TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol. 12, No. 9, September
2014, pp. 67
5
0
~ 675
7
DOI: 10.115
9
1
/telkomni
ka.
v
12i9.506
1
6750
Re
cei
v
ed
No
vem
ber 5, 20
13; Re
vised
Apr 23, 201
4; Accept
ed Ju
ne 17, 201
4
Finite-Time Stabilization of Networked Control Systems
with Packet Dropout
Yanling Shang*
1
, Ye Yuan
2
1
School of Softw
a
r
e, An
ya
ng
Normal U
n
iv
ersit
y
,
An
yan
g
45
500
0, Chin
a
2
School of Mat
hematics a
nd
Ph
y
s
ics, Suzh
ou
Un
iversit
y
o
f
Science an
d
T
e
chnolog
y,
Suzho
u
21
500
9, Chin
a
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: hnnhs
yl
@1
2
6
.com
A
b
st
r
a
ct
The problem
of
finite-t
ime stabili
z
a
tion for ne
tworked control
system
s
wi
th b
o
th
sensor-to-c
ontrol
l
e
r
and c
ontrol
l
er-t
o-actuator
pac
ket drop
out
s is
investi
gate
d
in
this pa
per.By
u
s
ing th
e iterativ
e ap
proac
h, the
NCSs with bounded packet
dr
opout is
modeled as
switc
hed linear system
s
.
Sufficient
c
o
nditions for finite-
tim
e
stabili
z
ation of the
underlying
systems are derived
via linear matr
ix inequalities
(LMIs). Lastly, an
illustrativ
e
exa
m
p
l
e is g
i
ven t
o
de
mo
nstrate
the effectiven
e
ss of
the propo
sed resu
lts.
Ke
y
w
ords
: net
worked control
system
s, pack
e
t dr
opout, finite-time stability, LMIs
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
Networked control syste
m
s
(NCSs)
are fee
dba
ck cont
r
o
l sy
st
em
s wit
h
net
wo
rk
cha
nnel
s use
d
for
the com
m
unication
s. Comp
ared wi
th the traditio
nal p
o
int-to
-p
oint wi
ring, th
e
use
of the
communi
catio
n
chann
els can redu
ce
th
e cost
s of
cable
s
an
d p
o
w
er,
sim
p
lify the
installatio
n
a
nd mainte
nan
ce of the wh
ole syste
m
, and incre
a
se the reli
ability. The NCSs
h
a
ve
many indu
stri
al appli
c
ation
s
in a
u
tomobi
les, ma
n
u
fact
uring plant
s,
aircrafts,
a
nd HVAC syste
m
s
[1]. Howeve
r, the insertion
of the com
m
unicati
on n
e
twork in fe
e
dba
ck
co
ntrol
loop ma
ke
s
the
analysi
s
and
desi
gn of an
NCS compli
cated be
cau
s
e
it introduce
s
some p
r
o
b
le
ms existing in
the
net
wo
rk i
n
t
o
cont
r
o
l sy
st
e
m
s
su
ch a
s
li
mit
ed
commu
nicatio
n
ba
nd
width, net
work-in
d
u
c
ed
del
ay,
packet
s
di
so
rde
r
a
nd
p
a
ckets lo
ss whi
c
h
ofte
n ha
ppe
n i
nevitably du
ring i
n
form
ation
transmissio
n see the
references [2-8]a
n
d
the referen
c
e
s
cited the
r
ein.
Among a
nu
mber
of issu
es a
r
isi
ng fro
m
su
ch
a fra
m
ewo
r
k, pa
cket lo
ss
of NCSs i
s
an
importa
nt issue to be add
resse
d
and h
a
s be
en re
ce
iving great at
tentions. Fo
r instan
ce, Xio
ng
and L
a
m [9]
studie
d
the p
r
oble
m
of
sta
b
ility and st
a
b
ilizatio
n of li
near sy
stems over n
e
two
r
ks
with boun
ded
packet loss. Bakule an
d De La Sen [
10] tackl
ed the pro
b
lem
of dece
n
trali
z
ed
stabili
zation of
netwo
rked
complex co
mposite
syst
ems
with non
linear p
e
rtu
r
b
a
tions. Wang
and
Yang [11] in
vestigated th
e probl
em of
state-fe
e
dba
ck
cont
rol sy
nthesi
s
for n
e
tworke
d co
n
t
rol
system
s with packet dro
p
o
u
t.
