Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
Co
mput
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
9
, No
.
5
,
Octo
ber
201
9
, pp.
3550
~
35
57
IS
S
N:
20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v
9
i
5
.
pp3550
-
35
57
3550
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
Effici
ent
e
rror
c
orre
cting
schem
e for c
h
aos
s
hi
f
t
k
ey
ing sign
als
Hikma
t N. Ab
dull
ah
1
,
Tham
ir
R
. Saee
d
2
,
Asaad H.
S
ahar
3
1
Coll
ege of
Infor
m
at
i
on
Eng
i
neer
ing,
Al
-
Nah
rai
n
Univer
sit
y
,
Ira
q
2
,3
Dep
art
m
ent
of
Elec
tr
ical Engi
n
ee
ring
,
Univ
ersity
of
T
ec
hno
log
y
,
Ir
aq
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Ju
l
31
, 2
01
8
Re
vised
Ma
r
19
, 2
01
9
Accepte
d
Apr
9
, 2
01
9
An
eff
ec
ti
v
e
err
or
-
cor
recti
on
sc
heme
base
d
on
norm
al
iz
ed
cor
r
el
a
ti
on
for
a
non
cohe
r
ent
c
haos
comm
unic
at
ion
s
y
stem
w
it
h
no
r
edundanc
y
bit
s
is
proposed
in
thi
s
pape
r
.
A
m
odi
fie
d
logi
sti
c
m
a
p
is
used
in
th
e
proposed
sche
m
e
for
gene
rating
two
seque
nce
s,
one
for
eve
r
y
d
at
a
bi
t
val
ue,
in
a
m
anne
r
tha
t
the
i
nit
ial
val
u
e
of
the
next
cha
o
ti
c
se
quenc
e
is
set
b
y
the
sec
on
d
val
ue
of
the
pre
sent
cha
o
ti
c
seque
nce
of
the
sim
il
ar
s
y
m
bol.
Thi
s
arr
ange
m
ent,
th
us,
has
the
cre
a
ti
on
of
succ
essive
cha
o
ti
c
s
equ
enc
es
wit
h
ide
ntica
l
cha
o
tic
d
y
n
amics
for
err
or
cor
re
ct
io
n
purpose.
The
det
ectio
n
s
y
m
bol
is
per
fo
rm
ed
prior
to
c
orre
ction,
on
th
e
basis
of
th
e
suboptimal
rec
e
ive
r
which
a
nchor
s
on
the
computat
ion
of
t
he
shortest
distance
exi
st
ing
bet
wee
n
th
e
re
ce
iv
ed
seque
nc
e
and
the
m
odifi
ed
logi
sti
c
m
ap’
s
cha
otic
tra
j
ec
tor
y
.
Th
e
result
s
of
t
he
sim
ula
ti
on
rev
e
al
not
ice
abl
e
Eb/
No
improvem
ent
b
y
the
proposed
sc
heme
over
the
p
rior
to
the
err
or
-
cor
recti
ng
sche
m
e
with
the
improvem
ent
inc
rea
sing
when
eve
r
the
r
e
is
increa
se
in
the
num
ber
of
sequ
enc
e
N.
Prior
t
o
the
err
or
-
cor
r
ec
t
ing
sche
m
e
when
N=8,
a
gai
n
of
1.
3
dB
i
s
ac
complished
in
E
b
/N
o
at
10
-
3
bit
err
or
prob
ability
.
On
th
e
basis
of
norm
al
i
ze
d
cor
re
la
t
ion,
t
he
m
ost
eff
i
cient
point
in
our
pro
posed
err
or
cor
recti
on
sch
e
m
e
is t
he absence
of
an
y
red
und
a
nt
bit
s
n
ee
ded
wi
th
m
ini
m
um
del
a
y
proc
edur
e
,
in
cont
rast
to
ea
rlier
m
et
hod
tha
t
was
base
d
on
suboptimal
m
et
hod
det
e
ct
io
n
and
cor
recti
on
.
Such
per
form
anc
e
would
ren
der
the
sche
m
e
good
ca
nd
ida
t
e
f
or
applications
r
equi
ring
high rates of
d
at
a
tr
ansm
ission.
Ke
yw
or
d
s
:
Chan
nel c
od
i
ng
Chaotic
s
hift
ke
yi
ng
Error co
rr
ect
i
on alg
ori
thm
Norm
al
iz
ed
cor
relat
ion
Subopti
m
a
l detec
ti
on
Copyright
©
201
9
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Asaa
d
H
. Saha
r
,
Dep
a
rtm
ent o
f
Ele
ct
rical
En
gi
neer
i
ng,
The U
niv
er
sit
y
of Tec
hnology
,
25 Sina
’a
h
stre
et
, Bag
hd
a
d 1
0120,
Iraq
.
Em
a
il
:
asaad.ha87@
gm
ail.co
m
1.
INTROD
U
CTION
Fo
r
a
co
uple
of
ye
ars
now,
c
haos
has
at
trac
te
d
a
great
dea
l
of
at
te
ntio
n
f
ro
m
var
io
us
sc
ho
la
rs
li
ke
eng
i
neer
s
,
m
ath
em
atici
ans,
a
nd
physi
ci
ans
[1
-
3].
The
stu
dy
tre
nd
ha
s
be
en
tra
ns
m
it
t
ing
from
searchi
ng
f
or
the
pro
ofs
of
t
he
e
xistence
of
cha
os
i
nto
s
ol
ic
it
ation
s
a
nd
thoro
ugh
hypo
theti
cal
researc
h
in
past
ye
ars
[4
]
.
Chaos
a
rr
a
nge
m
ents
go
t
fro
m
a
par
ti
cular
cat
egory
of
va
rio
us
eq
uatio
ns
are
not
spora
dic
an
d
subtl
e
to
first
ci
rcu
m
sta
nces,
and
it
is
har
d
to
forecast
their
i
m
pen
ding
ch
aract
ers
f
ro
m
pr
evi
ou
s
a
nnota
ti
on
s
[
5].
Be
cause
it
is
turn
i
ng
out
that
this
cha
otic
syst
e
m
can
be
easi
ly
execu
te
d
,
m
os
t
scho
l
ars
in
the
syst
em
s
and
ci
rc
uit
of
t
he
nonlinea
r
fiel
d
ha
ve
m
ajo
rly
been
fo
c
use
d
on
creati
on
exec
ution
c
on
ce
r
ning
cha
os
.
Cha
os
syst
e
m
s
co
m
m
un
ic
at
ion
s
a
re
am
on
g
the
at
te
ntio
n
-
gr
a
bbin
g
iss
ue
s
in
e
nginee
ring
fiel
d
[
6
-
8].
