I
nte
rna
t
io
na
l J
o
urna
l o
f
Rec
o
nfig
ura
ble a
nd
E
m
be
dd
e
d Sy
s
t
e
m
s
(
I
J
R
E
S)
Vo
l.
15
,
No
.
2
,
J
u
ly
20
26
,
p
p
.
350
~
3
63
I
SS
N:
2089
-
4864
,
DOI
:
1
0
.
1
1
5
9
1
/i
j
r
es
.
v
1
5
.
i
2
.
pp
3
5
0
-
3
63
350
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
r
es.ia
esco
r
e.
co
m
A
lo
w
-
co
st edge
-
AI
s
m
a
r
t
f
lo
o
r
m
a
t
using
m
ulti
-
po
i
nt
forc
e
sens
o
rs for real
-
ti
m
e
f
a
ll de
tec
tion
a
nd elderly sa
fe
ty
Sa
ha
po
ng
So
m
w
o
ng
1
,
Cha
t
re
e
H
o
m
k
h
iew
2
,
T
ha
nw
it
Na
em
s
a
i
3
,
At
hiro
t
M
a
no
4
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
P
r
i
n
c
e
o
f
S
o
n
g
k
l
a
U
n
i
v
e
r
si
t
y
,
H
a
t
Y
a
i
,
T
h
a
i
l
a
n
d
2
D
e
p
a
r
t
me
n
t
o
f
I
n
d
u
s
t
r
i
a
l
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
R
a
j
a
ma
n
g
a
l
a
U
n
i
v
e
r
si
t
y
o
f
Te
c
h
n
o
l
o
g
y
S
r
i
v
i
j
a
y
a
,
S
o
n
g
k
h
l
a
,
T
h
a
i
l
a
n
d
3
D
e
p
a
r
t
me
n
t
o
f
M
e
c
h
a
n
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
R
a
j
a
ma
n
g
a
l
a
U
n
i
v
e
r
si
t
y
o
f
Te
c
h
n
o
l
o
g
y
S
r
i
v
i
j
a
y
a
,
S
o
n
g
k
h
l
a
,
T
h
a
i
l
a
n
d
4
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
E
n
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
R
a
j
a
m
a
n
g
a
l
a
U
n
i
v
e
r
si
t
y
o
f
T
e
c
h
n
o
l
o
g
y
S
r
i
v
i
j
a
y
a
,
S
o
n
g
k
h
l
a
,
T
h
a
i
l
a
n
d
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
2
1
,
2
0
2
5
R
ev
i
s
ed
Ma
r
2
4
,
2
0
2
6
A
cc
ep
ted
Ma
y
3
1
,
2
0
2
6
T
h
is
stu
d
y
d
e
sc
rib
e
s
th
e
c
re
a
ti
o
n
o
f
a
sm
a
rt
f
lo
o
r
m
a
t
(S
F
M
)
th
a
t
in
teg
ra
tes
e
d
g
e
-
b
a
se
d
a
rti
f
icia
l
in
telli
g
e
n
c
e
(
AI
)
p
ro
c
e
ss
in
g
o
n
a
n
e
m
b
e
d
d
e
d
s
y
ste
m
to
id
e
n
ti
f
y
m
o
v
e
m
e
n
ts
su
c
h
a
s
sta
n
d
i
n
g
,
sitt
in
g
,
a
n
d
f
a
ll
in
g
to
im
p
ro
v
e
th
e
sa
f
e
t
y
o
f
th
e
e
l
d
e
rly
.
T
h
e
d
e
sig
n
i
n
c
o
r
p
o
ra
tes
n
in
e
f
o
rc
e
se
n
siti
v
e
re
sisto
r
(
F
S
R)
se
n
so
rs,
a
n
ES
P
3
2
m
icro
c
o
n
tr
o
ll
e
r,
a
n
d
a
m
u
lt
i
-
c
las
s
su
p
p
o
r
t
v
e
c
to
r
m
a
c
h
in
e
(
S
V
M
)
a
lg
o
rit
h
m
to
a
n
a
ly
z
e
th
e
se
n
so
r
d
a
ta
i
n
re
a
l
ti
m
e
o
r
lo
n
g
-
ti
m
e
imm
o
b
il
it
y
d
e
tec
ti
o
n
,
t
h
e
d
e
v
ice
w
il
l
a
u
to
m
a
ti
c
a
ll
y
s
w
it
c
h
o
n
a
n
d
a
c
ti
v
a
te
a
lar
m
s
to
a
lert
tele
-
c
a
re
g
iv
e
r
s
a
n
d
h
e
lp
e
rs
v
ia
T
e
le
g
ra
m
Bo
t
n
o
ti
f
ica
ti
o
n
s,
in
d
ica
to
r
li
g
h
ts,
a
n
d
s
p
e
a
k
e
rs
f
o
r
imm
e
d
iate
re
sp
o
n
se
s.
Ex
p
e
rime
n
tal
re
su
lt
s
d
e
m
o
n
stra
ted
th
a
t
t
h
e
c
las
sif
ica
ti
o
n
a
c
c
u
ra
c
y
wa
s
9
3
.
3
3
%
in
m
o
d
e
l
e
v
a
lu
a
ti
o
n
a
n
d
8
8
.
3
3
%
o
n
t
h
e
e
m
b
e
d
d
e
d
p
lat
f
o
r
m
,
re
sp
e
c
ti
v
e
l
y
,
w
it
h
a
n
F
1
-
sc
o
re
o
f
0
.
8
2
-
0
.
8
3
a
n
d
a
n
u
tt
e
rly
p
e
rf
e
c
t
f
a
ll
e
v
e
n
t
d
e
tec
ti
o
n
(
1
0
0
%
).
Da
ta
a
re
a
u
t
o
m
a
ti
c
a
ll
y
lo
g
g
e
d
in
G
o
o
g
le
S
h
e
e
ts
th
ro
u
g
h
W
i
-
F
i
f
o
r
tren
d
a
n
a
ly
sis
a
n
d
h
e
a
lt
h
m
o
n
it
o
rin
g
.