Sun
a
nd Qin
[12]
studi
ed NCS
s with
both se
nsor-to-co
ntroll
er a
nd
controlle
r-to
-
actuato
r
pa
cket dro
pout
s via switch
ed
system ap
proach. For
m
o
re detail
s
of th
e
literature rela
ted to netwo
rked p
r
obl
em
s with packe
t
drop
out, the read
er i
s
refe
rre
d to [13-1
8
]
and the refe
rences the
r
ein
.
It is wo
rth p
o
inting o
u
t m
o
st of exi
s
tin
g
literatu
r
e
relate to
stabi
lity and pe
rfo
r
man
c
e
crite
r
ia defin
ed over an i
n
finite-time interval. Ho
wever, the main attention in many pra
c
tical
appli
c
ation
s
is the beh
avio
r of t
he dynamical sy
stem
s over a fixed
finite-time; for example, large
values of the
state are no
t acce
ptable i
n
the pre
s
en
ce of satu
rati
ons [19, 20]. In this sen
s
e
it
appe
ars rea
s
onabl
e to define as sta
b
le
a system
whose state, g
i
ven some
in
itial condition
s,
remai
n
s
withi
n
pre
s
crib
ed
boun
ds in th
e fixed fini
te-time interval. For this p
u
rposes finite
-time
stable (FTS)
coul
d be use
d
[21, 22]. Rece
ntly, Amato
et al.
extends thi
s
con
c
ept to finite-time
boun
dedn
ess in [23].
To d
a
te, with th
e
aid of li
nea
r
matrix ine
qua
lities (LMIs) f
o
rmul
ation, m
o
re
results of finite-time stabi
lity and stabilization of
variou
s syste
m
s. For mo
re
details of the
literature
related to finite-time
stability, the reader
i
s
referred to [24-28], and t
he
references
therein.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Finite-Tim
e Stabilization of Ne
tworked
Control System
s with
Packet
Dropout (Y
anling Shang)
6751
Ho
wever,
to
the
be
st of
ou
r
kno
w
le
dge,
the
fini
te-time
stabil
i
ty and
stabi
lization
probl
em
s for
NCS
s
with p
a
cket d
r
opo
ut have not b
e
en fully investigated to dat
e. Espe
cially, for
the ca
se
where
both
sensor to
-con
troller
and
controlle
r-to
-
a
c
tuator pa
cket dro
pout
s
are
con
s
id
ere
d
simultaneo
usly
, very few
result
s re
late
d to NCS
s
are
availabl
e
in the
exist
i
n
g
literature, whi
c
h motivates
the study of this pa
per.
In this pa
per,
the finite-tim
e stabili
zatio
n
pro
b
lem
s
o
f
a cla
ss
of NCS
s with
b
ound
ed
packet dropo
ut is studie
d
. Firstly, we m
odel t
he NCS
s
with bo
und
ed pa
cket dro
pout as
swit
ched
linear
system
s. Then, the
con
c
e
p
ts of the finite-time
stability (FTS
) and
pro
b
le
m formulatio
n
are
given. The
main
contri
b
u
tion of this pape
r i
s
to
desi
gn a
st
ate-feed
ba
ck cont
rolle
r which
guarantee
s t
he re
sultin
g
the re
sulting
clo
s
ed
-lo
o
p
discrete
-tim
e syste
m
un
iform finite-ti
m
e
stable.
