
a
l
Univers
i
ty
. 200
7; 26: 96-9
8
.
[15]
Rissafi
Y, T
a
lbi L, Gh
ad
dar
M. Exp
e
rim
ent
al
C
har
acteriz
a
tion
of an U
W
B
Propa
gati
on Cha
n
n
e
l i
n
Und
e
rgro
un
d Mines.
IEEE Transactions on An
tennas and Propagation
. 2
012; 60(
1): 240
-246.
[16]
Ke G, Z
hang
d
u
i Z
,
Bo A, Rui
s
i H, Bin
gha
o
C,
Yuan
xua
n
L
,
Cesar BR. C
o
mpl
e
te Prop
a
gatio
n Mod
e
l
in T
unnels.
Ant
enn
as an
d W
i
reless Pro
p
a
gat
ion L
e
tters, IEEE
. 2013; 12:
741-
744.
Evaluation Warning : The document was created with Spire.PDF for Python.