Most
sc
hola
r
s
ha
ve
con
ce
ntrate
d
on
the
de
sig
ning
of
non
-
c
oh
e
r
ent
recog
niti
ons
that
do
no
t
r
equ
i
re
the
us
a
ge
of
pri
m
ary
sign
al
s
(unm
od
ulate
d
carriers
)
at
a
r
ecei
ver
f
or
de
m
od
ul
at
ion
.
I
n
no
rm
al
syst
e
m
co
m
m
un
ic
ation
,
cl
assifi
e
d
unde
r
coh
e
re
nt
rec
ogniti
on
,
pri
m
ary
signa
ls
re
quir
e
to
be
gen
e
rat
ed
s
o
th
at
as
t
hey
ar
rive
at
t
he
receiver
,
t
he
y
are
dem
od
ulate
d.
Ther
e
f
or
e,
the
no
rm
al
syst
e
m
s
of
com
m
u
nicat
ion
are
c
halle
ng
i
ng
when
the
non
-
co
her
e
nt
detect
ion
is
ap
plied.
O
n
the
c
on
t
rar
y,
the
det
ect
ion
ap
plyi
ng
cha
os
non
-
c
ohere
ntly
cou
ld
dem
od
ulate
the
data
with
the
abse
nc
e
of
pri
m
ary
sign
al
s
beca
use
chao
s
an
d
c
hao
ti
c
seq
ue
nc
es
po
sse
ss
disti
nct
char
act
eri
sti
cs.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Eff
ic
ie
nt erro
r
correcti
ng sc
he
me
fo
r c
haos
sh
if
t key
ing si
gnals
(Hik
m
at N
. A
bdulla
h)
3551
Th
us
, th
e no
n
-
coh
e
re
nt r
eco
gnit
i
on
is taken a
s a u
nique r
ec
ogniti
on
a
ppr
oa
ch
by m
eans
of
c
hao
s
. Th
e
op
ti
m
al
receiver
[
9]
as
well
as
DCS
K
(D
if
fere
ntial
cha
os
sh
ift
ke
yi
ng
)
[
10]
ar
e
fam
ou
s
f
or
i
ts
syst
e
m
s
of
cl
assic
non
-
co
her
e
ncy
.
Howe
ver,
the
op
ti
m
al
receiv
er
suffe
rs
fro
m
co
m
plica
te
d
cal
c
ulati
on
s
a
nd
diff
ic
ult
sign
al
detect
io
n
wh
e
n
the
le
ngth
of
c
hao
ti
c
sequ
e
nce
N
bec
om
es
lon
g.
A
rai
et
a
l.
[1
1]
pro
po
s
ed
an
a
ppr
oach
of
rec
ogni
zi
ng
sy
m
bo
ls by the
co
m
pu
ti
ng
fig
ur
es
of the m
ini
m
al
le
ng
th
fro
m
sign
al
s r
ecei
ved
t
o
cha
otic
m
ap
i.e., s
ubop
tim
a
l
re
cei
ver.
I
n
place
of
valuati
ng
the
PD
F,
the
s
uboptim
al
rece
iver
est
i
m
at
es
the
PD
Fs
th
rough
determ
inin
g
the
near
est
le
ngt
h
from
the
chao
t
ic
m
ap
to
signa
ls
that
wer
e
r
ecei
ved
.
A
par
t
from
the
app
li
cat
ion
of
cha
os
in
m
od
ulati
on
syst
e
m
s,
chao
s
pointed
ou
t
that
,
a
nu
m
ber
of
sch
olars
w
ho
hav
e
a
dv
a
nce
d
their
app
li
cat
ion
in
channel
co
ding
[12].
Ma
j
orl
y,
the
represe
ntati
on
al
dynam
ic
connecte
d
with
the
cha
otic
m
aps
are
de
vis
ed
as
a
crit
erion of e
rror co
rr
ect
in
g.
The n
on
-
re
dundant
or r
e
dund
ant syst
em
b
ased
on ch
a
otic c
h
an
nel c
od
i
ng.
The
basic
idea
of
any
erro
r
correct
ing
m
et
ho
d
is
re
dundan
cy
bits,
wh
ic
h
is
extra
bits
add
ed
to
th
e
data.
The
re
du
nd
a
ncy
bits
use
d
f
or
er
ror
de
te
ct
ion
an
d
correct
ion
th
at
m
ay
occu
r
on
the
data
trans
m
itted,
in
the
pr
ocess
of
tra
ns
m
issio
n
or
st
or
a
ge.
The
re
dunda
nc
y
bits
af
fect
t
he
data
rate
of
the
c
omm
un
ic
at
ion
syst
e
m
wh
ere
the
transm
issio
n
rate
co
nver
sel
y
pr
oport
io
nal
with
the
num
ber
of
re
dund
a
nt
bits
[13,
14
]
.
The
non
-
re
dundant
a
ppr
oac
h
is
the
finest
wh
e
n
t
he
tra
nsm
issi
on
rate
is
co
ns
ide
re
d
w
her
eas
the
re
dunda
nt
m
et
ho
ds
hav
e
go
t a
dv
a
nce
d
pe
rfor
m
ance o
f B
ER. Howe
ver, the prev
i
ou
s
works u
se
non
-
redu
nd
a
nt corr
ect
io
n
dep
e
nd
on
the
su
bo
ptim
a
l
receiver,
but
the
y
hav
e
a
lot
of
delay
process
because
the
c
orrecti
on
of
ea
ch
bit
dep
e
nds
on
al
l
sequ
e
nce
.
I
n
our
pre
vious
work
[
15
]
,
we
desig
ne
d
a
sub
op
ti
m
al
detect
i
on
with
the
m
od
i
fied
log
ist
ic
m
ap.
I
n
this
st
ud
y,
th
e
desig
n
of
sub
op
ti
m
al
receiver
to
reali
ze
a
com
bin
ed
c
haos
base
d
nonc
ohere
nt
m
od
ulati
on
a
nd
non
-
re
dunda
nt
er
ror
c
orrect
ing
c
odin
g
is
pro
posed
.
T
he
de
sign
e
d
syst
em
us
es
tw
o
s
ucc
essive
chao
ti
c
seq
ue
nc
es
base
d
on
the
lo
gisti
c
m
a
p
su
c
h
that
the
string
sta
rte
d
from
the
second
val
ue
to
the
end
of
the
first
seq
ue
nce
is
us
e
d
as
to
represent
t
he
strin
g
sta
rt
ed
from
the
first
value
of
th
e
nex
t
seq
ue
nc
e
if
it
represe
nts
the
sam
e
bit
value.
This
feat
ur
e
giv
es
the
recei
ver
a
dd
it
io
nal
inf
or
m
at
ion
not
only
fo
r
c
orrect
recovery
but
al
so
for
c
orrect
c
orrecti
on.