In the sequel
, the following notation wil
l
be used: Th
e symbol
s
n
R
and
nm
R
stand for
an n-dimen
s
i
onal Eu
clidea
n spa
c
e
and
the set of all
nm
real mat
r
ices,
respe
c
tively,
T
A
and
1
A
den
ote the
matrix tra
n
s
po
se
an
d
matrix inverse,
diag
A
,
B
re
presen
ts the
blo
c
k-
diago
nal m
a
trix of
A
and
B
,
0
P
stand
s fo
r a
positive-defini
t
e matrix,
I
is the unit matri
x
with app
rop
r
i
a
te dimen
s
io
ns, and
{1
,
2
,
}
Z
.
2. Problem Formulation a
nd Preliminaries
The frame
w
o
r
k of
NCS
s
consi
dered i
n
t
he p
ape
r
i
s
d
epicte
d
in
Fig
u
re
1. Th
e
proce
s
s to
be co
ntrolle
d is model
ed b
y
a linear discrete-time
syst
em.
(1
)
(
)
(
)
x
kA
x
k
B
u
k
(1)
Whe
r
e
kZ
is th
e time step,
()
n
x
kR
and
()
m
uk
R
are is syst
em state and
control input,
r
e
spec
tively.
A
and
B
are kn
own real con
s
ta
nt matrice
s
wi
th appro
p
ri
ate dimen
s
ion
s
.
Figure 1. Illustration of NCSs over Com
m
unication Network
We ma
ke the
following a
ssumption
s abo
ut the NCS:
1) Net
w
o
r
ks
exist betwe
en
sen
s
or a
nd c
ontrolle
r, and
betwe
en cont
rolle
r and a
c
t
uator;
2) The
sen
s
o
r
is clo
c
k driv
en; the cont
ro
ller and the a
c
tuator
are ev
ent driven;
3) The d
a
ta a
r
e tran
smitted
in a single p
a
cket at each
time step.
Let
12
{,
,
}
ii
, whi
c
h a
sub
s
e
que
nce
is of
{1
,
2
,
}
Z
, denot
e the
sequ
en
ce of
time points
of succe
s
sful
data transm
i
ssi
on from
the sam
p
ler t
o
the zero-o
rde
r
hold, a
nd
1
max
(
)
k
kk
i
si
i
be the maxi
mum pa
cket-l
oss upp
er
bo
und. The
n
th
e followin
g
concept and
mathemati
c
al
models a
r
e i
n
trodu
ce
d to c
aptu
r
e the n
a
ture of pa
cket losses.
The state fee
dba
ck
control
l
er law i
s
:
()
()
uk
K
x
k
(2)
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 67
50 – 675
7
6752
Whe
r
e
mn
KR
is to
be de
signe
d. From the vie
w
poi
nt of
the
zero-ord
er ho
ld, the control
input
is:
()
(
)
(
)
kk
ul
ui
K
x
i
For
1
1
kk
il
i
. The i
n
itial inputs are
set to zeros:
1
()
0
,
0
-
1
ul l
i
. He
nce th
e
cl
ose
d
-
loop sy
stem become
s
:
1
(1
)
(
)
(
)
,
1
kk
k
xl
A
x
l
B
K
x
i
i
l
i
(3)
From the
clo
s
ed-lo
op sy
ste
m
(3), we ca
n
obtain:
1
1
1
1
0
()
(
)
,
kk
kk
ii
ii
r
kk
k
r
x i
A
A
B
K
x
i
i
(4)
Define the p
a
c
ket drop
out pro
c
e
ss a
s
fo
llows:
1
()
kk
k
i i
i
(5)
Whi
c
h takes
values in the
finite state sp
ace
{1
,
2
,
,
}
s
.
Let,
1
1
1
()
1
()
0
()
(
)
(1
)
(
)
kk
kk
k
k
ii
ik
ii
r
ik
r
zk
x
i
zk
x
i
AA
BA
B
(6)
It is easily seen that the clo
s
ed
-loo
p system (4
) ca
n be de
scrib
ed by the follow in switch
ed
sy
st
em.
