I
nte
rna
t
io
na
l J
o
urna
l o
f
Rec
o
nfig
ura
ble a
nd
E
m
bedd
e
d Sy
s
t
em
s
(
I
J
RE
S)
Vo
l.
10
,
No
.
3
,
N
o
v
em
b
e
r
2
0
2
1
,
p
p
.
18
6
~
19
4
I
SS
N
: 2
0
8
9
-
4
8
6
4
,
DOI
: 1
0
.
1
1
5
9
1
/ij
r
es
.
v
10
.i
3
.
p
p
18
6
-
19
4
186
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
r
es.ia
esco
r
e.
co
m
Neura
l ne
t
imple
menta
tion o
f
stea
m pro
perti
es o
n F
PG
A
R.
V.
S.
K
rish
na
Dut
t
1
,
R.
G
a
nes
h
2
, P.
P
re
m
cha
nd
3
1
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
,
CVR Co
l
leg
e
o
f
En
g
i
n
e
e
rin
g
,
Hy
d
e
ra
b
a
d
,
I
n
d
ia
2
De
p
t
a
rtme
n
t
o
f
El
e
c
tr
o
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
,
CV
R
Co
ll
e
g
e
o
f
E
n
g
i
n
e
e
rin
g
,
Hy
d
e
ra
b
a
d
,
In
d
ia
3
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
,
Os
m
a
n
ia Co
ll
e
g
e
o
f
E
n
g
i
n
e
e
rin
g
,
Hy
d
e
ra
b
a
d
,
I
n
d
ia
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
n
2
,
2
0
21
R
ev
is
ed
J
u
l
2
7
,
2
0
21
Acc
ep
ted
Au
g
11
,
2
0
21
Re
a
l
ti
m
e
a
p
p
li
c
a
ti
o
n
s
li
k
e
m
o
d
e
l
p
re
d
icti
v
e
c
o
n
tro
l,
m
o
n
it
o
ri
n
g
a
n
d
d
a
ta
re
c
o
n
c
il
iatio
n
o
f
p
o
we
r
p
lan
ts
a
n
d
in
d
u
str
ial
p
r
o
c
e
ss
e
s
e
m
p
lo
y
n
o
n
li
n
e
a
r
m
a
th
e
m
a
ti
c
a
l
m
o
d
e
ls
a
n
d
re
q
u
ire
th
e
rm
o
d
y
n
a
m
ic
p
ro
p
e
rti
e
s
a
n
d
t
h
e
ir
d
e
riv
a
ti
v
e
s
o
f
w
o
rk
i
n
g
flu
i
d
s.
Ap
p
li
c
a
ti
o
n
s
li
k
e
su
p
e
r
h
e
a
ter
t
e
m
p
e
ra
tu
re
c
o
n
tro
l
b
a
se
d
o
n
e
n
e
rg
y
b
a
lan
c
e
a
n
d
re
a
l
t
ime
d
a
ta
re
c
o
n
c
il
iatio
n
,
re
q
u
ire
a
n
e
fficie
n
t
a
n
d
a
c
o
m
p
a
c
t
m
e
th
o
d
fo
r
sim
u
l
tan
e
o
u
s
e
sti
m
a
ti
o
n
o
f
th
e
rm
o
d
y
n
a
m
ic
p
ro
p
e
rti
e
s
,
a
n
d
th
e
ir
p
a
rt
ia
l
d
e
riv
a
ti
v
e
s
su
it
a
b
le
fo
r
imp
lem
e
n
tatio
n
in
f
ield
-
p
r
o
g
ra
m
m
a
b
le
g
a
te
a
rra
y
(
F
P
G
A
)
.
Ho
we
v
e
r,
th
e
c
o
m
p
lex
m
a
th
e
m
a
ti
c
a
l
fo
rm
u
la
ti
o
n
s
o
f
th
e
se
p
r
o
p
e
rti
e
s
p
ro
h
ib
it
d
irec
t
imp
lem
e
n
tatio
n
s
in
F
P
G
As
.
S
in
g
le
a
rti
ficia
l
n
e
u
ra
l
n
e
two
rk
(
ANN
)
a
rc
h
it
e
c
tu
re
is
u
se
d
to
re
p
lac
e
th
e
e
n
ti
re
c
o
d
e
in
h
ig
h
e
r
lev
e
l
l
a
n
g
u
a
g
e
s,
ru
n
n
in
g
in
t
o
a
fe
w
th
o
u
sa
n
d
li
n
e
s.
F
P
G
A
imp
lem
e
n
tatio
n
o
f
a
c
o
m
p
a
c
t
n
e
u
ra
l
n
e
two
r
k
f
o
r
th
e
e
n
ti
re
ra
n
g
e
o
f
th
e
rm
o
d
y
n
a
m
ic
p
r
o
p
e
rti
e
s is
p
re
se
n
ted
.
Larg
e
a
rg
u
m
e
n
ts
in
si
g
m
o
i
d
fu
n
c
ti
o
n
a
re
fa
c
to
re
d
i
n
to
a
p
ro
d
u
c
t
o
f
in
te
g
e
r
a
n
d
a
fra
c
ti
o
n
a
l
p
a
rt
wh
ich
is
re
p
re
se
n
ted
u
sin
g
se
ries
a
p
p
r
o
x
im
a
t
io
n
with
fiv
e
term
s
o
n
l
y
a
n
d
t
h
e
i
n
teg
e
rs
a
re
re
p
re
se
n
ted
in
lo
o
k
u
p
tab
le
(
LUT
)
.
Th
is
e
n
s
u
re
s
o
p
ti
m
u
m
st
o
ra
g
e
a
n
d
c
o
m
p
u
tati
o
n
a
l
b
u
rd
e
n
fo
r
th
e
a
b
o
v
e
a
p
p
li
c
a
ti
o
n
s.
T
h
e
AN
N
is
imp
l
e
m
e
n
ted
in
I
EE
E
7
5
4
f
lo
a
ti
n
g
p
o
i
n
t
wit
h
sy
n
th
e
sis
in
Xili
n
x
IS
E
d
e
sig
n
su
it
e
u
sin
g
Ve
ril
o
g
HD
L.
Th
e
re
su
lt
s
a
re
p
re
se
n
ted
fo
r
a
ty
p
ica
l
p
re
ss
u
re
v
e
rsu
s sa
tu
ra
ti
o
n
tem
p
e
ra
tu
re
.
K
ey
w
o
r
d
s
:
Ar
tific
ial
n
eu
r
al
n
etwo
r
k
FP
GA
I
E
E
7
5
4
f
lo
atin
g
p
o
in
t
T
h
er
m
o
d
y
n
am
ic
p
r
o
p
er
ties
Ver
ilo
g
HDL
Xilin
x
I
SE
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
R
.
V.
S.
Kr
is
h
n
a
Du
tt
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
C
VR
C
o
lleg
e
o
f
E
n
g
in
ee
r
in
g
Vastu
n
ag
ar
,
I
b
r
ah
im
p
t
n
am
,
H
y
d
er
ab
a
d
,
I
n
d
ia
E
m
ail: K
r
is
h
n
ad
u
tt.r
v
s
@
cv
r
.
a
c.
in
1.
I
NT
RO
D
UCT
I
O
N
R
ec
en
t
tr
en
d
s
in
p
o
wer
p
lan
t
an
d
in
d
u
s
tr
ial
ap
p
licatio
n
s
in
d
icate
d
ev
el
o
p
m
en
ts
lik
e
d
ata
r
ec
o
n
ciliatio
n
(
DR
)
,
Gu
o
et
a
l
.
[
1
]
m
o
d
el
p
r
e
d
ictiv
e
co
n
tr
o
l
(
MPC
)
,
W
an
g
et
a
l
.
[
2
]
an
d
o
p
tim
izatio
n
,
Nieg
o
d
ajew
et
a
l
.
[
3
]
to
im
p
r
o
v
e
p
er
f
o
r
m
an
ce
o
f
p
lan
ts
an
d
p
r
o
ce
s
s
es.
T
h
ey
ar
e
b
ased
o
n
th
e
f
ir
s
t
p
r
in
cip
le
th
er
m
o
d
y
n
am
ic
f
o
r
m
u
latio
n
s
as
well
co
m
p
lex
m
ath
em
atica
l
m
o
d
els.
T
h
ey
a
r
e
lim
ited
t
o
o
f
f
lin
e
s
im
u
latio
n
s
o
r
im
p
lem
e
n
ted
in
s
tan
d
al
o
n
e
d
esk
to
p
s
y
s
tem
s
.
D
ir
ec
t
im
p
lem
en
t
at
i
o
n
o
f
t
h
e
s
e
alg
o
r
ith
m
s
in
m
icr
o
co
n
tr
o
ller
s
an
d
f
ield
-
p
r
o
g
r
am
m
ab
le
g
ate
ar
r
ay
s
(
FP
GAs
)
,
o
f
ten
e
n
co
u
n
ter
ed
i
n
p
o
wer
an
d
p
r
o
ce
s
s
p
lan
t
d
is
tr
ib
u
ted
co
n
t
r
o
l
s
y
s
tem
s
(
DC
S)
,
is
d
if
f
icu
lt
an
d
co
m
p
u
tatio
n
ally
in
ef
f
icie
n
t
.
On
e
o
f
th
e
co
n
s
tr
ain
ts
is
ef
f
icien
t
im
p
lem
en
tatio
n
o
f
th
er
m
o
d
y
n
am
ic
p
r
o
p
er
ties
o
f
th
e
wo
r
k
in
g
f
lu
id
s
lik
e
wate
r
,
s
team
an
d
g
ass
es,
in
th
e
FP
GAs.
T
h
e
ap
p
licatio
n
s
r
eq
u
ir
e
r
ea
l
tim
e
p
r
ed
ictio
n
o
f
th
er
m
o
d
y
n
am
ic
p
r
o
p
e
r
ties
lik
e
en
th
al
p
y
,
al
o
n
g
with
th
eir
f
ir
s
t
an
d
s
ec
o
n
d
o
r
d
er
p
ar
tial
d
er
iv
ativ
es.
T
h
e
p
r
o
p
er
ties
ar
e
r
ep
r
esen
ted
b
y
n
o
n
lin
ea
r
Gib
b
s
f
r
ee
en
er
g
y
f
o
r
m
u
latio
n
s
[
4
]
.
T
h
es
e
f
u
n
ctio
n
s
a
r
e
im
p
lem
en
te
d
i
n
h
ig
h
e
r
lev
el
lan
g
u
ag
es
r
u
n
n
in
g
in
to
5
0
0
0
-
7
0
0
0
lin
es
o
f
co
d
e.
T
h
ese
ar
e
ev
al
u
ated
iter
ativ
ely
.