T
he
pro
po
se
d
non
-
r
edun
dan
t
e
rro
r
co
rr
ect
io
n
depends
on
norm
al
iz
ed
correla
ti
on
wit
h
m
ini
m
u
m
delay
becau
se
th
e
m
axi
m
u
m
delay
fo
r
c
orrecti
ng
one
bit
depends
on
previ
ous
a
nd
nex
t
seq
ue
nces
only
.
2.
THE
SYSTE
M
O
VE
RV
IE
W WITH
P
R
OPOSE
D
E
R
ROR
CORRE
CTING
ALG
ORI
TH
M
The
blo
c
k
diagr
am
of
the
c
hao
ti
c
s
hift
ke
yi
ng
syst
em
with
the
s
ubopti
m
al
receiver
i
s
show
n
i
n
Fig
ure
1. The
detai
ls of eac
h bloc
k
a
re
descri
bed
i
n
the
next
secti
on
s.
C
ha
ot
i
c
s
i
gn
a
l
ge
ne
r
a
t
or
f
or
‘
1
’
C
ha
ot
i
c
s
i
gn
a
l
ge
ne
r
a
t
or
f
or
‘
0
’
D
a
t
a
0
1
1
0
11
00
A
W
G
N
S
u
bopt
i
m
a
l
D
e
t
e
c
t
or
E
r
r
o
r
C
o
r
r
e
c
t
i
o
n
B
a
s
ed
o
n
C
h
a
o
t
i
c
D
y
n
a
m
i
c
s
X
0
,x
1
,x
2
,x
3
y
0
,y
1
,y
2
,y
3
M=
m
0
,m
1
,m
2
,m
3
,m
4
,m
5
,m
6
,m
7
R
=
r
0
,r
1
,r
2
,r
3
,r
4
,r
5
,r
6
,r
7
o
u
t
p
u
t
Fig
ure
1
.
T
he
s
uboptim
al
r
ecei
ver
of CS
K
sy
stem
w
it
h
er
ror
correcti
ng c
od
ing
3.
THE
TR
AN
S
MITTE
R
In
t
he
tra
ns
m
i
t
te
r
side,
t
he
m
essage
bits
are
m
od
ulate
d
by
gen
e
rati
ng
a
chao
ti
c
se
qu
e
nc
e
from
the
chao
ti
c
m
ap.
In
this
w
ork,
w
e
us
ed
a
m
od
ifie
d
lo
gisti
c
m
ap
[
15
]
w
hich
is
on
e
of
the
m
os
t
strai
gh
tfo
rw
a
r
d
chao
ti
c m
aps,
and it
desc
ribe
s
by E
q
uation
(
1)
.
+
1
=
−
(
2
2
−
0
.
25
)
(
−
1
≤
≤
1
)
(1)
wh
e
re
a
is
pos
it
ive
real
con
st
ant,
an
d
it
is
be
tween
0
<
a
≤
4
represe
nt
the
co
ntr
ol
pa
r
a
m
et
er
fo
r
t
his
m
ap
.
The
enc
odin
g
a
rch
it
ect
ure
of
CSK
shows
i
n
Fig
ur
e
2.
Wh
en
tra
ns
m
it
ted
K
bits
throu
gh
a
noisy
cha
nn
el
,
for
each
data
bit
N
sequ
e
nce
f
rom
identic
al
chao
ti
c
m
ap
gen
e
rates,
there
f
or
e
,
the
am
ou
nt
of
data
transm
itt
ed
tu
r
n
into
K
×N
.
For
each
sig
nal
bloc
k,
the
init
ia
l
value
is
rando
m
ly
sel
ect
ed
fo
r
sy
m
bo
l
"1"
(x
0
)
an
d
"0"
(y
0
)
a
t
the
beg
i
nn
i
ng.
Af
t
erw
a
rd,
the
in
it
ia
l
value
for
the
nex
t
sy
m
bo
l
"1"
an
d
"
0"
seq
uen
ce
s
will
be
ta
ken
from
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
5
,
Oct
ober
201
9
:
3
5
5
0
-
3
5
5
7
3552
the
seco
nd
val
ue
of
the
pre
vi
ou
s
se
quence
.
Fo
r
e
xam
ple,
assum
e
N
=
4
and
K
=
5,
and
t
he
data
bi
ts
are
01101. T
he
m
od
ulate
d si
gnal
vecto
r
S
which
i
s g
i
ven
as
fo
ll
ow
s:
S= (
S
0
, S
1
, S
2
, S
3
, S
4
)
=
(
y
0
, y
1
, y
2
, y
3
, x
0
, x
1
,
x
2
, x
3
,
x
4
, x
5
, x
6
, x
7,
y
4
, y
5
, y
6
,
y
7
, x
8
,
x
9
, x
10
, x
11
)
=
(
s
0
, s
1
,
s
2
, …
…, s
19
)
(2)
wh
e
re
y
1
,
the
s
econd
val
ue
of
the
sequ
e
nce
for
the
first
sy
m
bo
l
"0",
has
the
sam
e
value
of
y
4
,
the
fir
st
value
for
the
seq
ue
nc
e
fo
r
t
he
ne
xt
"0",
an
d
sim
i
la
rly
fo
r
the
f
ollow
i
ng
value
s
ti
ll
the
end
of
the
c
orres
pondin
g
sy
m
bo
l.
i.e.
(y
1
,y
2
,y
3
)
=(y
4
,y
5
,y
6
).
Fr
om
ano
t
he
r
ha
nd,
x
1
,
t
he
seco
nd
val
ue
of
t
he
se
quenc
e
for
t
he
fi
rst
s
ym
bo
l
"1",
has
the
s
a
m
e
value
of
x
4
,
the
first
va
lue
f
or
t
he
s
equ
e
nce
for
th
e
ne
xt
"1",
i.e
.
(x
1
,x
2
,x
3
)
=(x
4
,x
5
,x
6
).
Th
at
al
ways
th
e
ne
xt
se
qu
e
nc
e
is
ide
ntica
l
to
the
pre
vious
seq
uen
ce
in
(N
-
1)
values
.
T
hi
s
al
gorithm
give
s
the
receiver
ad
diti
on
featu
res
for
detect
ion
a
nd
correct
io
n
as
will
be
sho
wn
la
te
r.
Wh
e
n
the
sig
nal
tra
nsm
itted
thr
ough
noisy
channels
with
a
m
ean
of
zero
a
nd
a
var
ia
nce
σ
2
t
he
bl
ock
sig
na
l
beco
m
e
equ
al
to
R = S +
nois
e
= R = [
r
0,
r1,
r
2,
…. r
1
9
].
C
h
a
o
t
i
c
S
i
g
n
a
l
G
e
ne
r
a
t
or
“
0
”
C
h
a
o
t
i
c
S
i
g
n
a
l
G
e
n
e
r
a
t
o
r
“
1
”
101
01
I
n
p
u
t
d
a
t
a
M
=
(
M
0
,
M
1
,
M
2
,
M
3
,
M
4
)
=
(
y
0
,
y
1
,
y
2
,
y
3
,
x
0
,
x
1
,
x
2
,
x
3
,
x
4
,
x
5
,
x
6
,
x
7
,y
4
,
y
5
,
y
6
,
y
7
x
8
,
x
9
,
x
10
,
x
11
)
=
(
m
0
,
m
1
,
m
2
,
m
3
,
…
.