()
()
(1
)
(
)
(
)
ik
i
k
zk
A
B
K
z
k
(7)
Whe
r
e
1
()
kk
ik
i
i
is arbi
trarily switchi
ng sig
nal.
For sim
p
licity, at any arbitra
r
y discrete time
kZ
, the switching sig
nal
()
ik
is denoted
by
i
. Then, the clo
s
ed
-loo
p system (7
) ca
n
be rewritten
as:
(1
)
(
)
(
)
ii
zk
A
B
K
z
k
(8)
The gen
eral i
dea of finite-time stability concern
s
the b
ound
edne
ss of the st
ate of a system over
a finite-time interval for th
e given initial
condi
tio
n
s; t
h
is con
c
ept can be formali
z
ed th
rou
gh the
followin
g
definition, whi
c
h i
s
an exten
s
io
n to
discrete
-time system
s
of
the one given in [14].
Defini
tion 1:
(Finite-tim
e
st
ability (FTS)).
The
discrete-time swit
ched system
(1
)
(
)
,
i
x
kA
x
k
k
Z
Is
s
a
id to be FTS with res
p
ec
t to
12
(,
,
,
)
cc
N
R
wh
ere
12
0
cc
,
R
is a sy
m
m
et
ric p
o
sit
i
v
e
-
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Finite-Tim
e Stabilization of Ne
tworked
Control System
s with
Packet
Dropout (Y
anling Shang)
6753
definite matri
x
and
0
k
NZ
, if:
12
(
0
)
(
0
)
()
()
,
1
,
2
,
,
TT
x
Rx
c
x
k
R
x
k
c
k
N
The followi
ng
proble
m
will
be dre
s
sed in
this pape
r.
Problem 1.
For th
e di
screte-time
syst
em (1),
we f
i
nd a
net
worked
state
fe
edba
ck
controlle
r (5
) su
ch that the clo
s
ed
-loo
p system is FTS with re
spe
c
t to
12
(,
,
,
)
cc
N
R
.
We next provide a lemma
whi
c
h will play an important role in the late development.
Lemma 1.
The LMI
()
(
)
0
()
()
T
Yx
W
x
Wx
R
x
Is equivale
nt to:
1
()
0
,
()
()
()
0
T
Rx
W
x
R
x
W
x
Whe
r
e
()
()
,
(
)
(
)
TT
Yx
Y
x
R
x
R
x
and
()
Wx
depend o
n
x
.
3. Main Results
In this
section, we
will develop the stabi
liz
ation result
s
for
the closed-loop NCS (8).
T
h
e
followin
g
theorem p
r
e
s
ent
s a suffici
ent
conditio
n
for the finite-time stability of the con
s
id
ered
system
with a
r
bitra
r
y packe
t-loss process.
Theorem
1.
The
clo
s
ed
-l
oop
NCS (8) with
arbit
r
ary packet-l
o
ss pro
c
e
s
s i
s
FTS with
respec
t to
12
(,
,
,
)
cc
N
R
if, there exis
t
pos
i
tive definite matrix
nn
SR
a
nd
sc
a
l
a
r
1
such
that the following mat
r
ix inequalitie
s hol
d:
0
*
ii
SA
S
B
X
S
(
9
)
And,
N
ma
x
12
mi
n
(Q
)
c<
c
(Q
)
(10)
Whe
r
e
1/
2
1
1
/
2
QR
S
R
and
ma
x
()
and
mi
n
()
indicate the maximum and minimum
eigenvalu
e
of
the augment,
resp
ectively. T
hen st
ate feedba
ck co
ntroller is give
n by
-1
K = XS
.
Proof.
