Intern
ati
o
n
a
l
Journ
a
l of
Re
con
f
igur
able
and Embe
dded
Sys
t
ems
(I
JRES)
V
o
l. 4,
N
o
.
2
,
Ju
ly 20
15
, pp
. 14
2
~
16
0
I
S
SN
: 208
9-4
8
6
4
1
42
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJRES
Applicat
ion of
New Approach of
design fl
ow for
Hardware/S
o
ftware Emb
e
dded Syst
em
wi
th the Use of Desi
gn
Patt
erns
in Fuzzy cont
rol sys
t
em
Al
i
B
o
u
y
ah
ya
,
Y
a
ssi
ne
M
a
n
a
i
,
Jo
seph
H
a
ggè
ge
Departem
ent
of
Ele
c
tri
cal
Eng
i
n
eering
,
Labor
ato
r
y of Res
earch
i
n
Autom
a
tic
Co
ntrol,
Nation
a
l
Engineering s
c
h
ool of
Tunis, Univ
ersity
of
Tunis el Manar
,
BP 37
, Belvéd
ère, 100
2 Tunis, Tun
i
sia
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Dec 12, 2014
Rev
i
sed
Mar
27
, 20
15
Accepted Apr 24, 2015
This paper present a new method of
concep
tion of hardware/software
embedded s
y
stem design
methodolog
y
based
on use of desi
gn patter
n
approach
c
a
ll
ed
Abs
t
ract_fac
tor
y
. W
e
called th
is new design t
ool “smart
cel
l”. Th
e m
a
in idea of the con
c
eption of em
bed
d
ed s
y
s
t
em
s
des
i
gn is
bas
e
d
on the us
ed of
object-o
rien
ted
des
i
gn ULM
2
.0. W
h
en the s
m
art-cel
l is
im
plem
ented,
we justif
y
the
i
r u
s
es as
a design
tool
that allows, first, to
develop a s
p
ec
i
f
ied appl
ica
tion
of fuzz
y
con
t
r
o
ller c
a
ll
ed P
D
C (parall
e
l
distributed
conp
ensation)
. Seco
n
d
, th
e
specification of th
e gen
e
ration phas
e
s
of the
s
y
st
em
ar
chit
ectur
e d
e
sig
n
, and
ev
entua
l
l
y
p
a
rt
ition
i
ng th
e app
lic
ation
on heterogen
e
o
u
s platform b
a
sed on
h
a
rdware resource DSP and FPGA
software to
illustrate the proposed approach.
Keyword:
Abst
ract_Fact
ory
an
alytic m
o
d
e
l
H
Application/
Architecture
Ade
q
uacy
Design
p
a
ttern
Em
bedded
sy
st
em
MAC_
Cos
MAC_Op
eratio
n
MA
C_
Si
n
UM
L 2.
0
Copyright ©
201
5 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Ali Bou
y
ah
ya,
Depa
rtem
ent of Elect
ri
cal
E
n
gi
nee
r
i
n
g,
Nat
i
onal
En
gi
n
eeri
n
g
Sc
h
ool
of
Tu
ni
s,
Uni
v
ersity of
Tunis el Mana
r BP 37
, Belvé
d
ère,
1002 T
u
nis, T
unisia.
Em
a
il: al
i.b
ouyah
y
a@g
m
ail.
co
m
1.
INTRODUCTION
Th
ere are two orien
t
atio
ns i
n
em
bedded s
y
ste
m
s, the technologi
cal
fi
e
l
d a
n
d
t
h
e
m
e
t
h
o
dol
ogi
cal
one
. The m
e
t
hod
ol
o
g
i
cal
ori
e
nt
at
i
on [
1
]
t
r
y
t
o
devel
o
p t
h
e
em
bedde
d sy
s
t
em
desi
gn p
r
ocess by
e
x
am
i
n
i
ng
new
desi
gn
t
o
ol
s i
n
or
de
r t
o
decre
a
se t
h
e
c
o
m
p
l
e
xi
t
y
of e
m
bedded sy
st
e
m
s. There are
t
h
ree m
a
i
n
pr
o
b
l
e
m
s
d
u
ring
th
e syst
e
m
d
e
sig
n
:
th
e co
m
p
lex
ity, t
h
e h
a
rd
ware/so
f
t
w
are (HW/
SW) p
a
r
titio
n
i
n
g
and
th
e reusab
ility.
M
a
ny
fram
e
wor
k
s are de
vel
ope
d l
i
k
e t
r
an
sact
i
onal
en
vi
r
onm
ent
s
bet
w
een ap
pl
i
cat
i
on de
vel
o
pm
ent
and
architecture synthesis are use
d
to si
m
p
l
i
f
y
the desi
gn
p
r
oc
ess. Desi
gne
rs
ar
e recu
rring
to
raise th
e ab
st
raction
l
e
vel
,
f
r
om
R
e
gi
st
er T
r
ans
f
er
Level
(R
T
L
)
t
o
sy
st
em
l
e
vel
.
As a c
o
nseq
uence
,
a
gap
bet
w
ee
n a
ppl
i
cat
i
o
n
devel
opm
ent
a
n
d
arc
h
i
t
ect
ure
sy
nt
hesi
s a
p
p
ears [
2
,
3]
.
I
n
or
der
t
o
s
o
l
v
e
t
h
i
s
p
r
o
b
l
e
m
,
m
a
ny
desi
g
n
t
ool
s a
r
e
devel
ope
d i
n
or
der t
o
i
m
prove em
bed
d
e
d
sy
st
em
perfo
rm
ances [4,
5
]
. In t
h
e
fi
el
d
of c
o
nt
rol
s
y
st
em
im
pl
em
ent
a
t
i
o
n, m
a
ny
sol
u
t
i
ons
were
de
ve
l
ope
d
fo
r l
i
n
ea
r t
i
m
e i
nvari
an
t
(LTI
) c
o
nt
r
o
l
an
d em
bed
d
e
d
real
t
i
m
e
cont
r
o
l
ap
pl
i
cat
i
ons [
6
-
8
]
.
In [
9
]
,
a desi
g
n
m
e
t
hod
ol
o
g
y
based
on a t
r
a
n
sact
i
o
nal
m
o
d
e
l
whi
c
h i
s
i
n
s
e
rt
ed
betwee
n t
h
e a
pplication a
n
d the a
r
c
h
itecture is
pre
s
ente
d
.