T
h
e
p
r
o
p
o
se
d
S
F
M
is
lo
w
-
c
o
st,
f
o
ld
a
b
le,
p
o
rta
b
le,
a
n
d
c
a
p
a
b
le
o
f
su
p
p
o
rti
n
g
re
a
l
-
ti
m
e
m
o
n
it
o
ri
n
g
a
n
d
p
r
o
a
c
ti
v
e
sa
fe
t
y
m
a
n
a
g
e
m
e
n
t
in
t
h
e
e
ld
e
rly
.
T
h
is
in
n
o
v
a
ti
o
n
c
o
n
tri
b
u
tes
to
th
e
d
e
v
e
lo
p
m
e
n
t
o
f
s
m
a
rt
h
o
m
e
h
e
a
lt
h
c
a
re
sy
ste
m
s a
n
d
is i
n
li
n
e
w
it
h
th
e
g
o
a
l
o
f
a
c
h
iev
in
g
a
b
e
tt
e
r
q
u
a
l
it
y
o
f
li
f
e
.
K
ey
w
o
r
d
s
:
E
d
g
e
-
ar
ti
f
icial
i
n
tell
ig
e
n
ce
E
ld
er
ly
m
o
n
i
to
r
in
g
Fall
d
etec
tio
n
I
n
ter
n
et
o
f
t
h
i
n
g
s
h
ea
lth
ca
r
e
S
m
ar
t
f
lo
o
r
m
at
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
:
Sah
ap
o
n
g
So
m
w
o
n
g
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
E
n
g
i
n
ee
r
in
g
,
Fac
u
lt
y
o
f
E
n
g
i
n
ee
r
in
g
,
P
r
in
ce
o
f
So
n
g
k
la
Un
i
v
er
s
i
t
y
Hat
Yai
Dis
tr
ict,
So
n
g
k
h
la
9
0
1
1
0
,
T
h
ailan
d
E
m
ail:
s
a
h
ap
o
n
g
.
s
@
p
s
u
.
ac
.
th
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
W
o
r
l
d
Hea
lth
Or
g
a
n
izati
o
n
(
W
HO)
r
ep
o
r
ts
th
at
p
eo
p
le
ag
ed
6
5
an
d
o
v
er
h
av
e
a
2
8
-
3
5
%
r
is
k
o
f
f
alli
n
g
,
in
cr
ea
s
in
g
to
3
2
-
4
2
%
f
o
r
th
o
s
e
ag
ed
7
0
an
d
o
v
er
.
As
a
r
es
u
lt,
f
all
s
i
n
o
ld
er
ad
u
lts
ar
e
a
g
lo
b
al
h
ea
lt
h
p
r
o
b
lem
th
at
s
i
g
n
i
f
ica
n
tl
y
i
m
p
ac
ts
q
u
alit
y
o
f
li
f
e
an
d
t
h
e
ab
ilit
y
to
p
er
f
o
r
m
d
ail
y
ac
ti
v
it
ies
.
T
h
e
in
cid
en
ce
o
f
f
alls
is
3
0
–
5
0
%
in
t
h
e
co
m
m
u
n
it
y
an
d
a
s
h
i
g
h
as
4
0
%
i
n
n
u
r
s
in
g
h
o
m
es
[
1
]
.
P
r
ev
io
u
s
r
e
s
ea
r
ch
s
u
g
g
es
ts
t
h
at
f
alls
a
n
d
r
elate
d
in
j
u
r
ies
ar
e
a
m
aj
o
r
co
n
tr
ib
u
to
r
to
th
e
d
e
clin
e
in
p
h
y
s
ical
f
u
n
ctio
n
[
2
]
.
I
n
T
h
ailan
d
,
th
e
Dep
ar
t
m
en
t
o
f
Dis
ea
s
e
C
o
n
tr
o
l
s
tates
t
h
at
ap
p
r
o
x
i
m
atel
y
o
n
e
in
t
h
r
ee
o
ld
er
ad
u
lts
,
o
r
ap
p
r
o
x
im
a
tel
y
f
o
u
r
m
illi
o
n
p
eo
p
le,
ex
p
er
ien
ce
f
al
ls
ea
ch
y
ea
r
,
w
it
h
2
0
%
o
f
t
h
ese
in
j
u
r
ies
r
e
s
u
l
tin
g
i
n
d
is
ab
ilit
y
o
r
d
ea
th
.
T
h
is
n
ec
es
s
itates
i
m
p
r
o
v
e
m
e
n
t
s
in
en
v
ir
o
n
m
e
n
tal
s
a
f
et
y
,
p
r
o
m
o
ti
o
n
o
f
ex
er
cise
to
im
p
r
o
v
e
s
tr
en
g
t
h
an
d
b
alan
ce
,
an
d
h
ea
lth
as
s
es
s
m
e
n
t
s
to
ad
d
r
ess
r
i
s
k
f
ac
to
r
s
[
3
]
.
Falls
ar
e
o
f
ten
ca
u
s
ed
b
y
b
o
th
in
ter
n
a
l
f
ac
t
o
r
s
,
s
u
ch
as
m
u
s
cle
w
a
s
ti
n
g
.
Fall
s
ar
e
o
f
ten
ca
u
s
ed
b
y
b
o
th
in
tr
in
s
ic
f
ac
to
r
s
,
s
u
c
h
as
m
u
s
cle
w
as
tin
g
,
d
em
en
tia,
an
d
ch
r
o
n
ic
co
n
d
itio
n
s
,
a
n
d
ex
tr
in
s
ic
f
ac
to
r
s
,
s
u
c
h
as
u
n
e
v
en
s
u
r
f
ac
es
o
r
in
s
u
f
f
icien
t
lig
h
ti
n
g
[
4
]
.