T
h
ey
a
r
e
r
e
p
ea
ted
ly
ca
lled
in
th
e
ap
p
lic
atio
n
s
an
d
ar
e
n
o
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Reco
n
f
ig
u
r
a
b
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4
8
6
4
N
eu
r
a
l n
et
imp
leme
n
ta
tio
n
o
f
s
tea
m
p
r
o
p
erti
es o
n
F
P
GA
(
R.
V.
S.
K
r
is
h
n
a
Du
tt
)
187
am
en
ab
le
f
o
r
im
p
lem
en
tatio
n
in
m
icr
o
co
n
tr
o
ller
s
o
r
FP
GAs.
Ad
d
itio
n
ally
,
th
eir
p
ar
tial
d
er
iv
ativ
es
ar
e
co
m
p
u
ted
n
u
m
er
ically
b
y
r
e
p
ea
ted
ca
lls
to
t
h
ese
f
u
n
ctio
n
s
.
T
h
is
m
ak
es
t
h
em
u
n
s
u
itab
le
f
o
r
r
ea
l
tim
e
co
m
p
u
tatio
n
s
in
FP
GAs.
R
ec
en
t
r
esear
ch
is
f
o
c
u
s
ed
o
n
d
e
v
elo
p
in
g
ap
p
licatio
n
s
p
ec
if
ic
p
r
o
p
er
ties
o
f
wate
r
an
d
s
team
c
o
v
er
in
g
s
u
b
-
cr
iti
ca
l
an
d
s
u
p
er
-
c
r
itical
r
eg
io
n
s
im
p
lem
en
ted
in
Mo
d
elo
n
’
s
Mo
d
elica
t
h
er
m
al
p
o
wer
lib
r
ar
y
an
d
Sp
lin
e
-
B
a
s
ed
tab
le
lo
o
k
-
u
p
m
eth
o
d
,
[5
]
,
[
6
]
.
Ho
wev
er
,
m
an
y
o
f
th
ese
ef
f
o
r
ts
a
r
e
n
o
t
d
ir
ec
tly
s
u
ited
f
o
r
im
p
lem
en
tatio
n
in
FP
GA
s
u
s
ed
in
th
e
co
m
m
er
cial
DSC
p
latf
o
r
m
.
C
ar
d
aliag
u
et
an
d
E
u
v
r
ar
d
[
7
]
r
e
p
o
r
ted
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
(
ANNs
)
with
ex
p
licit
weig
h
ts
to
ap
p
r
o
x
i
m
ate
a
f
u
n
ctio
n
an
d
its
d
er
iv
ativ
es a
n
d
h
ig
h
lig
h
ted
u
s
ag
e
f
o
r
c
o
n
tr
o
l.
Hash
em
an
d
Sch
m
eiser
[
8
]
s
h
o
wed
th
at
a
lin
ea
r
co
m
b
i
n
atio
n
o
f
a
s
et
o
f
f
ee
d
f
o
r
war
d
ANNs
tr
ain
ed
to
a
p
p
r
o
x
im
ate
a
f
u
n
ctio
n
an
d
its
d
er
iv
ativ
es
i
m
p
r
o
v
e
ac
cu
r
ac
y
as
ag
ain
s
t
a
s
in
g
le
ANN.
W
h
ile
th
ese
lay
th
e
f
o
u
n
d
atio
n
f
o
r
u
s
in
g
ANNs
f
o
r
ap
p
r
o
x
im
atin
g
a
f
u
n
ctio
n
an
d
its
d
er
iv
ativ
es,
Kr
is
h
n
ad
u
tt
an
d
Kr
is
h
n
aiah
[
9
]
d
em
o
n
s
tr
ated
t
h
e
p
o
wer
o
f
a
f
ew
s
im
p
le
AN
Ns
to
s
im
u
ltan
eo
u
s
ly
ap
p
r
o
x
im
ate
s
team
p
r
o
p
er
ties
an
d
th
e
c
o
r
r
esp
o
n
d
in
g
d
er
iv
ativ
es
o
f
en
tire
s
team
p
r
o
p
er
t
ies;
th
u
s
r
ep
lacin
g
th
o
u
s
an
d
s
o
f
lin
es
o
f
c
o
d
e
in
h
ig
h
er
la
n
g
u
a
g
e.
T
h
ese
ANNs
ap
p
r
o
x
im
ate
th
e
c
o
m
p
lex
s
team
p
r
o
p
er
ties
to
th
e
r
eq
u
ir
ed
d
eg
r
ee
o
f
ac
c
u
r
ac
y
f
o
r
r
ea
l tim
e
ap
p
licatio
n
s
.
I
n
s
p
ite
o
f
th
e
av
aila
b
ilit
y
o
f
m
an
y
h
ig
h
-
le
v
el
lan
g
u
ag
e
to
o
l
s
,
th
e
p
r
o
g
r
ess
in
im
p
lem
en
tin
g
ANNs
in
FP
GA
a
s
s
ee
n
in
l
iter
atu
r
e
[
1
0
]
-
[
1
5
]
is
s
lo
w.
R
ec
en
t
s
u
r
v
ey
s
[
1
6
]
-
[
1
8
]
h
i
g
h
lig
h
t
th
at
FP
GA
b
ased
n
eu
r
al
n
etwo
r
k
ac
ce
ler
ato
r
s
ar
e
p
r
e
f
er
r
ed
o
v
e
r
th
e
ap
p
licatio
n
s
p
ec
if
ic
in
teg
r
at
ed
cir
cu
its
(
ASI
C
s
)
d
u
e
to
th
eir
en
er
g
y
ef
f
icien
c
y
with
a
s
m
a
ll
f
o
o
tp
r
in
t,
b
u
t
m
an
y
p
r
o
b
le
m
s
an
d
ch
allen
g
es
n
ee
d
to
b
e
ad
d
r
ess
ed
.
T
h
ese
b
r
in
g
o
u
t
t
h
e
ch
allen
g
es
in
tr
ain
in
g
d
ee
p
n
eu
r
al
n
etwo
r
k
s
(
DNNs
)
with
FP
GA
an
d
p
r
o
p
o
s
e
a
p
e
r
f
o
r
m
an
c
e
m
etr
ic
an
d
ev
al
u
ati
o
n
wo
r
k
f
lo
w
to
co
m
p
ar
e
th
e
FP
GA
-
b
ased
s
y
s
tem
s
f
o
r
DNN
tr
ain
in
g
s
p
ec
if
ic
to
co
m
p
u
ter
v
is
io
n
task
s
.
Ho
wev
er
,
f
o
r
a
p
r
etr
ain
ed
n
etwo
r
k
,
o
n
ly
d
e
v
i
c
e
u
tili
za
tio
n
an
d
e
n
er
g
y
ef
f
icien
cy
m
ay
b
e
co
n
s
id
er
ed
.
R
ev
iews
h
ig
h
lig
h
t
m
ajo
r
is
s
u
es
lik
e
d
ata
r
ep
r
esen
tatio
n
,
i
m
p
lem
en
tatio
n
o
f
in
n
er
p
r
o
d
u
ct
b
etwe
en
weig
h
t
m
at
r
ix
a
n
d
p
r
ev
io
u
s
lay
er
o
u
tp
u
ts
,
ac
tiv
atio
n
f
u
n
ctio
n
s
,
d
e
v
ice
u
tili
za
tio
n
,
en
er
g
y
ef
f
icien
cy
,
f
o
r
d
if
f
er
en
t a
p
p
licatio
n
s
.
No
n
-
lin
ea
r
ac
tiv
ati
o
n
f
u
n
ctio
n
is
th
e
m
ai
n
co
n
ce
r
n
o
f
m
an
y
r
esear
c
h
er
s
wh
ile
im
p
le
m
en
t
in
g
a
n
ANN
in
FP
GA.
Piazza
et
a
l
.
[
1
9
]
r
e
p
o
r
ted
ad
ap
ta
b
le
lo
o
k
u
p
ta
b
le
(
L
UT
)
-
b
ased
ac
tiv
a
tio
n
f
u
n
ctio
n
s
f
o
r
n
eu
r
o
n
s
,
in
lear
n
in
g
b
y
b
ac
k
war
d
d
if
f
er
en
ce
with
d
i
f
f
er
en
t
lear
n
in
g
r
ates
an
d
two
lo
o
k
u
p
tab
les,
o
n
e
f
o
r
weig
h
ts
an
d
th
e
f
o
r
c
o
ef
f
icien
ts
r
esp
ec
ti
v
ely
,
wh
er
ea
s
B
ieu
et
a
l
.
[
2
0
]
co
m
p
u
ted
th
e
s
ig
m
o
id
f
u
n
ctio
n
a
n
d
its
d
er
iv
ativ
e
in
d
ig
ital
h
a
r
d
war
e
b
y
a
s
u
m
o
f
s
tep
s
r
esu
ltin
g
i
n
ar
ea
-
ef
f
icien
c
y
.
R
ec
o
n
f
ig
u
r
atio
n
ca
p
ab
ilit
ies
o
f
th
e
Atm
el
FP
GA
ar
e
e
x
p
lo
ite
d
b
y
L
y
s
ig
h
t
et
a
l.
[
2
1
]
f
o
r
im
p
lem
en
tin
g
lar
g
er
ANN
with
i
n
d
iv
id
u
al
lay
er
s
o
f
th
e
n
etwo
r
k
with
tim
e
m
u
ltip
lex
in
g
o
n
th
e
lo
g
ic
a
r
r
ay
at
t
h
e
co
s
t
o
f
s
y
s
tem
p
er
f
o
r
m
a
n
c
e.
T
is
an
et
a
l
.
[
2
2
]
in
v
esti
g
ated
d
if
f
er
e
n
t
ap
p
r
o
x
i
m
atin
g
f
u
n
ctio
n
s
f
r
o
m
t
h
e
p
o
in
t
o
f
v
iew
o
f
h
ar
d
war
e
r
eso
u
r
ce
u
tili
za
tio
n
an
d
in
d
u
ce
d
er
r
o
r
s
an
d
co
n
cl
u
d
e
d
th
at
p
iece
wis
e
lin
ea
r
ap
p
r
o
x
im
atio
n
o
f
th
e
ac
tiv
atio
n
f
u
n
ctio
n
is
th
e
b
est.
Pro
b
lem
s
en
co
u
n
ter
ed
in
im
p
lem
en
tin
g
an
ANN
in
VHDL
ar
e
r
e
p
o
r
ted
b
y
P
.
Do
n
d
o
n
et
a
l
.
[
2
3
]
wh
er
ein
s
ig
m
o
id
ac
tiv
atio
n
is
ap
p
r
o
x
im
ated
b
y
s
am
p
lin
g
o
f
L
o
g
s
ig
f
u
n
ctio
n
with
ar
g
u
m
en
t
b
etw
ee
n
0
to
1
.