In
th
is way, th
e app
licatio
n
is refin
e
d in
an
in
term
ed
iate le
v
e
l wh
ich
contain
s
th
e arch
i
t
ectu
r
e p
a
ra
m
e
ters. Fro
m
th
is lev
e
l, th
e im
p
l
e
m
en
tatio
n
step
is
achi
e
ve
d i
n
or
der
t
o
ge
nerat
e
t
h
e R
T
L
arc
h
i
t
ect
ure.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
64
I
J
RES Vo
l. 4
,
N
o
. 2
,
Ju
ly 2
015
:
14
2
–
160
14
3
Our ap
pro
ach
o
f
d
e
si
g
n
t
r
y to
so
lv
e t
h
e com
p
lex
i
t
y
p
r
o
b
l
e
m
, it co
n
s
ists
to
d
e
v
e
lop
t
w
o in
term
ed
iate
envi
ro
nm
ent
s
i
n
or
der
t
o
m
i
nim
i
ze t
h
e gap
bet
w
ee
n a
ppl
i
c
at
i
on
devel
o
p
m
ent
and
arc
h
i
t
ect
ure sy
nt
hes
i
s [1
0,
11
, 12
, 20
, 21
, 22]
.
In o
r
der t
o
de
v
e
l
op re
usa
b
l
e
desi
g
n
t
o
ol
s i
n
di
ffer
e
nt
fi
el
d
s
, t
h
e desi
g
n
p
a
t
t
e
rns we
re us
ed [1
3,
20]
.
Ou
r a
p
p
r
oach
i
s
use
d
i
n
c
o
n
t
rol
fi
el
d
w
h
i
c
h
we
deve
l
o
p
a fuz
z
y
co
ntr
o
ller fo
r
no
nlin
ear sy
stem
(inverte
d
p
e
ndu
lu
m
)
. Man
y
research
es in
th
is fiel
d
are p
e
rform
e
d
.
Th
e
w
o
r
k
of
[1
4, 15
]
de
vel
o
p an object analysis
pat
t
e
rn f
o
r em
bed
d
e
d
sy
st
em
, fu
rt
her
,
a wra
ppe
r desi
gn pa
t
t
e
rn fo
r an ad
apt
i
ng
beha
vi
o
u
r o
f
t
h
e so
ft
IPs w
a
s
p
r
op
o
s
ed
in
[16
]
. Th
e reu
s
abilit
y o
f
In
tellectu
a
l Pro
p
e
rty (IP)
b
l
o
c
k
s
hav
e
b
e
en
p
e
rfo
rmed
ex
ten
s
i
v
ely fo
r
d
e
sign
h
a
rdware ap
p
licatio
ns and
IP
b
l
o
c
k
s
syn
t
h
e
sis [1
7-1
8
]
. Mak
[1
9
]
p
r
esen
ts a d
e
si
g
n
p
a
ttern
m
o
d
e
llin
g
in
Un
ified
Mod
e
llin
g
Lan
guag
e
(UML). Man
y
research
es
are p
e
rfo
r
m
e
d
for th
e reu
s
ab
i
lity p
r
o
b
l
em
in
o
r
d
e
r
t
o
devel
op
ne
w desi
gn t
o
ol
s
t
h
at
encaps
u
l
a
t
e
al
l
codesi
g
n
p
h
a
ses in
o
r
der to
i
m
p
l
e
m
e
n
t in
tellectu
a
l
p
r
op
erty
(IP
) bl
ocks
[2
0
]
. One at
t
e
m
p
t
pr
o
pose
d
i
n
[
2
1, 2
2
]
ha
ve as aim
t
o
devel
o
p
t
h
e
sma
r
tcell
desi
g
n
t
o
ol
s i
n
or
de
r
to
im
p
l
e
m
en
t
HW and
SW IP b
l
o
c
ks for
h
e
terog
e
n
e
ou
s
p
l
atfo
rm
s. Ou
r co
n
t
ribu
tio
n to
reso
lv
e th
e
reu
s
ab
ility
pr
o
b
l
e
m
consi
s
t
s
i
n
t
h
e sy
nt
h
e
si
s of IP
bl
oc
ks f
o
r
har
d
w
a
r
e
and s
o
ft
ware
sol
u
t
i
o
ns f
r
o
m
di
rect
acy
cl
ic gra
p
h
(D
AC
).
T
h
e
pr
op
ose
d
a
p
pr
oa
ch e
x
am
i
n
es t
h
e
Ab
stra
ct_Facto
ry
d
e
sign
patter
n
to pro
duce th
e app
licati
o
n and
sy
nt
hesi
s of
I
P
bl
oc
ks H
W
/
S
W (FP
GA/
DS
P).
Thi
s
pape
r i
s
or
ga
ni
zed as
fol
l
o
ws.
Sect
i
on
2 int
r
oduces the fuzzy
controller
PDC (Pa
r
allel
di
st
ri
b
u
t
e
d C
o
m
p
ensat
i
on) a
nd
Taka
gi
-S
u
g
e
no
f
u
zzy
sy
st
em
. Sect
i
on
3, t
h
e a
p
pl
i
cat
i
on
of t
h
e p
r
o
pos
ed
app
r
oach t
o
t
h
e i
nve
rt
ed p
e
n
dul
um
i
s
di
scussed i
n
de
tails. Th
e con
c
lusion
and
th
e
fut
u
re works are
presente
d
in
th
e last sectio
n of t
h
is p
a
p
e
r.
2.
FUZ
Z
Y
CONTROLLER
PDC (PARALL
EL DIST
RIB
U
TED
COMPENSATION)
Taka
gi
-S
uge
n
o
fuzzy
m
odel
i
s
a m
u
l
t
i
m
odel ap
pr
oach
ve
ry
use
d
t
o
m
odel
i
ze no
n l
i
n
ea
r
sy
t
e
m
s
by
co
nstru
c
tion
with
id
en
tificatio
n of inpu
t-o
u
t
p
u
t
d
a
ta.
Ma
ny
m
echanical syste
m
s are m
odeling
with T
-
S
fuzzy
syste
m
.
Th
e con
tin
ou
sT-S
fu
zzy m
o
d
e
l for a
no
n
l
i
n
ear system
is written
as fo
llows.
11
If
(
)
i
s
a
n
d
th
e
n
()
()
()
1
.