T
h
e
ar
ea
n
ea
r
th
e
b
ed
is
co
n
s
id
er
ed
a
h
i
g
h
-
r
i
s
k
ar
ea
f
o
r
f
alls
in
th
e
e
ld
er
l
y
.
T
h
is
i
s
b
ec
au
s
e
tr
a
n
s
i
tio
n
i
n
g
f
r
o
m
l
y
in
g
d
o
w
n
to
s
tan
d
i
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4864
A
lo
w
-
co
s
t e
d
g
e
-
A
I
s
ma
r
t flo
o
r
ma
t u
s
in
g
mu
lti
-
p
o
in
t fo
r
ce
s
en
s
o
r
s
fo
r
r
ea
l
-
time
…
(
S
a
h
a
p
o
n
g
S
o
mw
o
n
g
)
351
r
eq
u
ir
es
m
u
s
cle
s
tab
ilit
y
a
n
d
b
alan
ce
,
w
h
ic
h
ca
n
b
e
ch
al
len
g
in
g
w
i
th
o
u
t
ap
p
r
o
p
r
iat
e
s
u
p
p
o
r
t
d
ev
ices.
T
h
er
ef
o
r
e,
th
e
d
ev
elo
p
m
e
n
t
o
f
ass
i
s
ti
v
e
tech
n
o
lo
g
ie
s
,
s
u
c
h
as
s
m
ar
t
m
ats
o
r
f
o
r
ce
-
s
e
n
s
in
g
f
lo
o
r
s
y
s
te
m
s
,
is
cr
u
cial
to
d
etec
t,
p
r
ev
en
t,
an
d
r
ed
u
ce
th
e
s
ev
er
it
y
o
f
f
al
ls
.
Su
ch
tech
n
o
lo
g
ies
ca
n
s
i
g
n
if
ican
tl
y
i
m
p
r
o
v
e
s
a
f
et
y
,
p
r
o
m
o
te
in
d
ep
en
d
en
ce
,
a
n
d
im
p
r
o
v
e
th
e
q
u
alit
y
o
f
li
f
e
o
f
th
e
eld
er
ly
.
C
u
r
r
en
tl
y
,
s
en
s
o
r
tech
n
o
lo
g
y
p
la
y
s
a
k
e
y
r
o
le
in
in
tell
ig
e
n
t
elec
tr
o
n
ic
s
y
s
te
m
s
,
p
ar
ticu
lar
l
y
in
h
u
m
a
n
b
eh
av
io
r
m
o
n
ito
r
i
n
g
a
n
d
an
al
y
s
is
[
5
]
.
Fo
r
ce
s
en
s
o
r
s
h
av
e
r
ec
ei
v
ed
s
i
g
n
i
f
ica
n
t
atte
n
tio
n
i
n
h
ea
lt
h
r
esear
ch
d
u
e
t
o
th
eir
ab
ilit
y
to
co
n
ti
n
u
o
u
s
l
y
d
etec
t
an
d
as
s
es
s
m
o
v
e
m
e
n
t
p
atter
n
s
,
w
h
ic
h
ar
e
u
s
ef
u
l
f
o
r
p
r
ed
ictin
g
p
o
ten
tial
h
ea
lth
co
n
d
itio
n
s
in
ad
v
an
ce
[
6
]
,
[
7
]
.
I
n
teg
r
atin
g
s
en
s
o
r
tech
n
o
lo
g
y
w
i
th
t
h
e
in
t
er
n
et
o
f
th
i
n
g
s
(
I
o
T
)
an
d
d
ee
p
lear
n
in
g
d
ata
p
r
o
ce
s
s
in
g
h
as
en
h
a
n
ce
d
r
ea
l
-
ti
m
e
h
ea
lt
h
m
o
n
ito
r
i
n
g
f
o
r
t
h
e
eld
e
r
l
y
[
8
]
–
[
1
1
]
.
Dev
ices
s
u
ch
as
t
h
e
E
SP
3
2
,
R
asp
b
er
r
y
P
i,
an
d
A
r
d
u
i
n
o
h
av
e
b
ee
n
u
s
ed
to
tr
an
s
m
it
h
ea
lth
in
f
o
r
m
atio
n
to
ca
r
eg
i
v
er
s
i
n
a
ti
m
el
y
m
an
n
er
[
1
2
]
,
[
1
3
]
.
Fu
r
th
er
m
o
r
e,
th
e
ap
p
licatio
n
o
f
I
o
T
tech
n
o
lo
g
y
co
m
b
in
ed
w
it
h
ar
tif
icial
in
telli
g
en
ce
(
A
I
)
h
as
in
cr
ea
s
ed
th
e
ac
cu
r
ac
y
o
f
h
ea
lth
d
ata
an
al
y
s
is
,
s
u
p
p
o
r
ted
p
r
o
ac
tiv
e
m
ed
ical
d
ec
is
io
n
-
m
ak
i
n
g
,
an
d
en
ab
led
e
ar
l
y
d
etec
tio
n
o
f
r
is
k
f
ac
to
r
s
[
1
4
]
–
[
1
6
]
.
C
u
r
r
en
t
r
esear
ch
f
o
cu
s
e
s
o
n
d
ev
elo
p
in
g
f
all
d
etec
tio
n
an
d
b
eh
av
io
r
tr
ac
k
i
n
g
s
y
s
te
m
s
f
o
r
th
e
eld
er
ly
u
s
i
n
g
ad
v
an
ce
d
s
e
n
s
o
r
an
d
d
ata
p
r
o
ce
s
s
in
g
tech
n
o
lo
g
ie
s
to
in
cr
ea
s
e
d
etec
tio
n
ac
cu
r
ac
y
a
n
d
r
ed
u
ce
f
alse a
lar
m
r
ates.