Ng
a
h
et
al
.
[
2
4
]
u
s
ed
co
m
b
in
atio
n
o
f
s
ec
o
n
d
o
r
d
e
r
n
o
n
-
lin
ea
r
f
u
n
ctio
n
(
SONF)
an
d
d
if
f
er
en
tial
L
UT
f
o
r
im
p
lem
en
tin
g
an
ANN.
T
h
e
t
wo
-
s
tep
ap
p
r
o
ac
h
is
r
ep
o
r
ted
to
h
av
e
an
im
p
r
o
v
ed
ac
cu
r
a
cy
th
at
is
1
0
tim
es
b
etter
th
an
th
at
o
f
u
s
in
g
o
n
l
y
SONF
an
d
twice
b
etter
th
an
ju
s
t
u
s
in
g
L
UT
.
L
i
et
a
l
.
[
2
5
]
im
p
lem
e
n
ted
a
n
eu
r
o
n
b
lo
c
k
with
s
ig
m
o
id
f
u
n
ctio
n
u
s
in
g
t
h
e
C
OR
DI
C
alg
o
r
ith
m
.
So
m
e
o
f
th
e
a
p
p
licatio
n
s
o
f
ANN
im
p
lem
en
tatio
n
o
n
FP
G
A
in
clu
d
e
class
if
icatio
n
o
f
t
h
e
r
eg
io
n
o
f
p
ix
els
i.e
.
,
h
an
d
r
e
g
io
n
s
b
y
K
r
ip
s
et
a
l
.
[
2
6
]
u
s
in
g
t
h
r
ee
i
n
p
u
ts
r
ep
r
esen
tin
g
R
GB
v
alu
es
with
a
s
in
g
le
h
id
d
en
lay
er
an
d
o
n
e
o
u
tp
u
t
with
d
at
a
r
ep
r
esen
ted
b
y
in
teg
e
r
s
an
d
weig
h
ts
s
ca
led
u
p
an
d
r
o
u
n
d
e
d
o
f
f
to
th
e
n
ea
r
est
in
teg
er
.
R
ec
o
g
n
itio
n
o
f
d
ig
its
u
s
in
g
a
n
etwo
r
k
with
3
0
0
in
p
u
ts
an
d
1
0
o
u
tp
u
ts
with
a
s
in
g
le
n
eu
r
o
n
is
r
ep
o
r
te
d
b
y
L
atin
o
et
a
l
.
[
2
7
]
.
A
m
eth
o
d
o
f
co
n
f
ig
u
r
ab
le
ML
P
with
a
s
in
g
le
n
e
u
r
o
n
b
lo
ck
with
f
lo
a
tin
g
p
o
in
t
a
d
d
an
d
m
u
ltip
ly
u
n
its
alo
n
g
with
ac
t
iv
atio
n
f
u
n
ctio
n
as
L
UT
f
o
r
a
s
m
a
r
t
p
o
s
itio
n
s
en
s
o
r
o
f
s
o
lar
p
an
els
h
as
b
ee
n
s
tu
d
ied
b
y
Dą
b
r
o
wsk
i
et
a
l
.
[
2
8
]
.
A
n
eu
r
al
class
if
ier
with
f
i
x
ed
p
o
in
t
r
ep
r
esen
tatio
n
an
d
1
2
-
b
its
f
o
r
d
etec
tin
g
d
am
ag
ed
t
o
o
th
ed
g
ea
r
s
u
s
in
g
v
ib
r
o
ac
o
u
s
tic
s
ig
n
als is
h
ig
h
lig
h
ted
b
y
Po
lat
an
d
Yild
ir
im
[
2
9
]
.
T
h
ese
s
tu
d
ies
co
n
s
id
er
s
ig
m
o
i
d
f
u
n
ctio
n
with
a
r
g
u
m
e
n
t
in
t
h
e
r
an
g
e
(
0
-
1
)
.
FP
GA
im
p
lem
en
tatio
n
o
f
an
ANN
f
o
r
s
im
u
ltan
eo
u
s
p
r
e
d
ictio
n
o
f
th
er
m
o
d
y
n
am
ic
p
r
o
p
er
ties
an
d
t
h
e
ir
d
e
r
iv
ativ
es
r
e
q
u
ir
ed
in
ad
v
an
ce
d
ap
p
licatio
n
s
lik
e
MPS,
DR
,
an
d
o
p
tim
izatio
n
is
n
o
t
r
ep
o
r
ted
.
C
u
s
to
m
ANNs
s
u
itab
le
f
o
r
s
im
u
ltan
eo
u
s
esti
m
atio
n
o
f
b
o
th
en
th
alp
y
an
d
its
d
er
iv
ativ
es
s
u
itab
le
f
o
r
p
o
wer
p
lan
t
ap
p
licatio
n
s
ar
e
tak
en
u
p
f
o
r
im
p
lem
en
tatio
n
in
FP
GA.
I
n
t
h
e
f
o
llo
win
g
s
ec
tio
n
s
,
FP
GA
im
p
lem
en
tatio
n
o
f
a
s
in
g
le
n
e
u
r
al
n
etwo
r
k
ca
lled
Steam
Net
i
s
d
escr
ib
ed
.
T
h
e
s
im
u
ltan
eo
u
s
u
s
e
o
f
L
UT
an
d
T
ay
lo
r
s
er
ies
f
o
r
lar
g
e
ar
g
u
m
en
ts
o
f
s
ig
m
o
id
f
u
n
ctio
n
is
p
r
esen
ted
.
2.
CUST
O
M
NE
T
S F
O
R
T
H
E
RM
O
DYNA
M
I
C
P
RO
P
E
R
T
I
E
S O
F
WAT
E
R/
ST
E
A
M
Fig
u
r
e
1
r
ep
r
esen
ts
wate
r
an
d
s
team
p
r
o
p
er
ties
u
s
ed
in
i
n
d
u
s
tr
ial
ap
p
licatio
n
s
[
4
]
.
T
h
e
d
if
f
er
e
n
t
r
eg
io
n
s
r
ep
r
esen
t
s
u
b
co
o
led
1
)
.
Su
p
er
cr
itical
wate
r
/s
team
.
2)
.
Su
p
er
h
ea
t
s
team
.
3)
.
Satu
r
atio
n
.
4)
.
Hig
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
20
89
-
4
8
6
4
I
n
t J Reco
n
f
ig
u
r
a
b
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
10
,
No
.
3
,
No
v
em
b
er
2
0
2
1
:
18
6
–
19
4
188
tem
p
er
atu
r
e
s
team
.
5)
.
R
esp
ec
tiv
ely
.
T
h
ey
ar
e
r
ep
r
esen
ted
b
y
h
ig
h
ly
n
o
n
lin
ea
r
f
u
n
ctio
n
s
.
T
h
e
p
r
o
p
er
ties
lik
e
en
th
alp
y
,
e
n
tr
o
p
y
,
s
p
ec
if
ic
v
o
l
u
m
e,
ar
e
d
ep
en
d
en
t o
n
p
r
o
ce
s
s
p
ar
am
eter
s
lik
e
p
r
ess
u
r
e
an
d
tem
p
er
atu
r
e.
Fig
u
r
e
1
.
R
eg
io
n
s
o
f
s
team
p
r
o
p
er
ties
as p
er
I
APW
S
-
I
F9
7
T
h
e
en
th
alp
y
f
u
n
ctio
n
s
f
o
r
all
th
e
r
eg
io
n
s
ar
e
r
e
p
r
esen
ted
b
y
d
if
f
e
r
en
t
n
e
u
r
al
n
etwo
r
k
s
as
s
h
o
wn
in
Fig
u
r
e
2
.
T
h
e
n
etwo
r
k
s
Su
p
Net,
Su
b
C
Net,
SatVn
et
an
d
SatFn
et
r
ep
r
esen
t
en
th
al
p
y
i
n
s
u
p
er
h
ea
ted
,
s
u
b
co
o
led
,
s
atu
r
ated
v
ap
o
r
a
n
d
s
atu
r
ated
f
lu
id
r
eg
io
n
s
o
f
Fi
g
u
r
e
1
.
P2
T
n
etwo
r
k
is
u
s
ed
to
d
ec
id
e
th
e
zo
n
e
i
n
wh
ich
s
team
p
r
o
p
er
ty
is
r
eq
u
ir
ed
.
Fu
n
ctio
n
(
P,
T
)
in
Fig
u
r
e
2
r
ep
r
esen
ts
th
er
m
o
d
y
n
am
ic
p
r
o
p
er
ty
,
f
o
r
ex
am
p
le
en
th
al
p
y
,
as
a
f
u
n
cti
o
n
o
f
p
r
ess
u
r
e
P
an
d
tem
p
er
a
tu
r
e
T
.
A
s
in
g
le
n
e
u
r
al
n
etwo
r
k
,
Steam
Net,
with
co
m
p
ac
t
ar
ch
itectu
r
e
2
x
1
0
x
5
x
3
,
s
h
o
wn
i
n
Fig
u
r
e
3
,
is
u
s
ed
to
r
ep
r
esen
t
all
th
ese
d
if
f
er
en
t
n
etwo
r
k
s
.
T
h
e
ar
ch
itectu
r
e
is
ch
o
s
en
s
u
ch
th
at
th
e
ac
cu
r
ac
y
o
f
an
d
b
o
th
t
h
e
d
er
iv
ativ
es
an
d
ar
e
o
b
tai
n
ed
with
g
o
o
d
ac
cu
r
ac
y
.
Steam
Net
s
to
r
es
d
if
f
er
en
t
weig
h
t
an
d
b
ias
m
atr
ices
s
u
itab
le
f
o
r
d
if
f
er
en
t
r
e
g
io
n
s
o
f
Fig
u
r
e
2
.
Steam
Net
is
d
esig
n
ed
with
s
ig
m
o
id
ac
tiv
atio
n
f
u
n
ctio
n
r
ep
r
e
s
en
ted
b
y
(
1
)
an
d
(
2
)
.
Sig
m
o
id
f
u
n
ctio
n
tak
es
th
e
weig
h
ted
s
u
m
o
f
o
u
t
p
u
ts
f
r
o
m
th
e
p
r
ev
io
u
s
lay
e
r
.
FP
GA
im
p
lem
e
n
tatio
n
is
ac
h
iev
ed
b
y
u
s
in
g
s
in
g
le
p
er
ce
p
tr
o
n
with
s
ig
m
o
id
ac
tiv
atio
n
f
u
n
ctio
n
.