.
....
()
()
ip
i
p
ii
i
zt
M
a
n
d
z
t
i
s
M
xt
A
x
t
B
u
t
ir
yt
C
x
t
(
1
)
Whic
h
(
1
,
2
....
,
1
,
2
.....
)
ij
M
ir
j
p
is th
e fu
zzy set and
r is th
e nu
m
b
er o
f
m
o
d
e
l ru
les,
n
xt
is th
e
states vector;
m
ut
is th
e inpu
t v
e
cto
r
;
nn
i
A
, the states m
a
trix,
nm
i
B
the control a
nd
1
,
.
.
...
.,
p
zt
z
t
are know premise
va
riables.
Th
e T-S fu
zzy
m
o
d
e
l can
be
written
also
as
fo
llow:
11
11
,
rr
ii
i
i
i
ii
rr
ii
ii
wz
t
A
x
t
B
u
t
w
z
t
C
x
t
xt
y
t
wz
t
w
z
t
r: is th
e nu
mb
er of m
o
d
e
l ru
les.
Wi
t
h
1
1
1,
2
,
,
r
ii
j
j
j
i
i
r
i
i
wz
t
M
z
t
wz
t
hz
t
i
r
wz
t
(
2
)
The term
ij
j
M
zt
is th
e grad
e of
me
m
b
e
r
s
h
i
p
o
f
j
zt
in
ij
M
.
Since
1
0
0
1
.
....
.
r
i
i
i
wz
t
wz
t
i
r
,
W
e
ha
ve
1
01
1
i
r
i
i
hz
t
hz
t
(3
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RES I
S
SN
:
208
8-8
7
0
8
Ap
pl
i
c
at
i
o
n
of
N
e
w
Ap
pro
a
c
h
of
desi
g
n
f
l
o
w
f
o
r H
a
r
d
w
a
re/
Sof
t
w
are
E
m
be
dde
d
Syst
e
m
…
(
A
l
i
Bouy
a
h
ya
)
14
4
Th
e
fin
a
l
o
u
t
pu
t can b
e
written
as fo
llo
w
1
1
r
ii
i
i
r
ii
i
x
th
z
t
A
x
t
B
u
t
yt
h
z
t
C
x
t
(4
)
The sy
st
em
used i
n
t
h
i
s
pa
per
i
s
t
h
e
n
o
n
-
l
i
n
e
a
r i
n
vert
e
d
pe
n
dul
um
m
odel
e
d
by
Ta
kagi
-
S
uge
n
o
fuzzy
sy
st
em
mg
M
g
Fk
3
x
k
1
x
k
Fi
gu
re
1.
I
nve
r
t
ed pe
n
dul
um
The i
nve
rt
ed
p
e
nd
ul
um
i
s
m
o
del
i
s
ed i
n
x1
88
,
8
8
b
ecau
s
e it’
s no
t co
ntr
o
llab
l
e ar
ound
2
and
2
The m
e
m
b
ersh
i
p
f
u
nct
i
ons
1
,
1
,2
,
3
,4
i
hx
t
i
are obtained fr
om
th
e
pr
odu
cts
12
1
2
11
2
2
,,
e
t
FF
F
F
[2
3]
11
11
1
2
12
21
1
2
.
.
hx
F
F
hx
F
F
21
31
1
2
22
41
1
2
.
.
hx
F
F
hx
F
F
(
5
)
1
1
1
c
o
s(
)
0
.034
8
1
0
.0348
x
F
,
2
1
1
1c
o
s
(
)
10
.
0
3
4
9
x
F
,
1
11
2
1
1.5359
sin(
)
(
1
.535
9
1
)
x
x
F
x
,
2
11
2
1
1
.
53
59
(
s
i
n
(
)
)
(
1
.5
35
9
1
)
x
x
F
x
The
In
ve
rt
ed
p
e
nd
ul
um
m
ode
l
i
ng a
n
d
gi
ves t
h
e
fol
l
o
wi
n
g
m
a
t
r
i
ces:
12
0
1
00
0
1
00
2
5
.91
161
-
0
.27
8
9
0
39.71
095
0.294
096
-
0
.27
8
9
0
3
9
.710
95
,
0
0
01
0
0
01
-
1
.1
193
0
.
0
122
2
0
-
1
4
.
8
0
1
-
0.01
270
4
0
.01
222
0
-
14.80
1
AA
34
0
1
00
0
1
00
25.91161
-
0
.2789
0
1
.3819
0.2
94096
-
0
.2789
0
1
.38
1
9
,,
0
0
01
0
0
01
0.038951
0.00
04252
0
-
14.8
0
1
0
.0004421
0.0004252
0
-
14.801
AA
12
3
4
00
0
0
1.
7078
0.
05943
1.7078
0.
05943
,,
e
t
.
00
0
0
0.
02529
0.
02529
0.
02529
0.
02529
BB
B
B
Fi
nal
l
y
t
h
e
pen
dul
um
i
nvert
ed
i
s
m
odel
i
s
ed v
i
a Taka
gi
-S
ue
n
o
i
n
4
r
u
l
e
s as
f
o
l
l
o
w
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
64
I
J
RES Vo
l. 4
,
N
o
. 2
,
Ju
ly 2
015
:
14
2
–
160
14
5
11
1
11
22
2
11
33
1
12
44
2
12
Ru
l
e
1
:
if
is F
the
n
Ru
l
e
2
:
if
is F
th
e
n
Ru
l
e
3
:
if
is
F
th
e
n
Ru
l
e
4
:
if
is F
th
e
n
x
tA
x
t
B
u
t
x
yt
C
x
t
x
tA
x
t
B
u
t
x
yt
C
x
t
x
tA
x
t
B
u
t
x
yt
C
x
t
x
tA
x
t
B
u
t
x
yt
C
x
t
Th
e
PDC con
t
ro
l law is written
is t
h
e
n
e
x
t
equ
a
tio
n
1
1
1
r
ii
r
i
ii
r
i
i
i
wz
tF
x
t
ut
h
z
t
F
x
t
wz
t
(
6
)
i
F
rep
r
esen
ts the feedb
ack
matrices stab
iliz
in
g
th
e system in
clo
s
ed
l
o
op
. Th
is can b
e
fou
n
d
e
d
by th
e
ap
p
lication
o
f
t
h
e LM
Is techn
i
q
u
e
s (Lin
ear
Matrix
In
eq
u
a
l
ity) with
th
e u
s
e th
e LM
I too
l
b
o
x
i
n
Matlab
.