.
,
m
19
)
Fig
ure
2
.
The
pro
po
se
d
C
SK
encode
r
f
or er
r
or co
rr
ect
io
n
4.
NONC
OHE
R
ENT RE
CEI
VER WIT
H
P
ROP
OSED
E
RROR
COR
R
ECTION
The
transm
it
ted
sign
al
bl
ocks
are
recovere
d
by
the
receiver
f
ro
m
the
received
sig
nal
blo
cks
wh
il
e
the info
rm
ation
sym
bo
ls dem
odulate
d
it
. A
ga
in,
the
er
ror
c
orrecti
on is p
e
r
form
ed
by the
receiver
. Beca
us
e
we
took
the
nonc
oh
e
re
nt
receiv
er
into
co
ns
id
erati
on,
the
chao
ti
c
m
ap
util
iz
ed
at
the
transm
itter
fo
r
the
m
od
ulati
on
is
m
e
m
or
iz
ed
by
the
receiver
.
Nonetheless
,
the
init
ia
l
value
of
cha
os
within
the
tra
ns
m
it
te
r
is
nev
e
r
re
vealed
to
t
he
receive
r
.
T
he
e
rror
-
c
orrecti
ng
m
et
ho
d
w
hich
we
pr
opose
d
is
m
ade
up
of
t
he
s
uboptim
al
detect
or
a
nd
th
e
error
c
orrecti
on
on
the
basi
s
of
c
ha
otic
dy
nam
ic
s.
The
pro
po
se
d
detect
ion
/e
rro
r
co
rrec
ti
on
m
et
ho
d
bl
ock
diag
ram
wh
en
N=K
=4
is
il
lustrate
d
in
Fig
ure
3.
T
he
nonc
oh
e
re
nt
detect
ion
f
or
eve
ry
receive
d
blo
c
k
is,
first,
perform
ed
by the r
ec
ei
ver an
d every
sym
bo
l dem
od
ulate
d.
We
had our
subopti
m
a
l nonc
oh
e
re
nt
detect
ion
al
gorithm
,
wh
ic
h
we
intr
oduce
d
in
[
15]
,
ap
plied
in
t
his
w
or
k.
T
he
e
rror
-
c
orrecti
ng
sch
e
m
e
is
perform
ed
by
the
receiver
after
each
sym
bo
l
is
de
m
od
ulate
d.
a
des
cripti
on
of
our
subopti
m
a
l
detect
or
op
e
rati
on
will
b
e m
ade av
ai
la
ble prio
r
to
g
i
vi
ng
a
n
e
xpla
nat
ion
of the
prop
os
e
d
er
r
or
c
orr
ect
ion
op
e
rati
on.
P
r
o
p
o
s
e
d
E
r
r
o
r
C
o
r
r
e
c
t
i
o
n
B
a
s
e
d
o
n
C
h
a
o
t
i
c
D
y
n
a
m
i
c
s
S
u
b
o
p
t
i
m
a
l
D
e
t
e
c
t
o
r
(
D
e
t
e
c
t
i
o
n
o
f
s
y
m
b
o
l
f
o
r
e
a
c
h
b
l
o
c
k
)
(r
0
,r
1
,r
2
,r
3
)
(r
4
,r
5
,r
6
,r
7
)
(r
8
,r
9
,r
10
,r
11
)
(r
12
,r
13
,r
14
,r
15
)
d
0
d
1
d
2
d
3
C
1
C
2
C
3
C
4
Fig
ure
3. Bl
oc
k diag
ram
o
f
th
e pro
posed
d
et
ect
ion
/e
r
ror
c
orrecti
on m
et
hod wh
e
n N=
K=
4
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Eff
ic
ie
nt erro
r
correcti
ng sc
he
me
fo
r c
haos
sh
if
t key
ing si
gnals
(Hik
m
at N
. A
bdulla
h)
3553
5.
THE
SU
B
OP
TIMAL
DET
ECTOR
The
detect
io
n
of
sym
bo
ls
is
accom
plished
thr
ough
the
co
m
pu
ta
ti
on
of
the
shortest
dist
ance
existi
ng
betwee
n
the
r
e
cei
ved
si
gn
al
a
nd
t
he
cha
otic
m
ap.
The
us
e
of
a
nonlinea
r
m
ap
in
this
w
ork
is
f
or
deter
m
ining
the
cl
os
e
st
m
a
p
to
the
recei
ve
d
po
i
nt
R.
Th
e
com
pu
ta
ti
on
of
this
distance
that
e
xists
bet
ween
the
point
R
an
d
the
tw
o
m
aps
i
s
pe
rfor
m
ed
by
obta
inin
g
t
he
nonlinea
r
m
ap
ta
ng
e
nt
e
qu
at
i
on
a
nd
un
der
ta
king
c
om
pu
ta
ti
on
f
or
the
m
ini
m
u
m
distance
f
ro
m
receive
d
point
R=
(
ri
,
ri+1)
wh
e
re
i
=
1,
2,
3,
4…,
N
–
1
to
the
point
of
the
ta
ng
e
nt.
T
he
r
ecei
ved
point
and
the
distan
ce
to
the
m
od
i
fied
lo
gisti
c
m
ap
f
unct
io
n
ta
ng
e
nt
ar
e
il
lustrate
d
i
n
Fig
ure
4.
Fig
ure
4
.
The
c
al
culat
ion
of m
ini
m
u
m
d
ist
an
ce usi
ng the ta
ng
e
nt
of the
no
nlinear
m
ap
The
s
hortest
di
sta
nce,
base
d
on
Fig
ure
4,
f
ro
m
the
recei
ve
d
po
i
nt
to
t
he
sym
bo
l’s
tw
o
functi
ons.
Eq
uation (
3) is us
e
d
i
n
c
om
pu
ti
ng
for
"
1"
1
=
√
(
−
1
)
2
+
(
(
)
−
2
)
2
(
3)
wh
e
re t
he nonl
inear m
ap
f
unc
ti
on
sym
bo
li
zed
by f(x) is
give
n by
(
)
=
−
(
2
2
−
0
.