Z
=
∑
(
+
)
(
1
)
=
1
(
1
+
−
)
(
2
)
W
h
er
e
an
d
ar
e
weig
h
t
m
at
r
ix
b
etwe
en
lay
er
s
an
d
an
d
b
ias v
ec
to
r
at
lay
er
r
esp
ec
tiv
ely
.
is
th
e
in
p
u
t
v
ec
t
o
r
in
th
e
th
lay
er
.
I
n
th
e
Steam
Net,
ty
p
ical
v
alu
es
o
f
an
d
ar
e
in
t
h
e
r
a
n
g
e
(
-
1
6
.
1
7
3
5
to
+2
3
.
1
7
4
)
.
Fo
r
n
o
r
m
alize
d
v
alu
es
o
f
,
th
is
r
esu
lts
in
lar
g
er
v
al
u
es
o
f
|
Z
|
,
g
r
ea
ter
th
a
n
1
.
Acc
u
r
ate
ex
p
o
n
e
n
tiatio
n
o
f
|
Z
|
,
r
esu
lts
in
h
ig
h
er
o
r
d
er
ter
m
s
o
f
T
ay
lo
r
s
e
r
ies wh
ich
r
esu
lts
in
an
o
v
er
f
l
o
w
in
FP
GA
d
u
e
to
f
ac
to
r
ial
te
r
m
s
in
d
en
o
m
in
ato
r
s
in
s
er
ies
r
ep
r
esen
tatio
n
.
Ho
wev
er
,
ac
ce
p
tab
le
ac
cu
r
ac
y
with
f
ew
ter
m
s
in
s
er
ies is
o
b
tain
ed
wh
en
|
Z
|
<
1
.
Hen
ce
,
th
e
ex
p
o
n
en
tial
f
u
n
c
tio
n
ap
p
ea
r
in
g
in
(
2
)
is
r
ec
ast
as
in
(
3
)
an
d
(
4
)
to
f
a
cilitate
it
s
im
p
lem
en
tatio
n
in
FP
GA.
=
(
3
)
=
+
(
4
)
Ar
g
u
m
en
t
Z
is
s
p
lit
in
to
a
a
n
d
b
,
s
u
ch
t
h
at
-
1
<=
<=
1
an
d
b
is
an
in
teg
er
;
is
a
s
et
o
f
p
r
e
d
e
f
in
ed
co
n
s
tan
ts
(B
j
)
s
to
r
ed
in
lo
o
k
u
p
tab
le
(
L
UT
)
.
T
er
m
is
o
b
tain
ed
b
y
T
ay
lo
r
s
er
ies u
s
in
g
(
5
)
.
=
1
+
(
1
+
(
1
+
(
2
+
(
3
+
(
4
+
5
)
)
)
)
(
5
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Reco
n
f
ig
u
r
a
b
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4
8
6
4
N
eu
r
a
l n
et
imp
leme
n
ta
tio
n
o
f
s
tea
m
p
r
o
p
erti
es o
n
F
P
GA
(
R.
V.
S.
K
r
is
h
n
a
Du
tt
)
189
wh
er
e
1
to
5
ar
e
co
n
s
tan
ts
r
ep
r
e
s
en
tin
g
(
1
/2
!
)
to
(
1
/6
!
)
.
T
h
e
a
b
o
v
e
im
p
le
m
en
tatio
n
in
v
o
lv
es
m
u
ltip
ly
,
ad
d
an
d
c
o
m
p
ar
at
o
r
b
lo
ck
s
.
R
eL
u
,
(
6
)
,
wh
ich
is
an
o
th
er
o
p
ti
o
n
,
wid
el
y
u
s
ed
in
n
e
u
r
al
n
e
ts
,
in
v
o
lv
es
o
n
ly
a
co
m
p
ar
ato
r
.
I
t
is
s
im
p
le
an
d
r
eq
u
ir
es
o
n
ly
a
co
m
p
ar
at
o
r
as
o
p
p
o
s
ed
to
m
an
y
ad
d
er
s
,
m
u
ltip
lier
s
an
d
co
m
p
ar
ato
r
s
in
s
ig
m
o
id
.
=
m
ax
(
0
,
)
(
6
)
Sav
in
g
in
d
ev
ice
u
tili
za
tio
n
is
s
ee
n
in
R
eL
U
a
s
co
m
p
ar
ed
to
s
ig
m
o
id
v
id
e
T
ab
le.
1
.
Ho
wev
er
,
R
eL
u
is
le
s
s
ac
cu
r
ate
as
s
h
o
wn
in
Fig
u
r
e
4
.
Hen
ce
,
Steam
Net
r
eg
r
ess
io
n
is
im
p
lem
en
ted
u
s
in
g
s
ig
m
o
i
d
.
Ma
r
g
in
al
g
ain
in
d
ev
ice
u
tili
za
tio
n
in
Steam
Net
ca
n
b
e
ac
h
iev
ed
b
y
r
e
d
u
cin
g
th
e
n
u
m
b
er
o
f
ter
m
s
in
(
5
)
,
v
id
e
Fig
u
r
e
5.
Fig
u
r
e
2
.
Steam
Net
f
o
r
d
if
f
e
r
e
n
t w
ater
-
s
team
zo
n
es
Fig
u
e
3
.
Steam
Net
ar
ch
itectu
r
e
T
ab
le1
.
Dev
ice
u
ltil
izatio
n
s
u
m
m
ar
y
(
esti
m
ated
v
al
u
es)
Lo
g
i
c
u
t
i
l
i
z
a
t
i
o
n
U
sed
A
v
a
i
l
a
b
l
e
U
t
i
l
i
z
a
t
i
o
n
A
c
t
.
F
u
n
c
t
i
o
n
S
i
g
m
o
i
d
R
e
L
u
S
i
g
m
o
i
d
R
e
L
u
S
i
g
m
o
i
d
R
e
L
u
N
o
.
S
l
i
c
e
LU
Ts
8
5
6
7
31
1
5
0
7
2
0
1
5
0
7
2
0
5%
0%
N
o
.
LU
T
-
F
F
p
a
i
r
s
0
0
8
5
6
7
31
0%
0%
N
o
B
o
n
d
e
d
I
O
B
s
64
64
6
0
0
6
0
0
1
0
%
1
0
%
Fig
u
r
e
4
.
R
eg
r
ess
io
n
er
r
o
r
-
Sig
m
o
id
v
s
R
eL
u
Fig
u
r
e
5
.
T
s
at
er
r
o
r
with
tr
u
n
c
ated
Sig
m
o
id
Fig
u
r
e
6
s
h
o
ws
im
p
lem
en
tatio
n
o
f
(
5
)
u
s
in
g
g
en
e
r
ic
f
lo
atin
g
-
p
o
in
t
u
n
it
(
FP
U)
,
b
ased
o
n
th
e
I
E
E
E
7
5
4
3
2
-
b
it
f
lo
atin
g
f
o
r
m
at.
FP
U
p
r
o
v
id
es
o
p
e
r
atio
n
s
lik
e
ADD,
MU
L
T
I
PLY,
DI
VI
DE
an
d
-20
-10
0
10
0
50
100
150
200
%e
r
r
o
r
(
T
s
a
t
)
Pr(kg/cm
2
)
S
i
g
mo
i
d
R
e
Lu
0
0,5
1
0
50
100
150
200
%e
r
r
o
r
(
T
s
a
t
)
P
r
(
k
g
/
c
m
2
)
6
t
h
O
r
d
e
r
5
t
h
O
r
d
e
r
4
t
h
O
r
d
e
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
20
89
-
4
8
6
4
I
n
t J Reco
n
f
ig
u
r
a
b
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
10
,
No
.
3
,
No
v
em
b
er
2
0
2
1
:
18
6
–
19
4
190
SUB
T
R
AC
T
.
T
h
e
Steam
Net
r
eq
u
ir
es
f
o
u
r
o
r
f
i
v
e
co
ef
f
icien
ts
,
d
ep
en
d
in
g
o
n
ac
cu
r
ac
y
f
o
r
an
d
n
in
e
co
n
s
tan
ts
f
o
r
.
Fig
u
r
e
7
s
h
o
ws
th
e
p
er
ce
n
t
er
r
o
r
i
n
s
ig
m
o
id
i
m
p
lem
en
ted
in
Ver
ilo
g
as
p
er
(
5
)
as
co
m
p
a
r
ed
to
s
tan
d
ar
d
n
u
m
er
ical
lib
r
a
r
y
f
u
n
ctio
n
s
.
Fig
u
r
e
6
.
E
x
p
.
f
u
n
ctio
n
with
l
ar
g
e
ar
g
u
m
en
ts
Fig
u
r
e
7
.
Sig
m
o
id
er
r
o
r
in
v
er
i
lo
g
HDL
3.
ST
E
AM
NE
T
I
M
P
L
E
M
E
N
T
AT
I
O
N
O
N
F
P
G
A
All
lay
er
s
o
f
Steam
Net
s
h
o
wn
in
Fig
u
r
e
8
ar
e
im
p
lem
en
ted
b
y
a
s
in
g
le
n
eu
r
o
n
m
o
d
el.
I
n
p
u
ts
to
ea
ch
o
f
th
e
ANNs
ar
e
n
o
r
m
alize
d
to
b
e
in
th
e
r
a
n
g
e
(
0
.
1
–
0
.
9
)
,
b
y
(
7
)
to
av
o
i
d
asy
m
p
to
tic
s
a
tu
r
atio
n
in
s
ig
m
o
id
f
u
n
ctio
n
.
E
q
u
atio
n
(
8
)
de
-
n
o
r
m
alize
s
th
e
o
u
tp
u
ts
to
g
et
ac
t
u
al
v
alu
es in
en
g
in
ee
r
in
g
u
n
its
.
=
0
.
8
(
−
)
(
−
)
+
0
.
1
(
7
)
=
(
−
0
.
1
)
(
−
)
0
.
8
+
(
8
)
Fu
n
ctio
n
P2
T
in
th
e
Steam
Net,
P2
T
with
p
r
ess
u
r
e
(
Pr
-
k
g
/
cm
2
)
as
in
p
u
t
an
d
s
atu
r
atio
n
t
em
p
er
atu
r
e
(T
sat
-
o
C
)
as
o
u
tp
u
t
u
s
in
g
a
cu
s
to
m
th
r
ee
-
lay
er
n
etwo
r
k
is
r
ea
lized
u
s
in
g
Xilin
x
I
SE
Desig
n
s
u
ite
an
d
Ver
ilo
g
HDL
with
Vir
tex
f
am
ily
F
PG
A.