The m
a
trices
i
F
corresponding to
each
ru
le a
r
e
written as
foll
ow:
1
41
.2
12
3
2
8.
17
26
-
7
6.
17
26
119
.1
81
6
F
(
7
)
2
2
3
.5
44
6.530
1
-
3
2
.6
125
34
.29
5
6
F
(8
)
3
22.
8
407
8.
8
906
-
4
5
.
7
8
9
2
1
22.
6
646
F
(9
)
4
14.
65
84
1
1
.
704
4
-
17.
06
76
5
3
.
431
5
F
(1
0)
Fo
r th
e m
o
d
e
lisatio
n
and
co
ncep
tio
n d
e
tails
for th
e PDC con
t
ro
ller, see
[23
]
.
Aft
e
r
we de
sc
ri
be t
h
e Ta
ka
g
i
-Su
g
e
no
fuzz
y
sy
st
em
,
t
h
e m
odel
i
s
at
i
on o
f
t
h
e pe
n
dul
u
m
i
nvert
ed
wi
t
h
t
h
i
s
technique and
we pre
s
ent the
PDC cont
ro
l
law. W
e
p
r
esen
t
th
e approach
of conc
eption of Hardwa
re/Soft
w
ar
e
Em
b
e
d
d
e
d Syste
m
with
th
e
Use of
Design
Pattern
s
for th
e
PDC co
n
t
ro
ller.
3.
DEFINITION OF THE
CONTE
X
T
OF T
H
E PROBLEM
In t
h
i
s
a
p
p
r
oc
h
of
de
si
g
n
em
bedde
d sy
st
em
[20]
.
We
de
fi
ne
fo
u
r
pri
n
ci
pal
s
’ act
o
r
s
wh
o a
r
e i
n
v
o
l
v
e
d
in
th
e op
eration
o
f
th
e un
ified
stru
ct
u
r
e. The first
is
related
t
o
th
e id
en
tificatio
n
of m
e
th
od
s
fo
r m
o
delin
g
sp
ecification
,
we call it th
e ap
p
licatio
n.
Th
e seco
nd
,
related
p
r
ob
lem with
th
e syn
t
h
e
sis of IPs b
l
o
c
k
s
HW
/SW, we call it th
e arch
it
ectu
r
e. Th
e th
i
r
d
is th
e
c
o
mm
unication bet
w
een the
diffe
r
ent st
ruct
ure,
that’s
mean
s th
e prob
lem
o
f
n
e
twork
i
n
g
sub
s
ystem
s
o
f
th
e
ov
erall syste
m
. Th
e fou
r
th
actor is th
e
p
a
rtitio
n
i
n
g
har
d
ware
/ so
ftware
.
T
h
e follo
win
g
fig
u
re
ill
ustrat
es t
h
e c
o
ntext
of the
unified st
ruct
ure. [20]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RES I
S
SN
:
208
8-8
7
0
8
Ap
pl
i
c
at
i
o
n
of
N
e
w
Ap
pro
a
c
h
of
desi
g
n
f
l
o
w
f
o
r H
a
r
d
w
a
re/
Sof
t
w
are
E
m
be
dde
d
Syst
e
m
…
(
A
l
i
Bouy
a
h
ya
)
14
6
Fi
gu
re
2.
U
n
i
f
i
e
d st
r
u
ct
u
r
e
[2
0]
The unified s
t
ructure
is de
com
posed
by
four
princi
pales actors: t
h
e
devel
o
ppe
m
ent of t
h
e
appl
i
cat
i
o
n, t
h
e sy
nt
hesi
s
of
t
h
e a
r
c
h
i
t
ect
ure,
th
e p
a
rtionn
ing
of
th
e h
a
rdware/software an
d th
e
com
m
uni
cat
i
on.
We
de
fi
ne al
l
t
h
i
s
act
or
s
on
e by
one
.
There
are t
w
o levels of a
b
s
t
raction i
n
the
design
o
f
the PDC con
t
roller. In
th
e
first lev
e
l of
ab
straction
,
t
h
e wh
o
l
e system
is
m
o
d
e
led
b
y
th
e
un
ifie
d
structure calle
d «Unifie
d
Structure system
level», it
brea
ks
down the whole system into
subsyste
m
s
, each of t
h
em
is
m
odele
d by a sec
o
nd-level unified st
ruct
ure
called «Uni
fied St
ruct
ure
sec
o
nd le
vel».
3. 1 Abs
t
r
a
cti
o
n
L
e
vel
s
of
t
h
e
U
n
i
f
i
e
d Str
u
cture
Th
is ap
pro
ach o
f
two
level
ab
straction
d
e
sig
n
propo
sed is illu
strated
by th
e fi
g
u
re
3 [20
]
.
In th
e
fo
llowing
sectio
n
s
, th
e correspo
nd
ing
applicatio
n
of
PDC controller a
n
d its arc
h
itecture a
r
e
de
velope
d
th
ro
ugh
th
e un
i
f
ied
st
ru
ct
u
r
es
co
rresp
ond
ing
to
th
e
fu
zzy mo
d
e
l
d
i
sting
u
i
sh
ed.
Fig
u
re 3
.
Mu
lti-lev
e
l d
e
sign
ab
straction
[20
]
For t
h
e i
m
pl
em
ent
a
t
i
on o
f
t
h
e ha
rd
wa
re and s
o
ft
ware
bl
ock
s
IPs
,
t
w
o envi
ro
nm
ent
s
i
n
t
e
rfaci
n
g
wi
t
h
cards
o
n
w
h
i
c
h t
h
e t
a
r
g
et
i
s
goi
ng t
o
i
m
pl
em
ent
t
h
e em
bedde
d c
ode e
x
a
m
i
n
ed, t
h
e C
C
S (C
o
d
e C
o
m
pos
er
Unified Structure
sy
ste
m
level
Decom
position
Verificatio
n
Syste
m
in
teg
r
ation
Syste
m
v
a
lid
ation
Un
i
f
ied
Stru
ct
u
r
e-seco
nd
level
H-
m
odel
Verificatio
n
MAC_Mode
l
Verificatio
n
Validation of
application
Synthesis of
r
e
ssour
ces
IPs
Verificatio
n
Partitionning of
h
a
rd
ware/so
f
tw
are
Verificatio
n
Esti
m
a
tio
n
of
ress
o
u
rces
Validation of
Architecture
schu
d
e
ling
Valid
atio
n o
f
Partitio
nn
ing
A
pplica
tion
A
r
chite
cture
com
m
unication
P
a
rtition
i
ng
U
n
ified
st
ruct
ure
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
64
I
J
RES Vo
l. 4
,
N
o
. 2
,
Ju
ly 2
015
:
14
2
–
160
14
7
Studi
o 3.1) to interface
with soft
wa
re processor
DSP, and the Xilinx IS
E
envi
ronm
ent interfacing with
FPGA
boa
r
d
[
2
4,
25]
.