E
m
p
lo
y
ed
tech
n
o
lo
g
ie
s
in
cl
u
d
e
f
o
r
ce
s
en
s
o
r
s
,
ac
ce
ler
o
m
e
ter
s
,
s
m
ar
t
f
lo
o
r
s
y
s
te
m
s
,
I
o
T
p
latf
o
r
m
s
,
an
d
A
I
[
1
7
]
–
[
1
9
]
.
Fo
r
ce
s
en
s
o
r
s
an
d
ac
ce
ler
o
m
eter
s
h
a
v
e
b
ec
o
m
e
p
o
p
u
lar
d
u
e
to
th
eir
ab
ilit
y
to
ef
f
icie
n
tl
y
d
etec
t
ab
n
o
r
m
al
m
o
v
e
m
e
n
ts
.
Fo
r
ex
a
m
p
le,
A
l
-
Da
h
an
et
a
l
.
[
1
7
]
u
s
ed
an
MP
U6
0
5
0
s
en
s
o
r
co
u
p
l
ed
w
it
h
a
n
A
r
d
u
i
n
o
m
icr
o
co
n
tr
o
ller
to
m
ea
s
u
r
e
c
h
an
g
e
s
in
o
r
ien
ta
tio
n
an
d
ac
ce
l
er
atio
n
,
w
h
ile
Z
ito
u
n
i
et
a
l
.
[
1
8
]
an
d
L
i
et
a
l
.
[
1
9
]
d
ev
elo
p
ed
a
s
m
ar
t
f
lo
o
r
s
y
s
te
m
eq
u
ip
p
ed
w
it
h
f
o
r
ce
s
en
s
o
r
s
to
au
to
m
at
icall
y
clas
s
i
f
y
eld
er
l
y
b
eh
av
io
r
.
T
h
is
s
ig
n
i
f
ica
n
tl
y
i
m
p
r
o
v
es
t
h
e
ac
c
u
r
ac
y
o
f
f
all
d
etec
tio
n
.
Mi
n
v
i
elle
et
a
l
.
[
2
0
]
an
d
Fen
g
e
t
a
l
.
[2
1]
ex
ten
d
ed
th
i
s
ap
p
r
o
ac
h
b
y
u
s
in
g
a
n
et
w
o
r
k
o
f
g
r
o
u
n
d
-
b
ased
s
e
n
s
o
r
s
to
ac
cu
r
atel
y
d
etec
t t
h
e
lo
ca
tio
n
o
f
f
alls
.
I
n
AI
,
L
u
et
a
l
.
[
2
2
]
w
o
r
k
u
s
ed
3D
-
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
et
wo
r
k
s
(
C
NN
)
a
n
d
lo
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
(
L
ST
M
)
m
o
d
els to
a
n
al
y
ze
v
i
d
eo
t
o
d
is
tin
g
u
i
s
h
f
alls
f
r
o
m
n
o
r
m
al
m
o
tio
n
,
w
h
ile
C
h
an
e
t a
l
.
[
2
3
]
ap
p
lied
d
ee
p
n
eu
r
al
n
e
t
w
o
r
k
s
(
DNN
s
)
to
p
r
o
ce
s
s
d
ata
f
r
o
m
f
o
r
ce
an
d
ac
c
eler
atio
n
s
en
s
o
r
s
.
Fu
r
th
er
m
o
r
e,
W
an
g
et
a
l
.
[
2
4
]
u
s
ed
co
m
p
u
ter
v
is
io
n
tech
n
iq
u
es to
d
etec
t f
alls
f
r
o
m
ca
m
er
a
i
m
a
g
es,
an
d
I
b
r
ah
i
m
et
a
l.
[
2
5
]
d
ev
elo
p
ed
a
b
o
d
y
an
g
le
an
al
y
s
is
a
lg
o
r
it
h
m
to
r
ed
u
ce
th
e
f
al
s
e
alar
m
r
ate.
An
o
th
er
li
n
e
o
f
r
esear
ch
f
o
cu
s
es
o
n
m
o
n
ito
r
i
n
g
s
leep
b
eh
av
io
r
.
Usi
n
g
f
o
r
ce
,
s
tr
ain
,
a
n
d
v
ib
r
atio
n
s
e
n
s
o
r
s
,
f
o
r
ex
a
m
p
le,
Sch
r
e
m
p
f
et
a
l
.
[
2
6
]
an
d
W
altis
b
er
g
et
a
l
.
[
7
]
u
s
ed
f
o
r
ce
s
e
n
s
o
r
s
e
m
b
ed
d
ed
in
b
ed
s
to
d
etec
t
s
leep
b
eh
a
v
io
r
in
r
ea
l
ti
m
e.
K
u
tile
k
et
a
l
.
[
2
7
]
d
ev
elo
p
ed
a
s
tr
ai
n
s
en
s
o
r
f
o
r
an
al
y
zin
g
p
atie
n
t
m
o
v
e
m
e
n
t
,
an
d
Ha
m
za
et
a
l
.
[
2
8
]
u
s
ed
v
ib
r
atio
n
s
e
n
s
o
r
s
co
m
b
in
ed
w
i
th
m
ac
h
i
n
e
lear
n
in
g
tec
h
n
iq
u
es
to
class
if
y
m
o
v
e
m
e
n
t
b
eh
a
v
io
r
.
I
n
I
o
T
tech
n
o
lo
g
y
,
au
to
m
ated
d
etec
tio
n
an
d
aler
tin
g
s
y
s
te
m
s
h
av
e
b
ee
n
u
s
ed
to
m
o
n
ito
r
th
e
b
eh
a
v
io
r
o
f
t
h
e
eld
er
l
y
.
Fo
r
ex
a
m
p
le
,
Sar
a
n
an
an
d
C
h
ai
y
ab
u
t
[
2
9
]
u
s
ed
in
f
r
ar
ed
s
en
s
o
r
s
to
d
etec
t
f
alls
in
h
i
g
h
-
r
i
s
k
ar
ea
s
.