Fig
u
r
e
9
an
d
Fig
u
r
e
1
0
s
h
o
w
th
e
s
tr
u
ctu
r
al
m
o
d
el
o
f
s
team
Net
u
s
in
g
Xilin
x
I
SE
s
y
n
th
esis
to
o
l
a
n
d
th
e
d
e
v
ice
u
tili
za
tio
n
s
u
m
m
a
r
y
o
f
s
team
n
et
im
p
lem
en
ted
in
Vir
tex
XC
6
VC
X2
4
0
T
FP
GA
b
o
ar
d
,
r
esp
ec
tiv
ely
.
Fig
u
r
e
8
.
No
r
m
alizin
g
a
n
d
De
-
n
o
r
m
alizin
g
(
in
p
u
t
-
o
u
tp
u
t)
-0,
0
1
-0,
0
0
5
0
0
,0
0
5
0
,0
1
-1
-0,5
0
0,5
1
%e
r
r
o
r
A
r
g
u
me
n
t
(
|
a
|
<
=
1
)
o
f
S
i
g
mo
i
d
FPU
FPU
FPU
Pre
v
.
MU
L
ADD
MU
L
a
C
i
B
j
In
Ou
t
Co
n
v
e
rt
to
IEE
E
7
54
De
-
n
o
rm
a
li
z
e
No
rm
a
li
z
e
F
P
U Un
it
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Reco
n
f
ig
u
r
a
b
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4
8
6
4
N
eu
r
a
l n
et
imp
leme
n
ta
tio
n
o
f
s
tea
m
p
r
o
p
erti
es o
n
F
P
GA
(
R.
V.
S.
K
r
is
h
n
a
Du
tt
)
191
Fig
u
r
e
9
.
Steam
Net
s
tr
u
ctu
r
al
m
o
d
el
in
v
er
ilo
g
Fig
u
r
e
10
.
Dev
ice
u
tili
za
tio
n
o
f
Steam
Net
4.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
T
h
e
ab
o
v
e
s
ec
tio
n
d
etails
th
e
im
p
lem
en
tatio
n
o
f
Steam
Net
u
s
in
g
a
s
in
g
le
n
eu
r
o
n
m
o
d
el
wh
er
ein
d
if
f
er
en
t
weig
h
t
m
atr
ices
an
d
b
ias
v
ec
to
r
s
,
at
th
e
t
h
r
ee
l
ay
er
s
o
f
th
e
n
etwo
r
k
in
Fig
u
r
e
3
,
ar
e
s
to
r
ed
in
m
em
o
r
y
.
C
o
m
p
u
tatio
n
o
f
th
e
o
u
tp
u
t
is
ac
h
i
ev
ed
b
y
r
ep
ea
te
d
u
s
ag
e
o
f
s
in
g
le
n
eu
r
o
n
m
o
d
el
im
p
lem
en
ted
as
p
er
Fig
u
r
e
6
a
n
d
Fig
u
r
e
8
.
T
h
e
Fig
u
r
e
1
1
,
Fig
u
r
e
1
2
an
d
Fig
u
r
e
1
3
s
h
o
w
th
e
lay
er
-
1
,
l
ay
er
-
2
an
d
lay
e
r
-
3
in
p
u
t
-
o
u
tp
u
t
s
im
u
latio
n
r
esu
lt
s
r
esp
ec
tiv
ely
,
o
b
tain
ed
f
r
o
m
Xilin
x
I
SE
s
im
u
lato
r
f
o
r
a
t
y
p
ical
n
o
r
m
alize
d
in
p
u
t
o
f
0
.
1
k
g
/cm
2
.
T
h
e
s
im
u
latio
n
s
h
o
ws,
at
tim
e
t=0
,
f
o
r
lay
e
r
-
1
,
a
n
in
p
u
t
v
alu
e
o
f
x
=
ox
h
3
d
cc
cc
cd
p
r
o
d
u
ce
s
o
u
tp
u
ts
at
f
iv
e
n
e
u
r
o
n
s
as
f
x
1
=
ox
3
ee
3
9
af
7
,
f
x
2
=
ox
f
4
ea
1
cf
,
f
x
3
=3
ed
d
f
2
7
0
,
f
x
4
=3
e
d
6
d
ec
e
an
d
f
x
5
=3
f
0
d
5
1
9
5
.
Similar
ly
,
th
e
o
u
tp
u
ts
ar
e
s
h
o
wn
f
o
r
th
e
s
ec
o
n
d
a
n
d
th
ir
d
lay
e
r
s
.
T
h
is
r
esu
lts
in
a
n
o
r
m
alize
d
o
u
tp
u
t
o
f
T
sat
o
f
0
.
1
4
7
9
1
2
8
7
co
r
r
esp
o
n
d
in
g
to
6
3
.
6
9
9
6
9
6
d
eg
C
.
T
h
is
h
as
a
m
ax
im
u
m
er
r
o
r
o
f
0
.
6
1
%
as
s
h
o
wn
in
Fig
u
r
e
14.
Fig
u
r
e
11
.
L
ay
er
-
1
s
im
u
latio
n
r
esu
lts
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
20
89
-
4
8
6
4
I
n
t J Reco
n
f
ig
u
r
a
b
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
10
,
No
.
3
,
No
v
em
b
er
2
0
2
1
:
18
6
–
19
4
192
Fig
u
r
e
12
.
L
ay
er
-
2
s
im
u
latio
n
r
esu
lts
Fig
u
r
e
13
.
L
ay
er
3
s
im
u
latio
n
r
esu
lts
Fig
u
r
e
14
.
T
s
at
E
r
r
o
r
-
Steam
Net
v
s
ASME
s
team
tab
le
5.
CO
NCLU
SI
O
N
A
p
r
e
-
tr
ain
ed
c
o
m
p
ac
t
s
in
g
le
n
eu
r
al
n
et
f
o
r
s
im
u
ltan
eo
u
s
esti
m
atio
n
o
f
s
team
p
r
o
p
er
tie
s
an
d
th
eir
p
ar
tial
d
er
iv
ativ
es
is
tak
en
u
p
f
o
r
im
p
lem
e
n
tatio
n
in
FP
GA
wh
ich
i
s
u
s
ef
u
l
f
o
r
r
ea
l
tim
e
co
n
tr
o
l
an
d
m
o
n
ito
r
in
g
o
f
p
o
wer
p
lan
ts
.
T
h
is
tech
n
iq
u
e
e
n
ab
les
en
tire
s
team
p
r
o
p
e
r
ties
an
d
th
ei
r
d
er
iv
ativ
es
to
b
e
im
p
lem
en
ted
i
n
FP
GA
with
o
u
t
th
e
n
ee
d
f
o
r
p
o
r
tin
g
n
e
ar
ly
5
0
0
0
lin
es
o
f
C
/C
++
co
d
e
.
FP
GA
im
p
lem
en
tatio
n
o
f
o
n
e
n
etwo
r
k
s
h
o
win
g
th
e
r
elatio
n
s
h
ip
b
etwe
en
p
r
ess
u
r
e
an
d
s
atu
r
atio
n
tem
p
er
at
u
r
e
is
d
em
o
n
s
tr
ated
.
Ver
ilo
g
HDL
s
im
u
latio
n
r
esu
l
ts
o
b
tain
ed
f
o
r
XC
6
VC
X2
4
0
T
d
ev
ice
o
f
Vir
tex
f
a
m
ily
with
I
E
E
E
7
5
4
f
lo
atin
g
-
p
o
in
t
3
2
-
b
it
d
ata
r
ep
r
esen
tati
o
n
with
a
n
e
f
f
icien
t
s
ig
m
o
i
d
ac
tiv
atio
n
h
a
v
in
g
lar
g
e
a
r
g
u
m
en
ts
ar
e
p
r
esen
ted
.
E
r
r
o
r
i
n
th
e
s
ig
m
o
id
f
u
n
ctio
n
f
o
r
th
e
ar
g
u
m
en
t
r
a
n
g
e
(
-
30
-
3
0
)
is
with
in
0
.
0
2
%
wh
ich
g
iv
es
a
m
ea
n
ac
cu
r
ac
y
o
f
f
i
n
al
o
u
tp
u
ts
o
f
th
e
Steam
N
et
to
b
e
ar
o
u
n
d
0
.
5
%.
R
esu
lts
o
f
FP
GA
s
im
u
latio
n
o
n
a
tar
g
et
s
y
s
tem
s
h
o
w
th
e
cy
cle
tim
e
is
o
f
th
e
o
r
d
er
o
f
7
0
0
n
s
.
Dev
ice
u
tili
za
tio
n
f
o
r
th
e
n
etwo
r
k
ar
e
p
r
esen
te
d
u
s
in
g
th
e
a
p
p
r
o
ac
h
.
T
h
is
tech
n
iq
u
e
en
a
b
les
to
im
p
lem
e
n
tatio
n
o
f
m
a
n
y
tr
an
s
ce
n
d
en
tal
f
u
n
ctio
n
s
in
FP
GA
with
o
u
t
th
e
n
ee
d
f
o
r
p
o
r
tin
g
th
e
iter
ativ
e
c
o
m
p
lex
co
d
es;
h
en
ce
it
is
e
x
p
ec
ted
m
an
y
r
ea
l
tim
e
ap
p
licatio
n
s
lik
e
e
n
th
alp
y
-
b
ased
s
team
tem
p
er
atu
r
e
c
o
n
tr
o
l i
n
b
o
iler
s
,
o
f
ten
u
s
ed
i
n
p
o
we
r
an
d
p
r
o
c
ess
,
ca
n
b
e
ea
s
ily
d
ev
elo
p
e
d
.
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
10
60
110
160
%E
r
r
o
r
(
T
sa
t
)
P
r
(
k
g
/
c
m
2
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Reco
n
f
ig
u
r
a
b
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4
8
6
4
N
eu
r
a
l n
et
imp
leme
n
ta
tio
n
o
f
s
tea
m
p
r
o
p
erti
es o
n
F
P
GA
(
R.
V.
S.
K
r
is
h
n
a
Du
tt
)
193
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
e
au
th
o
r
s
ac
k
n
o
wled
g
e
th
e
s
u
p
p
o
r
t
a
n
d
en
c
o
u
r
a
g
em
en
t
p
r
o
v
id
e
d
b
y
th
e
m
a
n
ag
em
en
t
o
f
C
VR
C
o
lleg
e
o
f
E
n
g
i
n
ee
r
in
g
,
I
b
r
ah
i
m
p
atn
am
,
Hy
d
er
ab
a
d
,
I
n
d
ia.
RE
F
E
R
E
NC
E
S
[1
]
S
.
G
u
o
,
P
.
Liu
a
n
d
Z
.
Li
,
“
In
e
q
u
a
li
ty
c
o
n
stra
in
e
d
n
o
n
li
n
e
a
r
d
a
ta
r
e
c
o
n
c
il
iatio
n
o
f
a
ste
a
m
tu
r
b
i
n
e
p
o
we
r
p
lan
t
fo
r
e
n
h
a
n
c
e
d
p
a
ra
m
e
ter es
ti
m
a
ti
o
n
,
”
En
e
rg
y
,
v
o
l.