3.
2
Uni
f
i
e
d
St
ructure
S
y
ste
m
L
e
vel
Thi
s
sect
i
on c
ont
ai
n
s
t
h
e m
odel
i
n
g o
f
sy
st
em
-l
evel
appl
i
cat
i
on.
W
e
p
r
esent
t
h
e gl
o
b
a
l
m
odel
of
in
v
e
r
t
ed
p
e
ndulu
m
. Su
b
s
eq
u
e
n
tly, w
e
pr
esent th
e
m
o
d
e
ling
o
f
t
h
is syste
m
i
n
th
e state space p
l
an
t. Th
e sy
ste
m
lev
e
l g
r
an
u
l
arity
m
o
d
e
lin
g
ap
p
lication
u
s
ed
in
th
is appr
oach
to
is th
e
m
o
d
e
lisatio
n
in
th
e state sp
ace o
f
th
e
phy
si
cal
sy
st
e
m
(i
nvert
ed
pe
nd
ul
um
).
3.
3
Unified
Structure
Sec
o
n
d
Level
I
n
th
is section
,
w
e
p
r
esen
t a dif
f
e
r
e
n
t
second-
lev
e
l un
if
ied
str
u
ctur
es su
ch
as u
n
i
f
i
ed
structu
r
e inp
u
t
,
th
e un
ified
structu
r
e “con
tro
ller”,
“
p
hysical syste
m
” and “
o
utputs”.
3.
3.
1 Uni
f
i
e
d
Struc
t
ure Inp
u
t
Th
e
un
ified
stru
cture inpu
t con
t
ain
th
e fo
llowing
d
a
ta:
-
Th
e acqu
isitio
n
o
f
th
e si
g
n
a
l
d
e
liv
ered
b
y
the setpo
i
n
t
-
Th
e m
u
ltip
lica
tio
n
of th
e
referen
ce si
g
n
a
l
d
e
liv
ered
b
y
a matrix
g
a
i
n
to
adj
u
st th
e static g
a
in
o
f
t
h
e clo
s
ed
loop system
.
-
Transm
issio
n
of th
e sign
al su
pp
lied
to th
e con
t
ro
l d
e
v
i
ce.
3.3.2
Unified Structure
Contr
o
ller
Un
i
f
ied
st
ru
ct
ure co
n
t
ro
ller con
t
ain
s
t
h
e fo
llowing
d
a
ta:
-
Th
e acqu
isitio
n
o
f
th
e si
g
n
a
l
d
e
liv
ered
b
y
the un
ified stru
ctu
r
e inp
u
t
.
-
Gene
rat
i
o
n
of t
h
e c
ont
rol
si
gn
al
.
-
Tran
sm
i
ssi
on o
f
t
h
e
co
nt
r
o
l
si
gnal
t
o
the
phy
sical
syste
m
u
n
ified
stru
cture.
3.
3.
3 Uni
f
i
e
d
Struc
t
ure Ph
y
s
i
c
al
S
y
stem
The unified structure physical
co
nt
ai
n t
h
e f
o
l
l
owi
n
g
dat
a
:
-
Acqu
isitio
n of
th
e con
t
ro
l si
gn
al su
pp
lied
b
y
th
e co
n
t
ro
l
d
e
v
i
ce,
-
Ad
j
u
st
i
n
g t
h
e
c
ont
rol
si
gnal
t
o
t
h
e
phy
si
cal
pr
ocess.
-
Tran
sm
i
ssi
on o
f
t
h
e
o
u
t
p
ut
si
g
n
al
t
o
t
h
e
o
u
t
p
ut
de
vi
ce.
3.
3.
4 Uni
f
i
e
d
Struc
t
ure Ou
t
put
The u
n
i
f
i
e
d st
r
u
ct
u
r
e out
put
has
th
e fo
llowing
featu
r
es.
-
Acqu
isitio
n of
th
e sign
al
d
e
livered b
y
t
h
e
p
hysical syste
m
.
-
Gen
e
rating
th
e ou
tpu
t
sign
al.
Fi
gu
re
4.
U
n
i
f
i
e
d st
r
u
ct
u
r
e se
con
d
l
e
vel
[
2
0]
The
be
nefi
t
s
of i
m
pl
em
ent
i
ng
t
h
i
s
uni
fi
ed
desi
g
n
a
p
p
r
oach
by
m
a
steri
n
g t
h
e c
o
m
p
l
e
xi
ty
o
f
e
m
b
e
d
d
e
d syste
m
s stru
cture;
first is so
l
v
ing
th
e co
m
p
lex
ity
b
y
raising
t
h
e
lev
e
l of ab
stractio
n
.
Seco
nd
i
s
the
reu
s
ab
ility o
f
IPs b
l
o
c
k
s
to
min
i
mize th
e t
i
m
e-to
-m
ark
e
t, it
is reso
lv
ed
b
y
th
e d
e
v
e
l
o
p
m
en
t o
f
th
e u
n
i
fied
stru
cture. Fi
n
a
l
l
y, th
e p
r
o
b
l
em o
f
au
to
m
a
ti
n
g
o
f
t
h
e d
e
si
gn
is pro
cessed. In
ad
d
ition
to
th
ese adv
a
n
t
ages, th
is
ap
pro
ach illu
str
a
tes o
t
h
e
r
objectiv
es of
lo
cal o
r
d
e
r. In
t
h
e
d
e
v
e
l
o
p
m
en
t of
th
e
un
if
ied str
u
ct
u
r
e
second lev
e
l,
we e
x
am
i
n
e t
h
ree m
odel
s
f
o
r
t
h
e
devel
opm
ent
o
f
t
h
e a
ppl
i
cat
i
on. T
h
e
fi
r
s
t
m
odel
i
s
t
h
e m
odel
of
t
h
e
uni
fi
e
d
structure, t
h
e s
econd is t
h
e a
n
alytical
m
odel and
fina
lly is th
e m
o
d
e
l o
f
MAC_Op
eratio
n
on
wh
ich
we will
work to sy
nthe
size the arc
h
itecture.