Cho
o
s
e a Lyap
un
ov functional
candi
dat
e fo
r the system
(8) a
s
follo
ws:
1
()
()
()
T
Vk
x
k
S
x
k
(11)
Then, alon
g the solutio
n
of system (8
) we have:
1
1
(1
)
(
1
)
(
1
)
(
)
()
()
(
)
T
TT
ii
ii
Vk
x
k
S
x
k
x
k
A
BK
S
A
BK
x
k
(12)
Subs
tituting
-1
K =
XS
into (11
)
a
nd
pre
-
an
d po
st
-multiplying
b
y
-1
-1
di
ag
S
,
S
,
we can ob
tain
the equivale
n
t
condition.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 67
50 – 675
7
6754
11
1
1
1
0
*
ii
SS
A
S
B
X
S
S
(13)
From L
e
mma
1, (12) an
d (13), it follows
that:
()
()
Vk
Vk
(14)
Applying itera
t
ively (14), we can o
b
tain:
()
(
0
)
,
1
,
2
,
,
k
Vk
V
k
N
Noting that
1/
2
1
1
/
2
QR
S
R
and u
s
ing the
fact
1
we have:
1
1/
2
1
/
2
ma
x
ma
x
(0)
(
0)
(0)
(0)
(
0)
(Q
)
(
0
)
(
0
)
(Q
)
(
0
)
(
0
)
kk
T
kT
kT
NT
Vx
S
x
x
RQ
R
x
x
Rx
x
Rx
(15)
And,
1
1/
2
1
/
2
mi
n
()
()
()
(0
)
(
0
)
(Q
)
(
)
(
)
T
T
T
Vk
x
k
S
x
k
x
RQ
R
x
x
kR
x
k
(16)
Putting together (14)-(16
), we obtain:
ma
x
mi
n
(Q
)
()
()
(
0
)
(
0
)
(Q
)
N
TT
x
kR
x
k
x
R
x
(17)
From
(17), it
follows
that
(10) implies that, for all
1,
2
,
,
kN
2
()
()
T
x
kR
x
k
c
. Therefore, th
e
proof follo
ws.
Rem
a
r
k
2.
If
co
ndition
s
(9
) an
d
(10
)
in
Theo
rem
1 i
s
sati
sfied
wit
h
1
, then
system
(8) is
finite-t
ime s
t
able
with respec
t to
12
(,
,
,
)
cc
N
R
for all
0
k
NZ
and
it is also
asymptoticall
y
stable.
Theorem 2.
The
clo
s
ed
-l
oop
NCS (8) with
arbit
r
ary packet-l
o
ss pro
c
e
s
s i
s
F
T
S
with
respec
t to
12
(,
,
,
)
cc
N
R
if, there exist positive defin
ite matrix
nn
SR
and scala
r
s
0
,
0
and
1
s
u
c
h
that the following
matrix inequalities hold:
0
*
ii
SA
S
B
X
S
(18)
0
*
SS
R
R
(19)
0
*
RI
S
(20)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Finite-Tim
e Stabilization of Ne
tworked
Control System
s with
Packet
Dropout (Y
anling Shang)
6755
12
0
N
cc
(21)
Whe
r
e
1/
2
1
1
/
2
QR
S
R
and
ma
x
()
and
mi
n
()
indicate the maximum and minimum
eigenvalu
e
of
the augment,
resp
ectively. T
hen st
ate feedba
ck co
ntroller is give
n by
-1
K = XS
.
Proof.
According to The
o
rem 1, it suffices to
prove condition
(10
)
is
gua
ra
nteed
by (19)-
(21
)
.
Us
ing Lemma 1, it follows
that:
1
00
*
SS
R
SR
S
S
R
S
R
and
11
00
*
RI
SR
S
R
S
From the a
b
o
v
e two equati
ons, we have
:
1
RS
R
Whi
c
h mea
n
s that:
1/
2
1
1/
2
I
RS
R
I
Noting that
1/
2
1
1/
2
QR
S
R
,
we
c
an obtain the following relation:
I
QI
(22)
On the othe
r hand, from
(2
1), we have:
12
N
cc
(23)
Putting (22) a
nd (23
)
toget
her, we h
a
ve:
N
ma
x
11
2
mi
n
(Q
)
c<
c
(Q
)
N
c
(24)
This
compl
e
tes the p
r
oof.