25
)
(
4)
wh
e
re a=
4. Re
m
ov
al
o
f
t
he
e
qu
at
io
n’s s
quar
e r
oo
t
(3) gives
1
2
=
(
−
1
)
2
+
(
(
)
−
2
)
2
(
5)
Ob
ta
ini
ng the
distance y
ie
ld
(
1
)
2
=
16
3
+
(
8
2
−
6
)
−
2
1
(
6)
Find
i
ng the e
quat
ion’s
r
oo
ts
(
6) cal
ls f
or
it
t
o be e
qu
at
e
d
t
o ze
r
o
16
3
+
(
8
2
−
6
)
−
2
1
=
0
(
7)
Now,
try
in
g
to su
bst
it
ute
x=
(
x1,
x2,
x3
)
in
equ
at
io
n
(
1)
as a
m
et
ho
d
of
fi
nd
i
ng
(
y1,
y2
,
y3)
an
d
the
n
fin
ding
the
m
i
nim
u
m
dist
ance
for
"
1".
Sim
i
la
r
ste
ps
a
re
use
d
in
fin
ding
the
m
ini
m
u
m
distance
for
"
0"
.
The
cum
ulati
ve
dis
ta
nce
f
or
"
1"
(
∑
1
)
a
nd
"
0"
(
∑
0
)
is
com
pu
te
d
by
the
s
ubopti
m
a
l
receive
r
for
al
l
the
bits
seq
uen
ce
.
T
he
detect
or
m
akes
the
decisi
on
on
wh
ic
h
bi
t
is
"0"
or
"
1"
base
d
on
t
he
s
hortest
di
sta
nce
com
pu
te
d,
i
f
∑
0
˃
∑
1
, th
e
decodin
g o
f
the
sig
nal is
pe
rfor
m
ed
as "
1"
, and
i
f no
t i
t,
is dec
od
e
d
as
"
0".
6.
THE
PROPO
SED E
RROR
-
CORRE
CTING MET
H
OD
Af
te
r
the
dem
odulati
on
of
e
ach
sym
bo
l,
the
recei
ver
pe
rfor
m
s
the
error
-
co
rr
ect
in
g
m
et
ho
d.
T
he
basic
idea
of
th
e
schem
e
is
that
each
detect
ed
bit
is
par
ti
al
l
y
correla
te
d
(
N
-
1
cha
otic
sa
m
ples
ou
t
of
N
c
ha
otic
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
5
,
Oct
ober
201
9
:
3
5
5
0
-
3
5
5
7
3554
sam
ples
that
rep
rese
nt
eac
h
bit)
two
ti
m
es.
On
e
ti
m
e
with
its
procee
de
d
a
nd n
ext
ze
r
os
a
nd o
the
r
ti
m
e
wi
th
it
s
procee
ded
a
nd
nex
t
on
es
.
Ac
cordin
g
to
t
he
correla
ti
on
re
su
lt
s,
the
c
orre
ct
ion
is
done
dep
e
ndin
g
on
wh
ic
h
resu
lt
is
m
axim
u
m
.
If
the
co
rr
el
at
ion
res
ult
of
on
es
is
great
er
than
ze
ro
s
it
is
cor
rected
to
be
on
e
,
ot
herwise
,
it
is
correct
ed
to
be
zer
o.
For
ease
of
e
xpla
na
ti
on
,
c
onsider
the
f
ollow
i
ng
exam
ple.
Let
N=4,
k
=
5
,
da
ta
=
[01
101],
a
nd
a
n
e
rror
occurs
at
R
2,
then
t
he
flo
w
of
the
pro
po
s
ed
er
ror
c
or
recti
on
m
et
ho
d
will
be
as
sho
wn
i
n
Figure
5.
Af
te
r
the
detect
ion
,
the
rec
ei
ve
r
arr
a
ng
e
s
the
re
cei
ved
sam
ples
accor
ding
to
the
decode
d
sym
bo
ls
and
if
a
n
er
ror
occurs
w
he
n
the
receive
r
det
ect
s
sy
m
bo
l
it
us
e
s
the
fe
at
ure
of
eac
h
su
cc
essive
seq
ue
nc
es
is
identic
al
w
it
h
(
N
-
1)
an
d use
norm
al
iz
ed
corr
el
at
ion
s
how
n
i
n
e
qu
at
io
n (
8)
.
n
orm
al
iz
ed
correla
ti
on
=
∑
(
)
̂
(
−
1
)
=
1
√
∑
2
(
)
∑
̂
2
(
)
−
1
=
0
=
1
(8)
In
Fig
ur
e
5,
a
n
error
occ
ur
s
at
the
thir
d
bit.
T
he
recei
ver
c
om
pu
te
s
the
nor
m
al
iz
ed
cor
rel
at
ion
f
or
(
0)
betwee
n
the
thi
rd
s
eq
ue
nce
an
d
pr
e
vious
(0)
and
in
a
dd
it
io
n
to
no
rm
alized
correla
ti
on
f
or
the
ne
xt
(
0)
an
d
th
e
sam
e
ste
ps
us
ed
f
or
(
1)
(
by
usi
ng
pr
e
vious
(
1)
an
d
ne
xt
(
1)).
If
t
he
c
orrelat
ion
f
or
1
is
gr
ea
te
r
than
0,
it
m
eans
that t
he
t
hird b
i
t i
s 1
i
ns
te
ad o
f
0
.
T
he
recei
ve
r
us
es
t
he
al
gor
it
h
m
o
f
c
orrect
ion
f
or
eac
h
bit
on the
sym
bo
l
.
R
e
c
e
i
v
e
d
s
i
g
n
a
l
s
=
[
R
0
,
R
1
,
R
2
,
R
3
,
R
4
]
R
e
c
e
i
v
e
d
s
i
g
n
a
l
s
=
[
(
r
00
,r
01
,r
02
,r
03
)
,
(
r
10
,r
11
,r
12
,r
13
)
,
(
r
20
,r
21
,r
22
,r
23
)
,
(
r
30
,r
31
,r
32
,r
33
)
,
(
r
40
,r
41
,r
42
,r
43
)]
0
1
0
0
1
R
e
c
e
i
v
e
d
s
i
g
n
a
l
f
o
r
1
=
[
(
r
10
,r
11
,r
12
,r
13
)
,
(
r
20
,r
21
,r
22
,r
23
)
,
(
r
40
,r
41
,r
42
,r
43
)]
1
0
1
N
o
r
m
a
l
i
z
e
d
c
o
r
r
e
l
a
t
i
o
n
1
=
n
o
r
m
c
o
r
r
.
[
(
r
10
,r
11
,r
12
,r
13
)
,
(
r
20
,r
21
,r
22
,r
23
)]
+
n
o
r
m
c
o
r
r
.
[
r
20
,r
21
,r
22
,r
23
)
,
(
r
40
,r
41
,r
42
,r
43
)]
R
e
c
e
i
v
e
d
s
i
g
n
a
l
f
o
r
0
=
[
(
r
00
,r
01
,r
02
,r
03
)
,
(
r
20
,r
21
,r
22
,r
23
)
,
(
r
30
,r
31
,r
32
,r
33
)]
0
0
0
N
o
r
m
a
l
i
z
e
d
c
o
r
r
e
l
a
t
i
o
n
0
=
n
o
r
m
c
o
r
r
.