In the
architecture, eac
h sec
o
nd-level
u
n
i
f
i
e
d st
ruct
ure has a
p
r
oc
ess
f
o
r
t
h
e
sy
nt
hesi
s o
f
i
n
t
e
l
l
ect
ual
p
r
o
p
ert
y
bl
oc
ks
IPs
ha
r
d
wa
re and
s
o
ft
wa
re.
Har
d
w
a
re IPs
are
sy
nt
hesi
ze
d wi
t
h
VH
DL l
a
n
gua
ge,
w
h
i
l
e
t
h
e s
o
ft
ware
IPs
are
sy
nt
hesi
ze
d
w
i
t
h
t
h
e C
++
l
a
ng
ua
ge.
I
n
put
Controller
Physical system
Output
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RES I
S
SN
:
208
8-8
7
0
8
Ap
pl
i
c
at
i
o
n
of
N
e
w
Ap
pro
a
c
h
of
desi
g
n
f
l
o
w
f
o
r H
a
r
d
w
a
re/
Sof
t
w
are
E
m
be
dde
d
Syst
e
m
…
(
A
l
i
Bouy
a
h
ya
)
14
8
4.
APPLI
CATI
O
N DEVEL
O
PMENT
The de
vel
opm
ent of t
h
e application is pe
rformed accord
ing to the algorithm presente
d in Figure 5. It
co
nsists in
: first, th
e
d
e
v
e
l
o
pmen
t o
f
t
h
e mo
d
e
l
o
f
th
e
unified struct
ure,
second t
h
e
dev
e
l
opm
ent
of a
n
al
y
t
i
cal
m
odel
H, a
n
d
fi
nal
l
y
t
h
e
dev
e
l
opm
ent
of
a
wo
rk
m
odel
b
a
sed
o
n
t
h
e
M
A
C
_
Ope
r
at
i
o
n.
The
desi
gn
p
a
t
t
e
r
n
Ab
st
ract_
F
act
ory is resp
on
sible to
fu
lfill th
e
task
s o
f
th
e act
o
r
app
licatio
n. Fig
u
re 4
illu
strates th
e ap
p
licatio
n
actor.
Fi
gu
re 5.
App
licatio
n
Actor
[20
]
4.
1 Devel
o
pm
ent of
th
e Ana
l
yti
c
al
M
o
del
The
objective
of the a
n
alytical
m
odel is t
o
m
ode
l th
e un
ified
stru
cture
pr
oposed by t
h
e
state space
rep
r
ese
n
t
a
t
i
on
[2
0]
. T
h
e a
n
al
y
t
i
cal
m
odel
t
h
a
t
we
pr
o
pose
i
s
de
fi
ne
d
by
t
h
e
f
o
l
l
o
wi
ng
eq
u
a
t
i
on
[2
0]
,,
,
Y
(1
1)
0
0
{,
,
,
,
,
,
:
:
:
;:
;:
,
n
ii
i
n
ii
i
AB
C
D
X
y
u
u
i
np
ut
v
e
c
t
o
r
o
f
t
he
s
y
s
t
e
m
y
o
ut
pu
t
v
e
c
t
o
r
of
t
h
e
s
y
s
t
e
m
X
s
t
a
t
e
ve
ct
o
r
o
f
t
h
e
s
ys
t
e
m
Y
y
y
p
r
o
to
c
o
l
o
f
c
o
m
m
u
n
i
c
a
tio
n
f
f
pr
ot
o
c
ol
o
f
c
o
nt
r
o
l
XY
(1
2)
In
th
is equ
a
tion
s
, th
e m
o
d
e
l
e
n
capsulates the
m
odel of syste
m
wr
itten in t
h
e state space plant, the
state vector, the input
vect
or
an
d
th
e ou
tpu
t
v
ector
.
repr
e
s
ent
s
t
h
e
pa
rt
of t
h
e sy
st
em
di
st
ur
ba
nce.T
h
e
fu
nct
i
o
n set
s
t
h
e com
m
uni
cat
i
on bet
w
ee
n
har
d
ware a
n
d
soft
ware
res
o
urces.
In t
h
e analytical
m
odel H we
defi
ne yhree st
eps.
In the
first
step we
de
fine
the tra
n
sfe
r
f
u
nct
i
o
n
m
odel
o
f
t
h
e
uni
fi
ed st
ruct
ure.
The
se
con
d
step is to transform
each
transfe
r
function
to the state space repr
ese
n
tation
using a sim
p
lified com
p
anion
m
odel. The third ste
p
is to transf
orm
the state space compani
on m
odel
re
prese
n
tation
in the disc
rete dom
ain
usi
n
g t
h
e
ope
ra
t
o
r
to
devel
o
p recurrence
s e
q
uations.
Developm
ent o
f
unified
structure m
odel
D
e
velopm
ent o
f
anal
yt
ic
m
odel
,,
,
Y
Developm
ent o
f
work m
odel
bas
e
d in
M
A
C_O
p
eration
unit
A
pplica
tion
actor
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
64
I
J
RES Vo
l. 4
,
N
o
. 2
,
Ju
ly 2
015
:
14
2
–
160
14
9
4.
2 Devel
o
pm
ent of
M
o
del
o
f
W
o
rk
M
A
C
The de
vel
opm
ent
of t
h
e m
odel
of
w
o
r
k
M
A
C
co
nsi
s
t
s
t
o
t
r
ans
f
o
r
m
t
h
e recu
rre
nc
es equat
i
ons
devel
ope
d f
r
o
m
t
h
e anal
y
t
i
c
al
m
odel
t
o
t
h
e gra
p
hs o
f
t
a
sks
based
on t
h
e M
A
C
_
ope
rat
i
o
n. A M
A
C
_
O
p
erat
i
o
n
(Mu
ltip
ly and
Accu
m
u
late) is re
p
r
esen
ted b
y
Figu
re 6 [2
0
]
.
b
a
c
Fi
gu
re
6.