Rem
a
r
k
4
. We ca
n see th
at the co
nditi
ons i
n
The
o
rem 2 a
r
e not
LMIs. Ho
we
ver, once
we fix
an
, they can be tu
rned i
n
to LMIs ba
sed fea
s
ibility proble
m
whi
c
h can
be solved vi
a
existing software (fo
r
exam
ple the LM
I Control To
olbo
x of MATLAB).
5. An Illustrativ
e
Example
To illu
strate
the
effectivene
ss of th
e
p
r
op
osed
method,
we
present a
nume
r
ical
example. Co
nsid
er the sta
t
e-sp
ace plan
t model:
23
1
(1
)
(
)
(
)
12
1
x
kx
k
u
k
Since
pa
cket-loss p
r
o
c
e
s
s i
s
a
r
bitrary, we
can o
b
tain t
he p
a
cket-l
oss u
ppe
r b
oun
d
2
s
, for given
1
1
c
,
2
10
c
,
2
RI
,
100
N
, if let
1.11
, b
y
solving LMI
s
(2
0)-(2
3) b
y
Theo
rem 2, we can o
b
tain
state feedb
ack co
ntrolle
r:
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 67
50 – 675
7
6756
(
)
-
3
.
5
843
-
4
.
2463
(
)
uk
x
k
Whi
c
h en
su
re
the closed-l
o
op NCS is F
T
S
with respe
c
t to
12
(,
,
,
)
cc
N
R
.
5. Conclusio
n
In this pa
per,
the finite-tim
e stabili
zatio
n
pro
b
lem
s
o
f
a cla
ss
of NCS
s with
b
ound
ed
packet d
r
o
p
o
u
t is inve
stig
ated. The
m
a
in contri
buti
on of thi
s
p
aper is t
hat
both
sen
s
o
r-t
o
-
controlle
r an
d co
ntrolle
r-t
o-a
c
tuato
r
p
a
cket
dropo
uts have b
e
en taken int
o
acco
unt. The
sufficie
n
t co
n
d
itions fo
r fini
te-time sta
b
ili
zati
on of
the unde
rlying system
s are de
rived via LMI
s
formulatio
n. Lastly, an ill
ustrative
example is
give
n to dem
on
strate the effe
ctivene
ss
of the
prop
osed results.
Ackn
o
w
l
e
dg
ements
This
wo
rk i
s
sup
porte
d by
Nation
al Nature S
c
ien
c
e
Found
ation o
f
China
und
e
r
Grant
6107
3065
and
and the Key Progra
m
ofScien
c
e Te
ch
n
o
logy Re
sea
r
ch of Education De
partme
n
t
of Hena
nProv
i
nce u
nde
r Grant 13A120
0
16.
The a
u
tho
r
woul
d li
ke to
than
k the
e
d
itor a
nd th
e an
onymou
s
reviewers
for thei
r
con
s
tru
c
tive comment
s and
sugg
estio
n
s
for improvin
g the quality of the pape
r.
Referen
ces
[1]
GC W
a
lsh, H
Ye. Sche
dul
in
g of n
e
t
w
orke
d contro
l s
y
stems.
IEEE Control Syst. Mag.,
200
1; 21
:
57
−
65.
[2]
GC Walsh,
H
Ye, LG B
u
shn
e
ll. Sta
b
il
it
y a
n
a
l
y
sis
of
net
w
o
rked c
ontrol
s
ystems.
IEEE Trans. Contr
o
l
Syst. T
e
chnol.,
2002; 1
0
; 438-
446.
[3]
W
Zhang, MS Branick
y, SM Phill
ips.
Stab
ilit
y of n
e
t
w
o
r
k
e
d
control s
y
stem
s.
IEEE Trans.
Control Syst.
Mag
., 2001; 2
1
:
84-99.
[4]
D Yu
e, QL H
a
n, J L
a
m. Net
w
o
r
k-bas
ed
ro
bust H
∞
c
ontr
o
l of
s
y
st
ems
w
i
t
h
u
n
certa
i
nt
y.