[
(
r
00
,r
01
,r
02
,r
03
)
,
(
r
20
,r
21
,r
22
,r
23
)
]
+
n
o
r
m
c
o
r
r
.
[
r
20
,r
21
,r
22
,r
23
)
,
(
r
30
,r
31
,r
32
,r
33
)]
A
s
N
o
r
m
a
l
i
z
e
d
c
o
r
r
e
l
a
t
i
o
n
1
˃
N
o
r
m
a
l
i
z
e
d
c
o
r
r
e
l
a
t
i
o
n
0
⸫
t
h
e
d
e
c
o
d
e
d
s
i
g
n
a
l
=
(
01
1
01
)
C
or
r
e
l
at
i
on
1
C
or
r
e
l
at
i
on
0
Fig
ure
5
.
The
pro
po
se
d
e
rro
r c
orrecti
on m
eth
od
7.
SIMULATI
O
N RESULTS
The
perform
a
nce
of
t
he
propose
d
e
rror
-
correct
ing
m
e
thod
has
bee
n
asses
sed
wi
th
com
pu
te
r
si
m
ulati
on
s
thr
ough
MATL
A
B.
The
sim
ulatio
n
par
am
et
ers
are
as
fo
ll
ow
s
:
within
the
tr
ansm
itti
ng
sid
e,
we
su
pp
os
e
K
=3
2.
T
he
c
on
t
ro
l
par
am
et
er
is
set
as
a
=4
f
or
the
m
od
if
ie
d
log
ist
ic
m
ap.
The
cha
otic
seq
uen
c
e
’
s
le
ng
th
pe
r
1
bi
t
is
sel
ect
ed
to
ass
um
e
the
values
N=
4,
5,
6,
7
an
d
8.
Figure
6
il
lust
rates
the
BER
ve
rsus
Eb/N
0
wit
h
N
as
the
pa
ram
eter
f
or
t
he
sub
optim
al
receiver
without
er
ror
-
correct
ion.
Fig
ur
e
6
il
lustrate
s
tha
t
at
BER
=
10
-
3
,
4.2
dB
gain
i
n
Eb/N
0
is
acq
ui
red
on
ce
N
=
8
ove
r
N
=
4.
This
en
ha
ncem
ent
is
ref
e
rr
e
d
to
the
raise
of
sprea
di
ng
facto
r
of
t
he
cha
otic
sig
na
l.
Nev
e
rtheles
s,
the
pen
al
ty
f
or
t
his
en
ha
nc
e
m
ent
is
the
ra
ise
in
the intrica
cy
of the s
ubopti
m
a
l det
ect
ion sy
stem
r
egardin
g
t
he
am
ou
nt
of
m
at
he
m
at
ic
a
l op
erati
ons ca
rr
i
ed ou
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Eff
ic
ie
nt erro
r
correcti
ng sc
he
me
fo
r c
haos
sh
if
t key
ing si
gnals
(Hik
m
at N
. A
bdulla
h)
3555
Figure
6
.
BER
perform
ance f
or
a s
ubopti
m
al
receiver
for di
f
fer
e
nt v
al
ue of
N
Figure
7
il
lustrate
s
the
BER
perform
ance
ver
s
us
E
b/N0
for
the
pro
po
s
ed
te
ch
nique
a
s
con
t
rasted
with
the
sub
optim
al
receiver
pri
or
to
the
error
-
co
rr
ect
i
ng
wh
e
n
N
=
4,
6,
an
d
8
r
e
sp
ect
ively
.
It
m
ay
be
ob
s
er
ved
f
ro
m
this
fig
ur
e
th
at
the
e
nh
a
nc
e
m
ent
within
BER
is
enla
rg
ed
as
N
inc
re
ased.
F
or
exa
m
ple,
at
BER
=
10
-
3
,
the
at
ta
ined
ga
ins
withi
n
E
b/
N0
for
N
=
4,
6,
a
nd
8
a
re
0.6,
1,
an
d
1.3
dB
res
pecti
vel
y.
Eve
n
though
t
he
gai
n
values
of
t
he
co
ding
sc
hem
e
are
not
hi
gh,
it
s
perf
or
m
ance
is
deem
ed
e
xtrem
el
y
go
od
giv
e
n
that i
t ge
ner
at
e
s erro
r
c
orrecti
on w
it
ho
ut the necessi
ty
for
re
dundant
bits.
(a)
(b)
(c)
Figure
7
.
BER
perform
ance bef
ore a
nd afte
r e
rror co
rr
ect
io
n
(a)
N=
4,
(
b)
N=6, (c
) N=
8
Figure
8
s
how
s
the
BER
pe
r
form
ance
of
th
e
pro
posed
sys
tem
a
s
com
par
ed
with
th
e
sy
stem
wo
r
k
in
[12]
that
us
es
li
near
chao
t
ic
te
nt
m
ap
when
N=
4
for
both
cases.
From
this
figure,
it
m
ay
be
ob
ser
ve
d
that
the
pro
pose
d
m
et
ho
d
has
be
tt
er
perform
ance
in
noisy
ch
ann
el
.
At
BE
R
=
10
-
3
,
2.4
5
dB
gai
n
in
E
b
/N
0
is
ob
ta
ine
d
us
in
g
the
pro
posed
m
e
tho
d
ov
e
r
th
e
pr
e
vious
m
eth
od.
The
im
pr
ov
em
ent
introd
uced
by
the
propos
e
d
m
et
ho
d
is
ref
e
r
red
to
the
nat
ure
of
the
traject
or
y
of
cha
otic
m
ap
and
the
spreadi
ng
fact
or
o
f
sig
nal.
Th
e
us
e
of
chao
ti
c
m
ap
with
values
a
lt
ern
at
ing
bet
ween
posit
ive
an
d
ne
gati
ve
val
ues
act
s
to
re
duce
BE
R
in
no
isy
c
ha
nn
el
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
5
,
Oct
ober
201
9
:
3
5
5
0
-
3
5
5
7
3556
On
e
of
the
rest
rict
ion
as
pects
to
the
c
ha
o
ti
c
cod
i
ng
sc
hem
e
is
the
raise
of
com
pu
ta
ti
on
al
com
plexity
reg
a
rd
i
ng
the
a
m
ou
nt
of
m
ultip
li
cat
ion
s r
e
quired.
O
ur p
r
op
os
e
d
e
rror
co
rrec
ti
ng
te
c
hn
i
que
e
nhances
th
e
delay
par
am
et
er
as
con
t
rasted
with
te
ch
nique
in
[12]
wh
ic
h
a
s
well
e
m
plo
ys
subopti
m
a
l
t
echn
i
qu
e
.