1
0
3
,
p
p
.
2
1
5
–
2
3
0
,
2
0
1
6
,
d
o
i:
1
0
.
1
0
1
6
/j
.
e
n
e
r
g
y
.
2
0
1
6
.
0
2
.
1
5
8
.
[2
]
D.
Wan
g
,
X.
Wu
a
n
d
J
.
S
h
e
n
,
“
An
e
fficie
n
t
r
o
b
u
st
p
r
e
d
icti
v
e
c
o
n
tro
l
o
f
m
a
in
ste
a
m
tem
p
e
ra
tu
re
o
f
c
o
a
l
-
fired
p
o
we
r
p
la
nt
,
”
E
n
e
rg
ies
,
v
o
l
.
13
,
n
o
.
15
,
p
.
3
7
7
5
,
2
0
2
0
,
d
o
i:
1
0
.
3
3
9
0
/en
1
3
1
5
3
7
7
5
.
[3
]
P
.
Nie
g
o
d
a
jew
,
M
.
M
a
re
k
,
W
.
El
sn
e
r
,
a
n
d
Ł
.
K
o
wa
lcz
y
k
,
“
P
o
w
e
r
p
lan
t
o
p
t
imis
a
ti
o
n
e
ffe
c
ti
v
e
u
s
e
o
f
th
e
n
e
ld
e
r
-
m
e
a
d
a
p
p
ro
a
c
h
,
”
Pr
o
c
e
ss
e
s
,
v
o
l.
8
,
n
o
.
3
,
p
.
3
5
7
,
2
0
2
0
,
d
o
i:
1
0
.
3
3
9
0
/p
r
8
0
3
0
3
5
7
.
[4
]
W.
Wag
n
e
r
a
n
d
H.
-
J.
Kre
tzs
c
h
m
a
r
,
In
ter
n
a
ti
o
n
a
l
S
tea
m
T
a
b
les
-
Pro
p
e
rties
o
f
W
a
ter
a
n
d
S
tea
m
b
a
se
d
o
n
th
e
In
d
u
stria
l
Fo
rm
u
l
a
ti
o
n
I
AP
W
S
-
I
F9
7
,
S
p
r
in
g
e
r
S
c
ien
c
e
&
Bu
sin
e
s
s M
e
d
ia
,
2
0
0
7
.
[5
]
M
.
Åb
e
rg
.
,
J
.
Wi
n
d
a
h
l
.
,
H
.
Ru
n
v
i
k
.
,
F
.
M
a
g
n
u
ss
o
n
.
,
“
Op
t
imiz
a
ti
o
n
-
frien
d
l
y
t
h
e
rm
o
d
y
n
a
m
ic
p
r
o
p
e
rti
e
s
o
f
wa
ter
a
n
d
ste
a
m
,
”
Pro
c
e
e
d
in
g
s o
f
th
e
1
2
t
h
I
n
ter
n
a
ti
o
n
a
l
M
o
d
e
li
c
a
C
o
n
fer
e
n
c
e
,
2
0
1
7
,
d
o
i:
1
0
.
3
3
8
4
/ec
p
1
7
1
3
2
4
4
9
.
[6
]
“
G
u
id
e
li
n
e
o
n
th
e
fa
st
c
a
lcu
lati
o
n
o
f
ste
a
m
a
n
d
wa
ter
p
ro
p
e
rti
e
s
with
th
e
s
p
li
n
e
-
b
a
se
d
tab
le
l
o
o
k
-
u
p
m
e
th
o
d
(S
BTL
)
,
”
T
h
e
In
ter
n
a
ti
o
n
a
l
Ass
o
c
ia
ti
o
n
fo
r
t
h
e
Pro
p
e
rties
o
f
W
a
ter
a
n
d
S
te
a
m
,
IAPW
S
G
1
3
-
1
5
,
Ju
l
y
2
0
1
5
.
[On
li
n
e
].
A
v
a
il
a
b
le:
h
tt
p
:/
/www
.
i
a
p
ws
.
o
rg
/rel
g
u
i
d
e
/S
BTL
.
h
tm
l
.
[7
]
P
.
Ca
rd
a
li
a
g
u
e
t
a
n
d
G
.
Eu
v
ra
rd
,
“
Ap
p
ro
x
ima
ti
o
n
o
f
a
f
u
n
c
ti
o
n
a
n
d
it
s
d
e
ri
v
a
ti
v
e
with
a
n
e
u
ra
l
n
e
t
wo
rk
,
”
Ne
u
ra
l
Ne
two
rk
s
,
vol
.
5
,
n
o
.
2
,
p
p
.
2
0
7
-
2
2
0
,
1
9
9
2
,
d
o
i:
1
0
.
1
0
1
6
/S
0
8
9
3
-
6
0
8
0
(
0
5
)
8
0
0
2
0
-
6
.
[8
]
S
.
Ha
sh
e
m
a
n
d
B
.
S
c
h
m
e
ise
r,
“
A
p
p
r
o
x
ima
ti
n
g
a
fu
n
c
ti
o
n
a
n
d
it
s
d
e
riv
a
ti
v
e
s u
sin
g
MSE
-
o
p
ti
m
a
l
li
n
e
a
r
c
o
m
b
in
a
ti
o
n
o
f
train
e
d
fe
e
d
fo
rwa
r
d
n
e
u
ra
l
n
e
t
wo
rk
s
,
”
Pro
c
.
o
f
1
9
9
3
W
o
rl
d
Co
n
g
re
ss
o
n
Ne
u
ra
l
Ne
two
rk
s
,
vol
.
1
,
p
p
.
6
1
7
-
6
2
0
,
1
9
9
3
.
[9
]
R.
V.S
.
Krish
n
a
d
u
tt
a
n
d
J
.
Kris
h
n
a
iah
,
“
Ne
u
ra
l
n
e
ts
f
o
r
th
e
rm
o
d
y
n
a
m
ic
p
r
o
p
e
rti
e
s
”
,
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
Fi
rs
t
In
ter
n
a
t
io
n
a
l
c
o
n
fer
e
n
c
e
o
n
C
o
mp
u
ta
ti
o
n
a
l
I
n
telli
g
e
n
c
e
a
n
d
I
n
f
o
rm
a
ti
c
s,
Ad
v
a
n
c
e
s
in
I
n
telli
g
e
n
t
S
y
ste
ms
a
n
d
Co
mp
u
t
in
g
,
v
o
l
.
5
0
7
,
2
0
1
7
,
d
o
i
:
1
0
.
1
0
0
7
/9
7
8
-
9
8
1
-
10
-
2
4
7
1
-
9
_
6
8
.
[1
0
]
A.
R.
Om
o
n
d
i
a
n
d
J.
C.
Ra
jap
a
k
se
,
“
Ne
u
ra
l
n
e
two
rk
s
in
F
P
G
As
,
”
Pro
c
e
e
d
in
g
s
o
f
th
e
9
t
h
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Ne
u
r
a
l
In
f
o
rm
a
ti
o
n
P
ro
c
e
ss
in
g
,
2
0
0
2
.
ICONIP
'0
2
.
,
2
0
0
2
,
p
p
.
9
5
4
-
9
5
9
v
o
l.
2
,
d
o
i:
1
0
.
1
1
0
9
/ICONIP
.
2
0
0
2
.
1
1
9
8
2
0
2
.
[1
1
]
J.
M
.
G
ra
n
a
d
o
,
M
.
A.
Ve
g
a
,
R.
P
e
re
z
,
J.
M
.
S
a
n
c
h
e
z
a
n
d
J.
A.
G
o
m
e
z
,
“
Us
in
g
F
P
G
As
to
imp
lem
e
n
t
a
rti
ficia
l
n
e
u
ra
l
n
e
tw
o
rk
s,
”
2
0
0
6
1
3
th
I
EE
E
I
n
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
El
e
c
tro
n
i
c
s,
Circ
u
it
s
a
n
d
S
y
ste
ms
,
2
0
0
6
,
p
p
.
9
3
4
-
9
3
7
,
d
o
i
:
1
0
.
1
1
0
9
/IC
ECS
.
2
0
0
6
.
3
7
9
9
4
3
.
[1
2
]
H.
Hu
,
J.
Hu
a
n
g
,
J.
Xin
g
a
n
d
W.
Wan
g
,
“
Ke
y
issu
e
s
o
f
F
P
G
A
im
p
lem
e
n
tatio
n
o
f
n
e
u
ra
l
n
e
two
rk
s
,
”
2
0
0
8
S
e
c
o
n
d
In
ter
n
a
t
io
n
a
l
S
y
mp
o
si
u
m
o
n
In
telli
g
e
n
t
In
fo
rm
a
t
io
n
T
e
c
h
n
o
lo
g
y
A
p
p
li
c
a
ti
o
n
,
2
0
0
8
,
p
p
.
2
5
9
-
2
6
3
,
d
o
i:
1
0
.
1
1
0
9
/II
TA.
2
0
0
8
.
2
3
9
.
[1
3
]
A.
M
u
th
u
ra
m
a
li
n
g
a
m
,
S
.
Him
a
v
a
th
i
a
n
d
E.
S
ri
n
iv
a
sa
n
.
,
“
Ne
u
ra
l
n
e
two
rk
im
p
lem
e
n
tatio
n
u
sin
g
F
P
G
:
A
issu
e
s
a
n
d
a
p
p
li
c
a
ti
o
n
s
,
”
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
,
v
o
l.
2
,
n
o
.
1
2
,
2
0
0
8
,
d
o
i:
1
0
.
5
2
8
1
/ze
n
o
d
o
.
1
0
8
4
4
0
2
.
[1
4
]
J
.
M
isra
a
n
d
I
.
S
a
h
a
,
“
Artifi
c
ial
n
e
u
ra
l
n
e
tw
o
rk
s
i
n
h
a
rd
wa
re
:
A
su
rv
e
y
o
f
tw
o
d
e
c
a
d
e
s
o
f
p
r
o
g
re
ss
,
”
Ne
u
ro
c
o
mp
u
ti
n
g
,
v
o
l
.
74
,
n
o
.
1
-
3
,
p
p
.
2
3
9
-
2
5
5
,
2
0
1
0
,
d
o
i:
1
0
.
1
0
1
6
/
j.
n
e
u
c
o
m
.
2
0
1
0
.
0
3
.
0
2
1
.
[1
5
]
K
.
G
u
o
,
S
.
Zen
g
,
J
.
Yu
,
Y
.
Wan
g
a
n
d
H
.
Ya
n
g
.