M
ac_
Ope
r
at
i
o
n [
2
0]
The
recu
rre
nce
eq
uat
i
o
n
o
f
a
4t
h
or
de
r i
n
ver
t
ed pe
n
dul
um
, i
t
con
v
ert
e
d i
n
t
o
a set
of
t
a
sks
as t
h
e
gra
p
hs s
h
o
w
n
in
Figur
e
7
.
Fi
gu
re 7.
Tra
n
s
f
orm
a
tion of
a recurre
n
ce
e
quatio
n
with
t
h
e
MAC un
it [20
]
Th
e n
e
x
t
sectio
n
is d
e
v
o
t
ed
to
th
e syn
t
h
e
sis o
f
em
b
e
d
d
e
d syste
m
o
f
o
u
r PDC stab
iliza
tio
n
con
t
ro
l law for
i
nve
rt
ed pe
nd
u
l
um
sy
st
em
.
5.
SYNTHESIS OF
THE ARCHITECT
URE
OF THE
CONTROL L
A
W PDC
The actor
of
Arc
h
itecture is
the responsi
b
le for sy
nt
hesis
the hard
wa
re IPs an
d so
ftw
a
re IPs that
realize the target architecture
.
It is
withi
n
this architecture
as the em
be
dded system
perform
s
the de
dicated
appl
i
cat
i
o
n. T
w
o m
e
t
hods o
f
devel
o
pi
n
g
so
ft
ware a
nd
ha
r
d
wa
re IPs are
di
st
i
n
g
u
i
s
he
d.
The fi
rst
m
e
t
h
od i
s
t
o
devel
op t
h
e c
ode t
o
i
m
pl
em
ent
on t
a
rge
t
boar
d
by
ex
am
i
n
i
ng t
h
e cor
r
es
po
n
d
i
n
g f
unct
i
o
ns
. The
seco
nd
m
e
t
hod i
s
t
h
e e
xpl
oi
t
a
t
i
on
of
e
x
i
s
t
i
n
g
com
p
i
l
e
rs i
n
M
a
t
l
a
b s
u
ch
as R
eal
Ti
m
e
Wo
rks
h
op
(R
T
W
) e
nvi
ro
nm
ent
.
5.
1.
De
vel
o
p
m
ent
of
Arc
h
i
t
e
c
ture I
P
S
o
f
t
Thi
s
sect
i
on e
x
am
i
n
es t
h
e pr
obl
em
of i
n
t
e
g
r
at
i
on
of t
h
e cont
rol
l
a
w
on
a DSP ca
rd
us
i
ng t
h
e T
L
C
(Tar
get
La
ng
u
a
ge C
o
m
p
i
l
e
r)
of R
eal
Ti
m
e
Wo
r
k
sh
o
p
co
m
p
il
er and
de
vel
o
pm
ent
env
i
ro
nm
ent
C
C
S
IDE
T
I
D
S
P.
b
e
lon
g
s
t
o
th
e TI C200
0 D
S
P f
a
m
ily, t
h
e TI
TMS 320
C281
2.
x1[k]
x2[
k
]
x3 [
k
]
x4
[k]
x1[
k
+1]
x2[
k
+
1
]
x3 [
k
+
1
]
x4
[k+
1
]
y
[
k
]
MA
C
MAC
MAC
MAC
MAC
MAC
MAC
MAC
MAC
MAC
MAC
b
c
MAC
y
=
a+b*c
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RES I
S
SN
:
208
8-8
7
0
8
Ap
pl
i
c
at
i
o
n
of
N
e
w
Ap
pro
a
c
h
of
desi
g
n
f
l
o
w
f
o
r H
a
r
d
w
a
re/
Sof
t
w
are
E
m
be
dde
d
Syst
e
m
…
(
A
l
i
Bouy
a
h
ya
)
15
0
5.
2.
De
vel
o
p
m
ent
of
Arc
h
i
t
e
c
ture I
P
H
a
rd
Thi
s
sect
i
on e
x
am
i
n
es t
h
e pr
obl
em
of i
n
t
e
g
r
at
i
on
of t
h
e cont
rol
l
a
w
on
a DSP ca
rd
us
i
ng t
h
e T
L
C
(Tar
get
La
ng
u
a
ge C
o
m
p
i
l
e
r)
of R
eal
Ti
m
e
Wo
r
k
sh
o
p
co
m
p
il
er and
de
vel
o
pm
ent
env
i
ro
nm
ent
C
C
S
IDE
T
I
D
S
P. Belong
s t
o
th
e TI C200
0 D
S
P f
a
m
ily, t
h
e TI
TMS 320
C281
2.
6.
IMPLEME
N
TATION OF
THE CONT
ROL LA
W PDC ON
HETEROGE
NEOUS PLATFORM
In t
h
i
s
sect
i
o
n,
we p
r
ese
n
t
t
h
e im
pl
em
ent
a
t
i
on
of t
h
e co
nt
rol
l
a
w P
D
C
.
It
begi
ns
by
b
r
eaki
ng t
h
e
cont
rol
sy
st
em
of a
n
i
n
vert
e
d
pen
d
u
l
u
m
i
n
a set
of
uni
fied s
t
ructures. T
h
e
second
phase is the de
velopment of
th
e ap
p
lication as a set o
f
g
r
ap
h
s
task
s. Th
e last p
h
a
se
i
s
t
h
e sy
nt
hesi
s of
har
d
ware an
d s
o
ft
ware I
P
s bl
o
c
ks o
f
het
e
r
oge
ne
ous
archi
t
ect
u
r
e.
6.
1.
Unified Desi
gn Structure
At th
e first lev
e
l o
f
abstractio
n
(system lev
e
l), th
e co
n
t
ro
l system o
f
in
v
e
rted
p
e
ndu
lu
m
i
s
decom
pose
d
i
n
t
o
f
o
u
r
s
u
b
-
sy
s
t
em
s (fuzzy
m
odel
s
T
S
) a
r
e
m
odeled by a second-level
un
ified
st
ru
ct
u
r
e. Th
i
s
d
eco
m
p
o
s
ition
is p
e
rfo
rm
ed
usin
g th
e
d
e
sign p
a
ttern
Ab
tract_
F
actory presen
ted
i
n
Fi
g
u
re 8
.