Na
m
k
h
u
n
et
a
l
.
[
3
0
]
u
s
ed
w
ir
el
ess
s
en
s
o
r
s
to
m
o
n
i
to
r
b
ed
g
etti
n
g
o
u
t,
w
h
ile
C
o
cc
o
n
ce
ll
i
et
a
l
.
[
3
1
]
d
ev
elo
p
e
d
an
I
o
T
-
in
teg
r
ated
s
m
ar
t
f
lo
o
r
s
y
s
te
m
to
m
ea
s
u
r
e
u
s
er
w
ei
g
h
t a
n
d
m
o
v
e
m
en
t.
Fro
m
a
liter
atu
r
e
r
ev
ie
w
r
ev
ea
led
th
at
cu
r
r
en
t
r
esear
c
h
tr
en
d
s
f
o
cu
s
o
n
t
h
e
ap
p
licatio
n
o
f
A
I
,
m
ac
h
i
n
e
lear
n
in
g
(
M
L
)
,
an
d
ed
g
e
co
m
p
u
ti
n
g
to
en
h
a
n
ce
r
ea
l
-
ti
m
e
d
ata
an
al
y
s
is
a
n
d
r
ed
u
ce
t
h
e
r
esp
o
n
s
e
ti
m
e
o
f
au
to
m
ated
aler
t
s
y
s
te
m
s
.
R
es
ea
r
ch
er
s
h
av
e
t
h
er
ef
o
r
e
d
ev
elo
p
ed
a
s
y
s
te
m
to
s
u
p
p
o
r
t
d
ai
l
y
liv
i
n
g
at
h
o
m
e
,
s
p
ec
if
icall
y
s
m
ar
t
f
lo
o
r
s
o
r
m
ats
th
at
r
esp
o
n
d
to
th
e
b
eh
a
v
i
o
r
an
d
n
ee
d
s
o
f
ea
ch
eld
er
l
y
p
er
s
o
n
.
Su
c
h
s
y
s
te
m
s
h
elp
r
ed
u
ce
h
ea
lt
h
r
is
k
s
,
in
cr
e
ase
s
af
et
y
d
u
r
i
n
g
m
o
v
e
m
e
n
t,
a
n
d
s
h
o
r
ten
r
esp
o
n
s
e
ti
m
es to
em
er
g
e
n
cies
s
u
c
h
as
f
alls
o
r
s
u
d
d
en
f
all
s
,
v
ia
r
ea
l
-
ti
m
e
aler
ts
to
ca
r
eg
iv
er
s
.
T
h
is
s
ig
n
i
f
ica
n
tl
y
i
m
p
r
o
v
e
s
th
e
q
u
al
it
y
o
f
lif
e
an
d
o
v
er
al
l
ef
f
icien
c
y
o
f
eld
er
l
y
ca
r
e.
T
h
is
r
esear
ch
d
if
f
er
s
f
r
o
m
p
r
ev
io
u
s
s
tu
d
ie
s
th
at
p
r
i
m
ar
il
y
f
o
cu
s
o
n
ca
m
er
a
-
b
ased
o
r
w
ea
r
a
b
le
s
en
s
o
r
s
f
o
r
f
all
d
etec
tio
n
b
y
p
r
esen
t
i
n
g
a
co
s
t
-
ef
f
ec
t
iv
e,
f
o
ld
ab
le
s
m
ar
t
f
lo
o
r
m
a
t
(
SF
M)
th
at
i
n
teg
r
ate
s
n
i
n
e
f
o
r
ce
s
en
s
iti
v
e
r
esi
s
to
r
(
FSR
)
s
en
s
o
r
s
an
d
e
m
b
ed
d
ed
A
I
p
r
o
ce
s
s
i
n
g
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
S
VM
)
d
ir
ec
tl
y
o
n
th
e
E
SP
3
2
m
icr
o
co
n
tr
o
ller
.
T
h
is
ed
g
e
-
AI
ap
p
r
o
ac
h
r
ed
u
ce
s
r
elian
ce
o
n
clo
u
d
,
en
s
u
r
in
g
lo
w
lat
en
c
y
,
h
ig
h
p
r
iv
ac
y
,
an
d
r
ea
l
-
ti
m
e
r
e
s
p
o
n
s
e.
F
u
r
t
h
er
m
o
r
e,
th
e
i
n
te
g
r
atio
n
o
f
I
o
T
aler
ts
(
T
eleg
r
a
m
B
o
t
A
P
I
)
an
d
au
to
m
atic
clo
u
d
d
ata
lo
g
g
i
n
g
(
Go
o
g
le
S
h
ee
ts
)
en
ab
les
co
n
ti
n
u
o
u
s
b
e
h
av
io
r
tr
ac
k
in
g
,
w
h
ic
h
i
s
e
s
s
e
n
tial
f
o
r
d
ev
elo
p
in
g
a
s
ca
lab
le,
in
telli
g
e
n
t e
ld
er
l
y
s
a
f
et
y
s
y
s
te
m
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
2
.
1
.
Sy
s
t
em
s
o
v
er
v
iew
T
h
is
r
esear
ch
is
t
h
e
d
ev
elo
p
m
en
t
o
f
a
f
o
ld
ab
le
s
m
ar
t
f
lo
o
r
p
late,
w
h
ic
h
i
n
teg
r
ate
s
n
i
n
e
F
S
R
s
e
n
s
o
r
s
an
d
e
m
b
ed
d
ed
A
I
p
r
o
ce
s
s
in
g
SVM
d
ir
ec
tl
y
o
n
th
e
E
SP
3
2
m
icr
o
co
n
tr
o
ller
.