,
“
A
su
rv
e
y
o
f
F
P
G
A
b
a
se
d
n
e
u
ra
l
n
e
two
rk
a
c
c
e
lera
to
r
,
”
ACM
T
ra
n
sa
c
ti
o
n
s
o
n
Rec
o
n
f
ig
u
ra
b
le
T
e
c
h
n
o
l
o
g
y
a
n
d
S
y
ste
ms
,
v
o
l.
1
2
,
n
o
.
1
,
p
p
.
1
-
2
6
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
4
5
/3
2
8
9
1
8
5
.
[1
6
]
Z.
Li
,
Y.
Wan
g
,
T.
Z
h
i
a
n
d
T.
C
h
e
n
,
“
A
su
r
v
e
y
o
f
n
e
u
ra
l
n
e
two
r
k
a
c
c
e
lera
to
rs
,
”
Fro
n
ti
e
rs
o
f
Co
mp
u
ter
S
c
ien
c
e
,
v
o
l.
11
,
n
o
.
5
,
p
p
.
7
4
6
–
7
6
1
,
2
0
1
7
,
d
o
i
:
1
0
.
1
0
0
7
/s1
1
7
0
4
-
0
1
6
-
6
1
5
9
-
1
.
[1
7
]
L
.
S
.
M
e
z
h
e
r,
"
Ha
m
m
in
g
n
e
u
ra
l
n
e
two
rk
a
p
p
l
ica
ti
o
n
with
F
P
G
A
d
e
v
ice
,
"
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Rec
o
n
fi
g
u
r
a
b
le
a
n
d
Emb
e
d
d
e
d
S
y
ste
ms
(IJ
RE
S
)
,
v
o
l.
1
0
,
n
o
.
1
,
p
p
.
3
7
-
4
6
,
2
0
2
1
,
d
o
i:
1
0
.
1
1
5
9
1
/
ij
re
s.v
1
0
.
i
1
.
p
p
3
7
-
46
.
[1
8
]
Y.
Tao
,
R.
M
a
,
M
.
S
h
y
u
a
n
d
S
.
Ch
e
n
,
“
Ch
a
ll
e
n
g
e
s
in
e
n
e
rg
y
-
e
ffi
c
ien
t
d
e
e
p
n
e
u
ra
l
n
e
two
rk
trai
n
in
g
w
it
h
F
P
G
A,
”
2
0
2
0
I
EE
E/
CVF
Co
n
fer
e
n
c
e
o
n
Co
mp
u
ter
V
isio
n
a
n
d
Pa
tt
e
rn
Re
c
o
g
n
i
ti
o
n
W
o
rk
sh
o
p
s
(CVP
RW
)
,
2
0
2
0
,
p
p
.
1
6
0
2
-
1
6
1
1
,
d
o
i:
1
0
.
1
1
0
9
/CV
P
RW5
0
4
9
8
.
2
0
2
0
.
0
0
2
0
8
.
[1
9
]
F
.
P
iaz
z
a
,
A.
U
n
c
in
i
a
n
d
M
.
Zen
o
b
i,
“
Ne
u
ra
l
n
e
two
rk
s
with
d
i
g
it
a
l
LUT
a
c
ti
v
a
ti
o
n
f
u
n
c
ti
o
n
s,
”
Pro
c
e
e
d
in
g
s
o
f
1
9
9
3
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
N
e
u
ra
l
Ne
two
rk
s
(IJ
CNN
-
93
-
Na
g
o
y
a
,
J
a
p
a
n
)
,
1
9
9
3
,
p
p
.
1
4
0
1
-
1
4
0
4
v
o
l
.
2
,
d
o
i
:
1
0
.
1
1
0
9
/IJCNN
.
1
9
9
3
.
7
1
6
8
0
6
.
[2
0
]
V
.
Bieu
,
J
.
P
e
p
e
rstra
e
te,
J
.
Va
n
d
e
wa
ll
e
a
n
d
R
.
Lau
we
re
in
s,
“
Clo
se
a
p
p
ro
x
ima
ti
o
n
s
o
f
si
g
m
o
id
fu
n
c
ti
o
n
s
b
y
su
m
o
f
ste
p
s
fo
r
VLS
I
imp
lem
e
n
tati
o
n
o
f
n
e
u
ra
l
n
e
two
r
k
s
,
”
T
h
e
S
c
ie
n
ti
fi
c
An
n
a
ls,
S
e
c
ti
o
n
:
I
n
fo
rm
a
ti
c
s
,
v
o
l
.
4
0
,
n
o
.
1
,
p
p
.
31
-
5
0
,
1
9
9
4
.
[2
1
]
P.
Ly
si
g
h
t
,
J.
Law
a
n
d
D.
G
irma
.
,
“
Artifi
c
ial
n
e
u
ra
l
n
e
t
imp
le
m
e
n
tatio
n
o
n
a
fi
n
e
g
ra
in
e
d
F
P
G
A
,
”
in
Fi
e
ld
-
Pro
g
ra
mm
a
b
le L
o
g
ic A
rc
h
it
e
c
tu
r
e
s,
S
y
n
th
e
sis
a
n
d
A
p
p
li
c
a
ti
o
n
s
,
v
o
l.
8
4
9
,
1
9
9
4
,
1
0
.
1
0
0
7
/3
-
5
4
0
-
5
8
4
1
9
-
6
_
1
2
6
.
[2
2
]
A
.
Ti
sa
n
,
S
.
On
i
g
a
,
D
.
M
ic
a
n
d
A
.
Bu
c
h
m
a
n
,
“
Dig
i
tal
imp
lem
e
n
t
a
ti
o
n
o
f
th
e
si
g
m
o
i
d
fu
n
c
ti
o
n
fo
r
F
P
G
A
c
ircu
it
s
,
”
ACTA
T
ECHNICA
NAP
OCEN
S
I
S
El
e
c
tro
n
ics
a
n
d
T
e
lec
o
mm
u
n
ica
ti
o
n
s
,
v
o
l.
5
0
,
n
o
.
2
,
p
p
.
1
5
-
2
0
,
2
0
0
9
.
[2
3
]
P
.
Do
n
d
o
n
,
J.
Ca
rv
a
l
h
o
,
R.
G
a
rd
e
re
,
P
.
Lah
a
ll
e
,
G
.
Tse
n
o
v
a
n
d
V
.
M
lad
e
n
o
v
,
“
Im
p
lem
e
n
tatio
n
o
f
a
fe
e
d
-
fo
rwa
rd
Artifi
c
ial
Ne
u
ra
l
Ne
two
r
k
in
VH
DL
o
n
F
P
G
A,
”
1
2
th
S
y
mp
o
si
u
m
o
n
Ne
u
ra
l
Ne
tw
o
rk
Ap
p
li
c
a
ti
o
n
s
i
n
El
e
c
trica
l
En
g
i
n
e
e
rin
g
(N
EURE
L
)
,
2
0
1
4
,
p
p
.
37
-
4
0
,
d
o
i:
1
0
.
1
1
0
9
/NEUREL
.
2
0
1
4
.
7
0
1
1
4
5
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
20
89
-
4
8
6
4
I
n
t J Reco
n
f
ig
u
r
a
b
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
10
,
No
.
3
,
No
v
em
b
er
2
0
2
1
:
18
6
–
19
4
194
[2
4
]
S
.
Ng
a
h
,
R
.
A
.
Ba
k
a
r,
A
.
Emb
o
n
g
a
n
d
S
.
Ra
z
a
li
.
,
“
Two
-
ste
p
s
i
m
p
lem
e
n
tatio
n
o
f
si
g
m
o
i
d
fu
n
c
t
i
o
n
fo
r
a
rti
f
icia
l
n
e
u
ra
l
n
e
two
r
k
in
fiel
d
p
r
o
g
ra
m
m
a
b
le
g
a
te
a
rra
y
,
”
AR
PN
J
o
u
rn
a
l
o
f
En
g
i
n
e
e
rin
g
a
n
d
A
p
p
li
e
d
S
c
i
e
n
c
e
s
,
v
o
l.
1
1
,
n
o
.
7
,
p
p
.
4
8
8
2
-
4
8
8
8
,
2
0
1
6
.
[2
5
]
Z.
Li
,
Y.
Hu
a
n
g
a
n
d
W.
Li
n
,
“
F
P
G
A
imp
lem
e
n
tatio
n
o
f
n
e
u
r
o
n
b
l
o
c
k
fo
r
a
rti
ficia
l
n
e
u
ra
l
n
e
two
rk
,
”
2
0
1
7
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
El
e
c
tro
n
De
v
ice
s
a
n
d
S
o
li
d
-
S
t
a
te
Circ
u
it
s
(EDS
S
C)
,
2
0
1
7
,
p
p
.
1
-
2
,
d
o
i
:
1
0
.
1
1
0
9
/
EDS
S
C.
2
0
1
7
.
8
1
2
6
4
31
.
[2
6
]
M
.
Kri
p
s,
T.
Lam
m
e
rt
a
n
d
A.
Ku
m
m
e
rt,
“
F
P
G
A
imp
lem
e
n
tatio
n
o
f
a
n
e
u
ra
l
n
e
two
r
k
f
o
r
a
re
a
l
-
ti
m
e
h
a
n
d
trac
k
in
g
sy
ste
m
,
”
Pro
c
e
e
d
in
g
s
Fi
rs
t
IEE
E
In
ter
n
a
t
io
n
a
l
W
o
rk
sh
o
p
o
n
El
e
c
tr
o
n
ic
De
sig
n
,
T
e
st a
n
d
Ap
p
li
c
a
t
io
n
s
'2
0
0
2
,
2
0
0
2
,
p
p
.
3
1
3
-
3
1
7
,
d
o
i:
1
0
.
1
1
0
9
/DEL
T
A.
2
0
0
2
.
9
9
4
6
3
7
.
[2
7
]
C.
Latin
o
,
M
.
A.
M
o
re
n
o
-
Arm
e
n
d
a
riz
a
n
d
M
.
Ha
g
a
n
,
“
Re
a
li
z
in
g
g
e
n
e
ra
l
M
L
P
n
e
two
rk
s
with
m
in
ima
l
F
P
G
A
re
so
u
rc
e
s,
”
2
0
0
9
I
n
ter
n
a
ti
o
n
a
l
J
o
in
t
Co
n
fer
e
n
c
e
o
n
Ne
u
ra
l
Ne
two
rk
s
,
2
0
0
9
,
p
p
.
1
7
2
2
-
1
7
2
9
,
d
o
i
:
1
0
.
1
1
0
9
/IJCNN
.
2
0
0
9
.
5
1
7
8
6
8
0
.
[2
8
]
D.
Dą
b
ro
ws
k
i,
E
.
Ja
m
r
o
a
n
d
W
.