Fi
gu
re
8.
Déc
o
m
posi
t
i
on o
f
e
m
bedded
sy
st
em
wi
t
h
desi
g
n
pat
t
e
rn
A
b
st
rac
t
_Fact
o
r
y
[
20]
Em
bedded
Sy
s
t
em
Abst
ract Factory
Create App ()
Create Arch ()
App
licatio
n
Out
A
p
PhA
pp
Cm
dA
pp
In
A
pp
Arc
hitecture
Out
A
rc
PhArc
h
Cm
dArc
h
In
Arc
h
U
S
-In
pu
t
Creat
e A
pp ()
Creat
e A
r
ch
()
US-C
o
n
tr
ol
Creat
e A
pp ()
Creat
e A
r
ch
()
US-
p
hy
si
q
syst
Creat
e A
pp ()
Creat
e A
r
ch
()
US-
O
utp
u
t
Creat
e A
pp ()
Creat
e A
r
ch
()
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
64
I
J
RES Vo
l. 4
,
N
o
. 2
,
Ju
ly 2
015
:
14
2
–
160
15
1
Aft
e
r
dec
o
m
posi
t
i
on
of t
h
e
sy
st
em
i
n
st
ruct
ure
uni
fi
e
d
f
i
rst
l
e
vel
,
we
devel
op
u
n
i
f
i
e
d st
r
u
ct
ur
es
seco
nd l
e
vel
.
We defi
ne f
o
ur u
n
i
f
i
e
d st
ru
ct
ures seco
n
d
l
e
vel
.
Uni
f
i
e
d
st
ruct
ure i
n
p
u
t
,
wi
ch c
ont
a
i
n t
h
e
acq
u
i
sition
of th
e inp
u
t
si
g
n
a
l
(setpo
in
t) m
u
ltip
lied
b
y
th
e static g
a
in
to
con
t
ro
l th
e clo
s
ed
loo
p
system
. Th
e
second
uni
fied structure is the controlle
r, it
contain t
h
e ele
m
entaries cont
ro
l laws with
stat
es feedba
ck for
each s
u
bsystem
s
of the pe
ndulum
inverte
d
syste
m
. The th
ird c
ontai
n the
m
odelisation of physical syste
m
and
t
h
e f
o
ort
h
co
nt
ai
n t
h
e
o
u
t
p
ut
s
i
gnal
.
Whe
n
we
exa
m
i
n
es t
h
e
pat
t
e
rn
de
si
g
n
Ab
s
t
ract
-Fact
o
r
y
d
e
vel
o
ped
by
Fi
gu
re
8,
we s
p
e
c
i
f
y
t
h
e
ne
xt
cl
asses de
vel
o
ped
by
C
+
+
wi
t
h
vi
s
u
al
St
udi
o1
0.
-
C
u
st
om
er:
i
s
t
h
e em
bedded
sy
st
em
.
-
Abst
ract classe
s: App
licatio
n, Arch
itectu
r
e,
-
C
onc
ret
e
cl
ass
e
s:
In
p
u
t
,
c
ont
r
o
l
l
e
r,
Phy
s
i
cal
Sy
st
em
, out
put
.
-
A
b
st
r
act Pr
oducts: I
n
A
p
p
,
Cmd
a
pp
, Ph
app
,
ou
tAp
p
,
I
n
ar
ch,
C
m
d
a
r
c
h
,
Ph
arch
,
Ou
tar
c
h
,
In
Figu
re
9
,
we p
r
esen
t little
p
a
rty of C++ co
d
e
of d
e
sign
p
a
ttern
Abstact-Facto
r
y t
h
at allo
ws to
decom
pose
t
h
e co
nt
r
o
l
sy
st
e
m
of t
h
e i
nve
rt
ed
pe
nd
ulum
. T
h
e CreateApp (), CreateArch () a
r
e
functions
decl
are
d
vi
rt
u
a
l
t
y
pe. A vi
rt
u
a
l
funct
i
o
n i
s
a fun
c
t
i
on
defi
ned i
n
a cl
ass and i
s
i
n
t
e
n
d
e
d
t
o
be o
v
er
ri
d
d
en i
n
deri
ved
cl
asses
.
Fi
gu
re
9.
Desi
gn
pat
t
e
r
n
A
b
s
t
ract
-Fact
o
r
y
6.
2.
Devel
o
pme
n
t
of Ap
plicati
o
n
for In
ver
t
ed
Pendulum
S
y
s
t
em
6.
2.
1.
H
A
n
al
yti
c
al
Mo
del
o
f
the Co
ntr
o
l
S
y
st
e
m
The
devel
opm
ent of the a
n
alytical
m
odel H is ba
sed
on the state space of each s
u
bsystem
.
There are
three
pha
ses in the analytical m
odel H. The
first phase
consists in
th
e
d
e
v
e
lop
m
en
t o
f
t
h
e tran
sfer functio
n
m
odel of the unified sec
o
nd l
e
vel struct
ure
.
The sec
o
nd
phase is to trans
f
orm
each
trans
f
er function in the
state space
usi
n
g a sim
p
lified represe
n
tation suc
h
as m
odi
fi
ed c
o
m
p
anion
m
odel. The
t
h
ird
phase i
n
vol
v
es the
t
r
ans
f
o
r
m
a
ti
on
of st
at
e repre
s
ent
a
t
i
on m
o
d
e
l
i
n
t
h
e di
scret
e
dom
ai
n usi
ng t
h
e o
p
erat
or
to
d
e
v
e
lop th
e
recurre
nce e
quations.
Th
e an
alytical m
o
d
e
l is represen
ted
b
y
th
e follo
wing
eq
u
a
ti
o
n
s
,,
,
(
1
3
)
Y
class abs
F
actory {
p
u
b
lic:
virt
ual Archite
cture c
r
eatearc
h
()
{
return ne
w a
r
chitecture; }
vi
rt
ual
ap
pl
i
ca
tion c
r
eateapp()
{
retur
n
new
application; }
class sysem
b
{
p
u
b
lic:
syse
m
b
(
a
b
s
Facto
r
y& f
actor
y)
{
architecture* Arc
h
itecture
= factory.c
r
eatea
rch();
ap
p
lication
*
Ap
p
licatio
n =
fa
ctory.createa
pp();
}
Evaluation Warning : The document was created with Spire.PDF for Python.