I
t
ca
n
d
etec
t
m
o
v
e
m
e
n
t
b
eh
av
io
r
an
d
d
etec
t
f
al
ls
,
s
itti
n
g
,
o
r
s
ta
n
d
in
g
i
n
t
h
e
s
a
m
e
p
o
s
itio
n
f
o
r
a
n
ab
n
o
r
m
a
ll
y
lo
n
g
ti
m
e,
w
h
ic
h
m
a
y
b
e
ca
u
s
ed
b
y
m
u
s
cle
s
tr
ai
n
.
T
h
e
s
y
s
te
m
w
il
l
p
r
o
v
id
e
an
au
d
ib
le
aler
t
an
d
s
en
d
a
n
o
ti
f
icatio
n
to
th
e
ca
r
e
g
iv
er
's
s
m
ar
tp
h
o
n
e
v
ia
th
e
I
n
ter
n
et
s
o
th
at
t
h
e
eld
er
l
y
ca
n
r
ec
eiv
e
t
i
m
e
l
y
as
s
is
ta
n
ce
.
T
h
e
o
v
er
all
s
y
s
te
m
i
s
s
h
o
w
n
i
n
Fi
g
u
r
e
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
9
-
4864
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
15
,
No
.
2
,
J
u
l
y
202
6
:
350
-
3
6
3
352
Fig
u
r
e
1
.
S
y
s
te
m
o
v
er
v
ie
w
T
h
e
SF
M
f
o
r
th
e
eld
er
l
y
i
s
d
es
ig
n
ed
to
d
etec
t,
an
al
y
ze
an
d
al
er
t
ab
o
u
t
p
o
ten
tiall
y
d
an
g
er
o
u
s
b
eh
av
io
r
s
in
r
ea
l ti
m
e.
I
t
w
o
r
k
s
th
r
o
u
g
h
t
h
r
ee
m
ai
n
p
ar
ts
:
i
)
a
r
u
b
b
er
f
lo
o
r
m
at
w
it
h
a
FS
R
s
e
n
s
o
r
,
ii
)
a
co
n
tr
o
l u
n
it u
s
i
n
g
E
SP
3
2
to
p
r
o
ce
s
s
d
ata
w
it
h
ML
alg
o
r
it
h
m
,
an
d
iii
)
a
d
is
p
la
y
an
d
aler
t s
y
s
te
m
v
ia
a
d
is
p
la
y
,
T
eleg
r
am
B
o
t A
P
I
an
d
w
ar
n
i
n
g
s
ig
n
al,
as
s
h
o
w
n
in
Fi
g
u
r
e
2
.
Fig
u
r
e
2
.
S
y
s
te
m
d
ia
g
r
a
m
T
h
e
SF
M
d
etec
ts
th
e
f
o
r
ce
g
e
n
er
ated
b
y
m
o
v
e
m
e
n
t.
I
n
s
id
e
th
e
m
a
t
ar
e
n
in
e
1
.
5
-
i
n
ch
F
S
R
s
en
s
o
r
s
,
w
h
ic
h
ar
e
r
esis
ta
n
ce
s
e
n
s
o
r
s
t
h
at
ca
n
ch
a
n
g
e
t
h
eir
elec
tr
ical
v
alu
e
d
ep
en
d
in
g
o
n
t
h
e
f
o
r
ce
ap
p
lied
.
T
h
ese
s
en
s
o
r
s
ar
e
p
o
s
itio
n
ed
in
s
tr
ate
g
ic
lo
c
atio
n
s
o
n
t
h
e
r
u
b
b
er
f
lo
o
r
m
a
t
to
m
ea
s
u
r
e
th
e
f
o
r
ce
g
en
er
a
ted
w
h
e
n
s
ta
n
d
i
n
g
,
s
itti
n
g
,
o
r
f
alli
n
g
.
T
h
e
d
etec
ted
v
alu
es a
r
e
s
e
n
t to
th
e
co
n
tr
o
l
u
n
it.
T
h
e
co
n
tr
o
l u
n
it th
e
n
co
l
lects,
an
al
y
ze
s
,
an
d
p
r
o
ce
s
s
es
th
e
d
ata
r
ec
eiv
ed
f
r
o
m
t
h
e
FS
R
s
e
n
s
o
r
s
u
s
in
g
AI
al
g
o
r
ith
m
s
o
n
t
h
e
E
S
P
3
2
,
th
e
m
ain
p
r
o
ce
s
s
o
r
.
I
t
ca
n
an
al
y
ze
m
o
v
e
m
en
t
b
eh
a
v
io
r
s
s
u
c
h
as
s
itt
in
g
,
s
ta
n
d
i
n
g
,
o
r
f
al
lin
g
,
an
d
d
etec
t b
e
h
av
io
r
s
t
h
at
m
a
y
b
e
s
i
g
n
s
o
f
an
e
m
er
g
e
n
c
y
,
s
u
c
h
as
f
alli
n
g
w
i
th
o
u
t
co
n
ti
n
u
o
u
s
m
o
v
e
m
e
n
t
o
r
ab
n
o
r
m
al
w
eig
h
t
b
ea
r
in
g
.
T
h
e
co
n
tr
o
l
u
n
it
t
h
en
s
en
d
s
t
h
e
d
ata
to
a
d
is
p
la
y
an
d
n
o
tif
ica
tio
n
s
y
s
te
m
v
ia
a
w
ir
e
less
n
et
w
o
r
k
.
T
h
is
s
y
s
te
m
i
s
in
s
talled
n
e
x
t
t
o
th
e
b
ed
to
d
etec
t
th
e
m
o
v
e
m
en
t
o
f
th
e
eld
er
l
y
p
er
s
o
n
a
s
th
e
y
g
et
o
u
t
o
f
b
ed
.
W
h
en
th
e
eld
er
ly
p
er
s
o
n
ch
an
g
es
f
r
o
m
l
y
in
g
d
o
w
n
to
s
tan
d
in
g
,
th
e
ir
b
o
d
y
m
u
s
t
a
d
j
u
s
t
to
im
m
ed
iatel
y
s
u
p
p
l
y
b
lo
o
d
to
th
e
b
r
ain
.