Aut
o
m
a
tica
.
200
5; 41: 999-
100
7.
[5]
X J
i
an
g, QL
H
an, SR
Li
u, A
Xu
e. A
ne
w
H
∞
stabil
i
zati
on
criterion
for
ne
t
w
ork
ed c
ontro
l s
y
stems,”
IEEE Trans. A
u
tomat. Control
. 2008; 5
3
: 10
25-1
032.
[6]
H Gao, T
Chen, J Lam. A ne
w
de
la
y s
y
st
e
m
appro
a
ch to
net
w
o
rk-b
ase
d
control.
Autom
a
tica
. 2
0
08;
44; 39-5
2
.
[7]
J Jameel. Perf
ormanc
e enh
a
n
ceme
nt of
w
i
r
e
le
ss comm
uni
cation s
y
stems
using transm
i
t and rece
iv
e
diversit
y
.
IJACT
. 2010; 2(3): 163-
171.
[8]
J H
u
, GY L
i
, Y
Son
g
. Si
gn
al f
r
aming
meth
od
s of a
u
tomotiv
e
em
bed
de
d
n
e
t
w
o
r
ke
d co
ntr
o
l s
y
stem
for
ban
d
w
i
d
th con
s
umptio
n optim
izatio
n.
JDCTA
. 4(4): 15-27, 2
010.
[9]
J Xi
on
g, J La
m. Stabiliz
atio
n of li
ne
ar
s
y
stems
over net
w
o
rks
w
i
th bou
n
ded pack
e
t
los
s
.
Automatica
.
200
7; 43(1): 80
-87.
[10]
L Baku
le, M
D
e
La
Se
n.
Dec
entrali
z
ed stabili
z
a
tion
of networ
ked complex com
p
osite
system
s wit
h
non
lin
ear p
e
rturbati
ons
. Proceedings of the 2009 IEEE Inter
national Conference
on Control and
Automatio
n
. 20
09; 1-3: 22
72-2
277.
[11]
YL W
ang, GH Yang.
State feed
back contr
o
l s
y
nthes
is for net
w
o
rke
d
control s
y
stems
w
i
th p
a
cke
t
drop
out.
Asian
Journ
a
l of Co
ntrol
. 200
9; 11(1)
: 49-58.
[12]
YG Sun, SY Qin. Stabil
i
t
y
and sta
b
il
izati
on of n
e
t
w
ork
ed co
nt
rol s
y
s
t
ems
w
i
t
h
b
o
u
nde
d pack
e
t
drop
out.
Acta Autom
a
tica Sinica
. 201
1; 37: 113-
118.
[13]
K
T
s
umura, H Ishii, H Hosh
in
a.
T
r
adeoffs bet
w
e
e
n
qu
antiz
ation a
nd p
a
ck
et loss in net
w
o
rked co
ntro
l
of linear s
y
stems.
Autom
a
tica
. 2009; 4
5
(12)
: 2963-2
9
7
0
.
[14]
Y Niu, T
Jia,
X W
ang, F
Ya
n
g
. Output-feed
back co
ntrol
d
e
sig
n
for N
C
S
s
subj
ect to q
u
antizati
on
an
d
drop
out.
Information Sci
enc
es.
2009; 1
79(2
1
): 3804-
38
13.
[15]
Y Ishido, K T
a
kaba, DE Que
v
edo. Stab
ilit
y
ana
l
y
sis of n
e
tw
o
r
ke
d contro
l
s
y
stems su
bje
c
t to packet-
drop
outs an
d finite-l
evel q
u
a
n
tizatio
n
.
Systems & Control Let
ters
. 2011; 60(
5): 325-3
32.
[16]
OC Imer, S Yuksel, T
Basar.
Optimal c
ontr
o
l
of LT
I s
y
ste
m
s over
unr
eli
abl
e comm
uni
cation
li
nks,
”
Autom
a
tica
. 2
0
06; 42(9): 1
429
-143
9.