Fig
ur
e
9
il
lustrate
s
the
c
om
pu
ta
ti
on
al
c
om
plexity
of
pro
posed
te
ch
niq
ue
by
m
eans
of
norm
al
iz
ed
correla
ti
on
ab
out
the
a
m
ou
nt
of
m
ulti
plica
ti
on
s
as
con
t
rasted
with
the
sub
op
ti
m
al
te
chn
iq
ue
.
It
m
a
y
be
vie
we
d
in
this
fig
ur
e
that
a
consi
der
a
ble
re
du
ct
io
n
i
n
c
ompu
ta
ti
ons
is
at
t
ai
ned
us
i
ng
t
he
pro
posed
te
ch
nique
a
nd
this
reducti
on
rises
as
N
increases
.
This
le
ssening
is
r
efer
red
t
o
cal
culat
ing
the
c
or
relat
ion
s
li
nke
d
to
the
ea
rlie
r
an
d
ne
xt
bits
on
ly
rather
t
han
al
l
bits
stream
of
as
in
the
co
nventio
nal
te
ch
niq
ue.
At
N=8,
the
num
ber
of
r
ed
uctio
n
acqu
i
red
is 6
8%.
Figure
8
.
BER
perform
ance f
or
propose
d
m
eth
od a
nd
m
et
ho
d
i
n [12] w
hen N
=
4
Figure
9
.
Com
pu
ta
ti
onal
c
omplexit
y o
f
c
haot
ic
cod
i
ng sc
hem
e
w
it
h N a
s a
pa
ram
et
er
for
error co
rr
ect
io
n
m
et
ho
d
8.
CONCL
US
I
O
N
Chaotic
dynam
ic
s
m
a
y
be
us
e
d
as
ext
ra
in
f
orm
at
ion
to
a
ppr
opriat
el
y
recov
er
the
co
nvey
e
d
data.
T
he
error
-
co
rr
ect
i
ng
syst
e
m
do
es
no
t
nee
d
any
redunda
nt
bit
sequ
e
nce
gi
ve
n
that
the
error
c
orrecti
on
us
es
the
chao
ti
c
dynam
ic
s
inh
e
ren
t
i
n
the
co
nvey
ed
s
ign
al
blo
c
ks
.
T
he
sc
hem
e
pr
ovides
en
ha
nce
m
ent
in
E
b/N0
over
tradit
ion
al
c
ha
otic
sh
ift
keyi
ng
sc
hem
e
and
this
en
ha
ncem
e
nt
is
au
gm
ented
as
th
e
am
ou
nt
of
the
s
eq
ue
nce
is
enlar
ged.
The
pro
po
se
d
er
ror
cor
recti
on
re
duces
the
delay
in
op
e
rati
on
a
s
con
tra
ste
d
w
i
th
oth
er
te
ch
ni
qu
es
since
the
co
rr
e
ct
ion
delay
reli
es
j
ust
on
c
ompu
ta
ti
ons
li
nke
d
to
the
ea
rlie
r
and
nex
t
bits.
The
ca
pab
il
it
y
of
the
syst
e
m
to
co
rrec
t
errors
without
redu
nd
a
nc
y
m
a
y
be
de
e
m
ed
as
t
he
r
adical
sect
io
n
of
th
e
perfor
m
ance
enh
a
ncem
ent.
ACKN
OWLE
DGME
NTS
The
aut
hors
w
ou
l
d
li
ke
to
th
ank
t
he
hea
d
a
nd
t
he
sta
ff
of
com
m
un
ic
at
ion
en
gin
ee
rin
g
la
borato
ry
of
el
ect
rical
eng
i
neer
i
ng
de
par
t
m
ent
at
the
university
of
te
c
hnology
for
th
ei
r
sup
port
a
nd
e
nd
le
ss
co
op
erati
on
thr
oughout t
his
work.
REFERE
NCE
S
[1]
G.
Kolum
ba´
n,
et
al
.
,
“
Diffe
r
ent
i
al
ch
aos
shift
ke
y
ing
:
A
robust
c
oding
for
cha
os
comm
unic
at
ion,
”
i
n
Proc
ee
ding
s
of
NDES’96
int
e
rnational
conf
er
enc
e
,
pp
.
87
-
92
,
Jan
1996
.
[2]
S
.
Fadhel,
et
al
.
,
“
Chaos
Im
age
Enc
r
y
p
ti
on
Me
thods:
A
Surve
y
Stud
y
,
”
Bulleti
n
of
El
e
ct
rica
l
Engi
ne
ering
an
d
Informatic
s
,
v
o
l. 6, pp. 99
-
104
,
Mar
2017
.
[3]
F.
C.
M. Lau an
d
C.
K
.
Tse,
“
Chaos
-
Based
Digi
t
al
Com
m
unic
ati
on
S
y
stems
,
”
Springer, 2003.
[4]
E
.
R.
Arbol
eda,
et
a
l
.
,
“
Chaot
ic
Rive
st
-
Sham
ir
-
Adler
m
an
Algorit
hm
with
Data
Enc
r
y
pt
ion
Stan
dar
d
S
che
dul
ing
,
”
Bul
letin
of El
ec
t
rical
Engi
ne
erin
g
and
Informati
c
s
,
v
ol. 6, pp. 219
-
227
,
Sep
2017
.
[5]
A
.
D
.
W
owor
and
V
.
B
.
L
iwan
douw,
“
Dom
ai
n
Exa
m
ina
t
ion
of
Chaos
Logi
stic
s
Functi
on
As
A
Ke
y
Gen
erator
i
n
Cr
y
ptogr
aph
y
,
”
IJE
CE
,
v
ol
.
8
,
p
p.
4577
-
4583
,
D
ec
2018
.
[6]
P.
Stavroulakis,
“
Chaos
Applic
ations i
n
Te
l
ec
om
m
unic
at
ions,
”
T
al
or & Fran
ci
s G
roup,
2006
.
[7]
A
.
Sam
bas,
et
a
l
.
,
“
A
New
Chaot
ic
S
y
s
te
m
wit
h
a
Pear
-
Shap
ed
Equi
l
ibr
ium
an
d
Its
Circ
ui
t
Si
m
ula
ti
on,
”
IJ
EC
E
,
v
ol.
8
,
pp
.
4951
-
4958
,
De
c
2018
.
[8]
H
.
A.
Abdullah
and
H
.
N.
Abdulla
h,
“
A
New
Chaot
ic
Map
for
Secur
e
Tra
nsm
ission,
”
TEL
KOMNIKA
Tele
communic
a
t
ion
Computing
El
e
ct
ronics
and
Control
,
v
o
l.
16,
pp.
1135
-
1142
,
J
un
2018
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Eff
ic
ie
nt erro
r
correcti
ng sc
he
me
fo
r c
haos
sh
if
t key
ing si
gnals
(Hik
m
at N
. A
bdulla
h)
3557
[9]
M.
Hasle
r
and
T
.