I
f
th
e
cir
cu
lato
r
y
s
y
s
te
m
d
o
es
n
o
t
r
esp
o
n
d
w
ell,
b
lo
o
d
w
ill
f
l
o
w
d
o
w
n
t
h
e
le
g
s
,
te
m
p
o
r
ar
il
y
r
ed
u
ci
n
g
b
lo
o
d
f
lo
w
to
th
e
b
r
ai
n
,
p
o
ten
tiall
y
ca
u
s
in
g
d
izzi
n
es
s
o
r
u
n
co
n
s
cio
u
s
n
es
s
.
T
h
is
s
y
s
te
m
i
s
a
p
r
o
to
ty
p
e
th
at
d
etec
ts
m
o
v
e
m
en
t b
eh
av
io
r
an
d
aler
ts
w
h
e
n
th
e
s
en
s
o
r
d
etec
ts
f
o
r
ce
f
o
r
a
p
r
o
lo
n
g
ed
p
er
io
d
.
I
t
is
as
s
u
m
ed
t
h
at
t
h
e
eld
er
l
y
p
er
s
o
n
m
a
y
f
all
to
th
e
f
lo
o
r
o
r
ex
p
er
ien
ce
s
p
a
s
ticit
y
a
n
d
b
e
u
n
ab
le
to
m
o
v
e
w
h
e
n
s
tep
p
in
g
o
n
th
e
f
lo
o
r
.
T
h
e
s
y
s
t
e
m
w
ill
s
en
d
an
alar
m
v
ia
a
lo
u
d
s
p
ea
k
er
,
b
u
zz
er
,
an
d
T
eleg
r
a
m
B
o
t
ap
p
licatio
n
to
in
f
o
r
m
t
h
e
eld
er
l
y
ca
r
eg
iv
e
r
an
d
p
r
o
v
id
e
ti
m
el
y
as
s
is
tan
c
e.
T
h
e
s
y
s
te
m
co
n
s
is
t
s
o
f
a
r
u
b
b
er
f
lo
o
r
m
at
w
it
h
an
FS
R
s
e
n
s
o
r
,
a
v
o
ltag
e
co
m
p
ar
ato
r
cir
cu
it b
o
x
,
an
d
a
co
n
tr
o
l b
o
x
,
as sh
o
w
n
i
n
Fi
g
u
r
e
3.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4864
A
lo
w
-
co
s
t e
d
g
e
-
A
I
s
ma
r
t flo
o
r
ma
t u
s
in
g
mu
lti
-
p
o
in
t fo
r
ce
s
en
s
o
r
s
fo
r
r
ea
l
-
time
…
(
S
a
h
a
p
o
n
g
S
o
mw
o
n
g
)
353
Fig
u
r
e
3
(
a)
s
h
o
w
s
a
m
o
d
el
o
f
th
e
s
y
s
te
m
in
s
tallatio
n
,
co
n
s
i
s
tin
g
o
f
a
p
r
ess
u
r
e
-
s
e
n
s
iti
v
e
r
u
b
b
er
p
a
d
w
it
h
a
n
FS
R
s
e
n
s
o
r
,
a
p
r
ess
u
r
e
co
m
p
ar
ato
r
cir
cu
it
b
o
x
,
an
d
a
m
ain
co
n
tr
o
l
b
o
x
.
Fig
u
r
e
3
(
b
)
s
h
o
w
s
th
e
ac
t
u
al
p
r
o
to
ty
p
e
o
f
th
e
d
ev
elo
p
ed
s
y
s
te
m
.
W
h
e
n
an
eld
er
l
y
p
er
s
o
n
s
tep
s
o
u
t
o
f
b
ed
,
th
e
s
en
s
o
r
d
etec
ts
th
e
p
r
ess
u
r
e
an
d
s
en
d
s
t
h
e
d
ata
to
th
e
co
n
t
r
o
l
u
n
it
to
an
al
y
ze
t
h
e
m
o
v
e
m
en
t
p
atter
n
.
I
f
co
n
tin
u
o
u
s
p
r
ess
u
r
e
is
d
etec
ted
f
o
r
an
u
n
u
s
u
all
y
lo
n
g
p
er
io
d
,
th
e
s
y
s
te
m
w
i
ll
s
e
n
d
an
aler
t
s
i
g
n
al
v
ia
s
p
ea
k
er
,
b
u
zz
er
,
a
n
d
T
el
eg
r
a
m
B
o
t
to
n
o
tif
y
th
e
ca
r
eg
i
v
er
f
o
r
ti
m
el
y
a
s
s
i
s
ta
n
ce
.
(
a)
(
b
)
Fig
u
r
e
3
.
Stru
ct
u
r
al
d
esig
n
: (
a
)
s
tr
u
ctu
r
al
d
esi
g
n
o
f
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
a
n
d
(
b
)
p
r
o
to
ty
p
e
o
f
th
e
d
ev
elo
p
ed
s
y
s
te
m
T
o
ass
e
m
b
le
th
e
SF
M
to
s
u
p
p
o
r
t
th
e
n
in
e
FS
R
s
e
n
s
o
r
s
,
w
e
d
esig
n
ed
a
s
tab
le
s
tr
u
ctu
r
e
an
d
ef
f
icie
n
t
s
ig
n
al
d
etec
tio
n
.
T
w
o
ca
r
p
et
m
ats
w
it
h
a
w
id
th
o
f
2
4
cm
,
a
len
g
t
h
o
f
1
1
1
cm
,
an
d
a
th
ick
n
es
s
o
f
3
m
m
w
er
e
p
r
ep
ar
e
d
an
d
ass
e
m
b
led
.
T
h
en
,
th
e
y
w
er
e
d
iv
id
ed
in
to
th
r
ee
eq
u
al
p
ar
ts
an
d
cu
t
in
to
j
ig
s
aw
p
u
zz
l
e
p
iece
s
f
o
r
ea
s
y
co
n
n
ec
t
io
n
.