[17]
Y Z
hao, H Gao,
T
Chen. F
u
zz
y
co
nstrain
e
d
pred
ictive c
ontrol of n
on-l
i
near s
y
stems
w
i
t
h
pack
e
t
drop
outs.
IET
Contro
l T
heory
& Applicati
ons
. 2010; 4(9): 16
65-1
677.
[18]
J Z
hang, Y Xi
a, Desig
n
of H
∞
fuzz
y
contro
llers for non
lin
ear s
y
stems
w
i
th rand
om dat
a drop
outs,”
Optima
l Co
ntro
l Appl
icatio
ns a
nd Metho
d
s
. 2
011; 32(
3): 328
-349.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Finite-Tim
e Stabilization of Ne
tworked
Control System
s with
Packet
Dropout (Y
anling Shang)
6757
[19]
EI-Gohar
y
.
Opt
i
mal contr
o
l of an an
gul
ar motion of a rig
i
d b
o
d
y
dur
ing i
n
fi
nite an
d finite-t
ime interv
als.
Appl. Math. Com
p
ut.,
2003; 1
41: 541-
55
1.
[20]
EI-Gohar
y
,
AS
AI-Ruzaiza. O
p
timal co
ntrol
of non-
homo
g
e
nous
pre
y
–
p
re
dator mo
dels
d
u
rin
g
infi
nit
e
and fin
i
te-time i
n
tervals.
Appl. Math. Comput
.
,
2003;1
46: 49
5-50
8.
[21] P
Dorato.
Shor
t tim
e
stability in linear time-v
arying system
s
.
Proc. IRE internatio
nal co
nve
n
tion rec
o
r
d
196
1; part 4: 83-87.
[22] L W
e
iss, EF
Infante. Fini
te time stabilit
y un
der pertur
b
in
g forces and o
n
prod
uct space
s
.
IEEE Trans.
on Auto
mat. C
ontrol
. 19
67; 1
2
: 54-59.
[23]
F
Amato, M Ariola, P D
o
rate.
F
i
nite-time co
n
t
rol
of lin
ear s
ystems subj
ect to
par
ametric u
n
certai
ntie
s
and d
i
sturb
anc
es.
Automatica.
2001;
37; 14
5
9
-14
63.
[24]
E Moulay
, W
Perruquetti
. Fi
nite tim
e
sta
b
il
it
y
an
d sta
b
il
iz
atio
n
of
a cl
as
s of co
ntin
uou
s s
y
st
ems. J.
Math. Anal. Appl.,
2006; 3
23; 143
0-14
43.
[25]
E Moula
y
, M Dambri
ne, N
Yega
nefar, W Perruq
uetti
. Finite-time stab
il
it
y
a
nd stab
iliz
ation of time-
del
a
y
s
y
stems.
Systems & Co
ntrol Letters.
2
008;
57: 5
61-5
66.
[26] Q Ming, Y Shen. F
i
nite-time H
∞
control for linear continuous s
y
st
em
w
i
th norm-bounded disturbanc
e.
Co
mmun. N
onl
ine
a
r Sci. Nu
mer. Simu
lat.
20
09; 14: 10
43-1
049.
[27]
F
Amato, M Ariola, C C
o
se
n
t
i
no. F
i
nite-tim
e control
of di
screte-
time li
n
ear s
y
stems: ana
l
y
sis a
n
d
desi
gn co
nditi
o
n
s.
Automatica.
2009; 46: 91
9
-
924.
[28]
X L
i
n, H D
u
S
Li. Unif
orm finite-time sta
b
il
it
y
a
nd fe
ed
ba
ck stabiliz
atio
n
for discrete-ti
me s
w
itche
d
line
a
r s
y
stems
and
its ap
plic
ation to
net
w
o
rk
ed co
ntrol s
y
st
ems
. Contro
l a
nd D
e
cisi
on
. 2
011; 6:
841-
846.
Evaluation Warning : The document was created with Spire.PDF for Python.