Schimm
ing,
“
Chaos
com
m
unic
a
ti
on
ove
r
nois
y
c
hanne
ls,
”
In
t.
J.
Bi
furcation
and
Chaos
,
v
ol.
10,
pp.
719
-
736
,
Ap
r
2000
.
[10]
Z
.
Li
u
and
J
.
Zh
ang,
“
Design
of
the
diffe
r
ent
i
al
c
haos
shift
ke
y
in
g
comm
unic
at
io
n
sy
st
em
base
d
on
DS
P
buil
d
er,
”
Computer
Mode
l
li
ng
&
New
Technol
ogie
s
,
v
ol
.
18
,
pp
.
138
-
143
,
2
014
.
[11]
S.
Arai
and
Y.
Nishio,
“
Suboptim
al
rec
ei
v
er
using
shortest
dista
nce
appr
oximat
i
on
m
et
hod
for
c
haos
shift
ke
y
in
g,
”
Journal
of
Signa
l
Proc
essing
,
v
o
l
.
13
,
pp
.
161
-
16
9
,
2009
.
[12]
S.
Arai,
et
al
.
,
“
Err
or
-
cor
r
ec
t
ing
sche
m
e
base
d
on
cha
ot
ic
d
y
n
a
m
ic
s
and
it
s
per
form
anc
e
for
no
n
-
cohe
ren
t
ch
ao
s
comm
unic
at
ions
,
”
Journal
o
f
No
n
-
li
near
Theory
and
it
s
Appl
i
cat
i
ons,
IEI
CE
,
v
o
l.
1,
pp
.
196
-
206
,
2010
.
[13]
W
.
C
.
Huffm
an and
V
.
Pless,
“
Fundam
ent
al
s of
Err
or
-
Corre
c
ti
ng
Codes,
”
Cambri
dge
Univer
si
t
y
P
ress,
2003.
[14]
V.
G.
Jadh
ao
and
P.
D
.
Ga
wand,
“
Perform
anc
e
Anal
y
sis
of
Li
n
ea
r
Block
Code,
Convo
lut
ion
code
an
d
Conca
t
ena
t
ed co
de
to
Stud
y
The
i
r
Com
par
at
iv
e E
ffe
ctivene
ss
,
”
v
ol
.
1
,
pp
.
53
-
61
,
J
un
2012
.
[15]
H
.
N.
Abdulla
h
,
et
a
l
.
,
“
Suboptim
al
Dete
c
ti
on
o
f
Modifie
d
Log
i
stic
Map
Based
Chaos
Shift
Ke
y
ing
Modulat
ion
,
”
U.P
.
B
.
S
ci. Bull.
,
Seri
es
C
Journal
,
v
o
l. 80, 2018.
BIOGR
AP
H
I
ES
OF
A
UTH
ORS
Hi
kmat.
N.
Ab
du
ll
ah
was
born
in
Baghda
d
,
Ira
q
in
1974.
He
o
bta
in
ed
his
B.
Sc
.
in
Elec
tri
c
al
Engi
ne
eri
ng
in
1995,
M.Sc.
in
Comm
unic
at
io
n
Engi
nee
r
ing
i
n
1998
at
Univer
sit
y
of
Al
-
Mus
ta
nsir
y
ah
,
I
raq
and
Ph.D.
in
Comm
unic
at
ion
Eng
ine
er
i
ng
in
2004
at
Univer
sit
y
of
Te
chno
log
y
,
Ir
a
q.
From
1998
to
2015
he
worked
as
associate
profe
ss
or
in
the
Elec
tri
c
al
Engi
ne
eri
ng
De
par
tment,
a
t
Al
-
Mus
ta
nsir
y
ah
U
nive
rsit
y
,
Ir
aq.
Since
the
beg
in
ning
of
2015
he
works
as
full
profe
ss
or
in
col
lege
of
Inform
at
i
on
Engi
nee
r
ing
at
Al
-
Nahra
in
Univer
sit
y
,
Ira
q.
From
2011
–
201
3
he
got
a
r
ese
ar
ch
awa
rd
from
Inte
rna
ti
ona
l
Inst
it
ute
of
Edu
catio
n
(IIE
/USA
)
a
t
Bonn
-
Rhei
n
-
Sie
g
unive
rsi
t
y
of
appl
i
ed
sci
ences,
Germ
an
y
.
He
is
a
sen
ior
m
e
m
ber
of
IE
EE
associa
t
ion
since
2014.
He
is
i
nte
rest
ed
in
subject
s
of
wire
le
ss
com
m
unic
at
io
n
sy
st
ems
and
cha
ot
ic
comm
unic
a
ti
ons
.
Th
amir
Ras
hed
Saeed
was
Bor
n
in
Baghda
d,
Ir
aq
on
Februa
r
y
10,
1965.
He
recei
ved
th
e
B.
Sc
.
degr
ee
from
m
il
it
ar
y
engi
ne
eri
n
g
col
l
ege
in
Ba
ghdad
in
1987
,
the
M.Sc
.
degr
e
e
from
m
il
it
a
r
y
engi
ne
eri
ng
coll
ege
in
Baghd
a
d
in
1994
and
Ph.D.
d
egr
ee
from
AL
-
Rashed
co
ll
eg
e
of
engi
ne
eri
ng
and
Scie
nc
e
in
Ba
ghdad
2003.
From
1994
to
20
03,
he
worked
with
m
il
ita
r
y
engi
ne
eri
ng
coll
ege
in
Baghda
d
as
a
m
ember
of
te
a
chi
ng
staff
.
From
2003
ti
ll
now,
he
worked
with
the
Univer
sit
y
of
Te
chno
lo
g
y
in
B
aghda
d
as
a
m
ember
of
te
a
chi
n
g
staff
.
C
urre
ntly
,
he
is
As
sist.
Profess
or
of
el
e
ct
ri
ca
l
eng
ine
er
ing
at
un
ivers
ity
of
Technol
og
y
.
His
m
aj
or
int
ere
sts
ar
e
in
digi
tal
signal
pr
oce
ss
ing,
digital
ci
rcu
i
t
design
f
or
DS
P
base
d
on
FP
G
A,
Rada
r,
sen
sors
net
work
and
Pat
te
rn
Recogniti
on.
As
aad
Hame
ed
Sah
ar
was
born
in
Baghd
ad,
I
raq
in
1987
.
He
obt
ai
ned
his
B.
Sc
.
i
n
Elec
tron
ic
s
Engi
ne
eri
ng
in
2010,
M.S.c
.
in
El
ectroni
cs
En
gine
er
ing
in
2013
at
Univer
sit
y
of
Te
chnol
og
y
,
Ira
q.
In
2015
h
e
joi
n
ed
Ph
.
D.
stu
d
y
in
E
lectr
oni
c
and
Com
m
unic
a
ti
on
Eng
ineeri
ng
at
Univer
si
t
y
of
Technol
og
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.