Cio
c
h
,
“
Ha
rd
wa
re
imp
lem
e
n
tatio
n
o
f
a
rti
ficia
l
n
e
u
ra
l
n
e
two
r
k
s
fo
r
v
ib
r
o
a
c
o
u
stic
sig
n
a
ls cla
ss
ifi
c
a
ti
on
,
”
ACT
A
PH
Y
S
ICA
PO
L
ONICA
A
,
v
o
l.
1
1
8
,
n
o
.
1
,
p
p
.
41
-
44
,
2
0
1
0
.
[2
9
]
O
.
P
o
lat
a
n
d
T
.
Yil
d
iri
m
,
“
F
P
G
A
imp
lem
e
n
tatio
n
o
f
a
g
e
n
e
ra
l
r
e
g
re
ss
io
n
n
e
u
ra
l
n
e
tw
o
rk
:
a
n
e
m
b
e
d
d
e
d
p
a
tt
e
rn
c
las
sifica
ti
o
n
sy
ste
m
,
”
Dig
it
a
l
S
ig
n
a
l
Pro
c
e
ss
in
g
,
v
o
l.
2
0
,
no
.
3,
p
p
.
8
8
1
-
886
,
2
0
1
0
,
d
o
i:
1
0
.
1
0
1
6
/
j.
d
s
p
.
2
0
0
9
.
1
0
.
0
1
3
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
R.
V.
S
.
K
r
ish
n
a
Dutt
is
c
u
rr
e
n
tl
y
p
ro
fe
ss
o
r
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
CVR
Co
ll
e
g
e
o
f
E
n
g
i
n
e
e
rin
g
,
Hy
d
e
ra
b
a
d
,
In
d
ia.
He
wo
rk
e
d
e
a
rli
e
r
a
s
G
e
n
e
r
a
l
M
a
n
a
g
e
r,
Co
rp
o
ra
te
Re
se
a
rc
h
&
De
v
e
lo
p
m
e
n
t,
BHE
L,
Hy
d
e
ra
b
a
d
,
In
d
ia.
He
h
a
d
su
c
c
e
ss
fu
ll
y
led
i
n
terd
isc
ip
l
in
a
ry
e
n
terp
rise
l
e
v
e
l
d
e
v
e
lo
p
m
e
n
t
p
r
o
jec
ts
in
th
e
las
t
fo
u
r
d
e
c
a
d
e
s.
Cu
rre
n
tt
l
y
fo
c
u
se
d
o
n
d
r
iv
i
n
g
in
n
o
v
a
ti
o
n
a
n
d
sta
rt
u
p
s
wi
th
g
o
v
e
rn
m
e
n
t
f
u
n
d
i
n
g
.
He
h
a
s
fe
w
p
u
b
li
c
a
ti
o
n
s
a
n
d
p
a
ten
ts
t
o
h
is
c
re
d
it
.
He
h
o
l
d
s
a
B
.
Tec
h
i
n
El
e
c
tri
c
a
l
e
n
g
i
n
e
e
rin
g
fro
m
NI
T,
Wara
n
g
a
l,
M
S
in
Ap
p
li
e
d
M
e
c
h
a
n
ics
fro
m
IIT
,
Ch
e
n
n
a
i
a
n
d
M
.
Tec
h
in
Co
m
p
u
ter
sc
ien
c
e
a
n
d
e
n
g
i
n
e
e
rin
g
,
U
n
iv
e
rsit
y
Co
ll
e
g
e
o
f
E
n
g
in
e
e
rin
g
,
Os
m
a
n
i
a
Un
iv
e
rsit
y
,
H
y
d
e
ra
b
a
d
,
I
n
d
ia.
He
h
a
s
n
e
a
rly
fo
u
r
d
e
c
a
d
e
s
o
f
e
x
p
e
rien
c
e
re
late
d
t
o
th
e
p
o
we
r
i
n
d
u
str
y
.
His
a
re
a
s
o
f
in
tere
st
a
re
a
p
p
li
c
a
ti
o
n
o
f
M
L
tec
h
n
i
q
u
e
s
to
i
n
d
u
strial
p
r
o
b
lem
s.
R.
G
a
n
e
sh
.
is
c
u
rre
n
tl
y
As
so
c
iate
P
ro
fe
ss
o
r
o
f
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
,
CVR
Co
ll
e
g
e
o
f
E
n
g
i
n
e
e
rin
g
,
H
y
d
e
ra
b
a
d
,
In
d
ia.
He
wo
r
k
e
d
a
s
a
tea
m
m
e
m
b
e
r
in
Re
se
a
rc
h
&
De
v
e
lo
p
m
e
n
t
p
ro
jec
ts
f
u
n
d
e
d
b
y
DRD
O,
ECIL
in
th
e
a
re
a
s
o
f
OFDM
a
n
d
M
u
lt
i
C
h
a
n
n
e
l
An
a
ly
z
e
rs.
He
h
a
s
g
u
id
e
d
1
5
M
.
Tec
h
P
ro
jec
ts
a
n
d
1
9
B
.
Tec
h
.
P
r
o
jec
ts.He
h
a
s
p
u
b
li
sh
e
d
a
t
o
tal
o
f
1
5
j
o
u
r
n
a
l
a
n
d
c
o
n
fe
re
n
c
e
p
a
p
e
rs.
He
h
a
s
c
o
o
r
d
in
a
te
d
a
n
d
c
o
n
d
u
c
ted
1
8
wo
r
k
sh
o
p
s/trai
n
in
g
p
ro
g
ra
m
s
fo
r
th
e
b
e
n
e
fit
o
f
fa
c
u
lt
y
a
n
d
stu
d
e
n
ts
i
n
th
e
fil
e
d
o
f
VL
S
I
u
sin
g
d
i
ffe
re
n
t
EDA
t
o
o
ls.
He
a
lso
d
e
li
v
e
re
d
1
5
train
in
g
se
ss
io
n
s
a
s
a
re
so
u
rc
e
p
e
rso
n
fo
r
VLS
I
wo
rk
sh
o
p
s
a
n
d
train
i
n
g
p
ro
g
ra
m
s.
He
o
b
tain
e
d
B.
Tec
h
d
e
g
re
e
in
El
e
c
tro
n
ics
a
n
d
C
o
m
m
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
,
M
.
Tec
h
in
VLS
I
S
y
ste
m
De
sig
n
fr
o
m
JN
TUH,
Hy
d
e
ra
b
a
d
a
n
d
Ad
v
a
n
c
e
d
P
o
st
G
ra
d
u
a
te
Dip
l
o
m
a
in
VLS
I
fro
m
Ve
d
a
n
t,
S
CL,
Ch
a
n
d
i
g
a
d
h
.
P
re
se
n
tl
y
,
h
e
is
p
u
rsu
i
n
g
P
h
.
D.
fro
m
JN
TUH,
Hy
d
e
ra
b
a
d
.
His
a
re
a
s
o
f
in
tere
st
a
re
th
e
d
e
sig
n
a
n
d
a
p
p
li
c
a
ti
o
n
s
o
f
VLS
I
,
4
G
/5
G
Co
m
m
u
n
ica
ti
o
n
s
a
n
d
AI/M
L
a
lg
o
rit
h
m
s
u
sin
g
VLS
I.
P
.
Pre
m
c
h
a
n
d
o
b
tain
e
d
B.
Tec
h
in
El
e
c
tri
c
a
l
En
g
in
e
e
ri
n
g
fr
o
m
NIT,
Ja
m
sh
e
d
p
u
r,
I
n
d
ia.
M
.
E
.
a
n
d
P
h
.
D
d
e
g
re
e
s
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
S
y
ste
m
s
En
g
i
n
e
e
rin
g
fr
o
m
An
d
h
ra
Un
i
v
e
rsity
,
Visa
k
h
a
p
a
tn
a
m
,
In
d
ia.
He
wo
r
k
e
d
a
s
a
lec
tu
re
r
in
th
e
De
p
t
.
o
f
CS
E,
An
d
h
ra
Un
i
v
e
rsity
.
He
late
r
jo
in
e
d
a
s
As
so
c
iate
p
ro
fe
ss
o
r,
i
n
th
e
De
p
t.
o
f
CS
E,
Un
i
v
e
rsity
C
o
l
leg
e
o
f
E
n
g
i
n
e
e
rin
g
,
Os
m
a
n
ia
Un
iv
e
rsity
,
Hy
d
e
ra
b
a
d
,
In
d
ia
a
n
d
late
r
b
e
c
a
m
e
a
P
ro
fe
ss
o
r
o
f
CS
E
i
n
t
h
e
sa
m
e
c
o
ll
e
g
e
.
He
se
rv
e
d
a
s
Dire
c
to
r
o
f
AICTE,
a
t
Ne
w
De
lh
i.
He
a
lso
se
rv
e
d
a
s
H
e
a
d
,
De
p
t.
o
f
CS
E,
C
h
a
irma
n
Bo
a
rd
o
f
stu
d
ies
in
CS
E,
De
a
n
F
a
c
u
lt
y
o
f
E
n
g
i
n
e
e
rin
g
a
t
U
n
iv
e
rsity
Co
l
leg
e
o
f
E
n
g
in
e
e
rin
g
,
Os
m
a
n
ia
Un
iv
e
rsity
,
H
y
d
e
ra
b
a
d
.
He
is
c
u
rre
n
tl
y
P
ro
fe
ss
o
r,
De
p
t
.
o
f
CS
E
a
n
d
De
a
n
,
F
a
c
u
lt
y
o
f
In
fo
rm
a
ti
c
s a
t
Os
m
a
n
ia Un
iv
e
rsit
y
,
Hy
d
e
ra
b
a
d
,
I
n
d
ia.
He
is a me
m
b
e
r
o
f
t
h
e
se
lec
ti
o
n
c
o
m
m
it
tee
o
f
IS
RO,
NRSA,
AD
RIN
a
n
d
N
G
RI
(
G
o
v
t.
o
f
In
d
ia
Org
a
n
iza
ti
o
n
s).
He
h
a
s
g
u
id
e
d
3
1
D
o
c
to
ra
l
re
se
a
rc
h
sc
h
o
lars
a
n
d
m
a
n
y
a
re
p
u
rsu
i
n
g
P
h
.
D
d
e
g
re
e
s
u
n
d
e
r
h
is
g
u
id
a
n
c
e
.
He
h
a
s
p
re
se
n
ted
se
v
e
ra
l
p
a
p
e
rs i
n
Na
ti
o
n
a
l
a
n
d
i
n
t
e
rn
a
ti
o
n
a
l
C
o
n
fe
re
n
c
e
s a
n
d
jo
u
rn
a
ls.
Evaluation Warning : The document was created with Spire.PDF for Python.