T
h
r
ee
1
.
5
-
in
ch
F
SR
s
e
n
s
o
r
s
w
er
e
in
s
talle
d
in
t
h
ese
p
o
s
itio
n
s
.
I
n
ad
d
iti
o
n
,
th
e
w
ir
es
w
er
e
p
lace
d
in
th
e
ce
n
ter
o
f
t
h
e
SF
M,
an
d
th
e
y
w
er
e
co
n
n
ec
ted
t
o
th
e
v
o
ltag
e
co
m
p
ar
i
s
o
n
cir
c
u
it
b
o
x
w
it
h
a
6
-
p
in
p
lu
g
,
a
s
s
h
o
w
n
i
n
Fi
g
u
r
e
4
.
T
h
en
,
w
e
p
lace
d
a
1
m
m
th
in
a
cr
y
lic
s
h
ee
t
o
n
t
h
e
s
en
s
o
r
s
to
i
m
p
r
o
v
e
t
h
e
f
o
r
ce
d
etec
tio
n
p
er
f
o
r
m
an
ce
o
f
th
e
FS
R
s
en
s
o
r
s
,
as th
e
t
h
i
n
ac
r
y
li
c
s
h
ee
t
h
elp
s
to
d
is
tr
ib
u
te
t
h
e
f
o
r
ce
ev
en
l
y
.
Fig
u
r
e
4
.
FS
R
s
e
n
s
o
r
in
s
talled
o
n
r
u
b
b
er
f
lo
o
r
m
at
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
9
-
4864
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
15
,
No
.
2
,
J
u
l
y
202
6
:
350
-
3
6
3
354
T
h
e
f
o
r
ce
d
ata
d
etec
ted
b
y
th
e
n
in
e
FS
R
s
e
n
s
o
r
s
i
s
tr
an
s
m
itte
d
to
a
v
o
ltag
e
co
m
p
ar
ato
r
b
o
x
lo
ca
ted
at
th
e
e
n
d
o
f
t
h
e
SF
M,
w
h
ic
h
co
n
v
er
t
s
t
h
e
a
n
alo
g
s
i
g
n
al
s
f
r
o
m
th
e
s
en
s
o
r
s
i
n
to
d
ig
i
tal
s
i
g
n
al
s
.
I
n
s
id
e
th
e
b
o
x
ar
e
n
in
e
v
o
lta
g
e
co
m
p
ar
ato
r
s
d
e
s
i
g
n
ed
u
s
in
g
t
h
e
I
C
L
M3
9
3
,
a
h
i
gh
-
p
r
ec
is
io
n
v
o
lta
g
e
co
m
p
a
r
ato
r
o
p
-
am
p
.
T
h
e
d
ig
ital
d
ata
is
th
e
n
s
en
t
to
a
p
r
o
ce
s
s
o
r
in
th
e
co
n
tr
o
l
b
o
x
to
an
al
y
ze
an
d
d
etec
t
t
h
e
eld
er
l
y
's
m
o
v
e
m
e
n
t
u
s
in
g
ML
al
g
o
r
ith
m
s
,
s
u
ch
as
ab
n
o
r
m
al
w
ei
g
h
t
d
is
tr
ib
u
tio
n
o
r
p
o
s
s
ib
le
f
a
lls
.
W
h
en
s
u
ch
e
v
e
n
ts
ar
e
d
etec
ted
,
th
e
s
y
s
te
m
w
il
l
s
en
d
an
alar
m
v
ia
t
h
e
m
o
n
i
to
r
in
ter
f
ac
e
an
d
T
eleg
r
a
m
B
o
t
A
P
I
t
o
im
m
ed
iatel
y
n
o
tify
t
h
e
ca
r
eg
iv
er
.
T
h
e
r
esear
ch
er
s
s
tu
d
ied
th
e
r
esp
o
n
s
e
b
eh
av
io
r
o
f
th
e
FS
R
s
en
s
o
r
w
h
e
n
s
u
b
j
ec
ted
to
ch
an
g
in
g
lo
ad
s
an
d
ev
al
u
ated
t
h
e
r
elatio
n
s
h
ip
b
et
w
ee
n
t
h
e
f
o
r
ce
a
n
d
th
e
ADC
s
i
g
n
al
le
v
el
o
b
tain
ed
f
r
o
m
th
e
v
o
ltag
e
d
i
v
id
er
cir
cu
it.
T
h
e
ex
p
er
im
e
n
t
s
w
er
e
p
er
f
o
r
m
ed
b
y
p
lacin
g
a
p
en
d
u
l
u
m
w
ei
g
h
t
r
an
g
in
g
f
r
o
m
1
to
1
5
k
ilo
g
r
a
m
s
o
n
th
e
FS
R
s
e
n
s
o
r
p
o
s
itio
n
a
n
d
m
ea
s
u
r
in
g
t
h
e
c
h
an
g
e
i
n
t
h
e
ADC
s
ig
n
al
to
u
n
d
er
s
tan
d
t
h
e
lo
w
f
o
r
ce
s
en
s
it
iv
i
t
y
a
n
d
li
m
ita
tio
n
s
o
f
F
SR
.
T
h
e
r
esp
o
n
s
e
b
e
h
av
io
r
o
f
th
e
FS
R
s
e
n
s
o
r
f
r
o
m
th
r
ee
r
ep
ea
ted
ex
p
er
i
m
en
ts
i
s
s
h
o
w
n
in
Fig
u
r
e
5
.
Fig
u
r
e
5
.
R
elatio
n
o
f
w
e
ig
h
t a
n
d
lev
el
o
f
ADC s
ig
n
al
Fig
u
r
e
5
s
h
o
w
s
t
h
at
th
e
A
D
C
v
alu
e
d
ec
r
ea
s
es
w
it
h
i
n
cr
ea
s
i
n
g
w
ei
g
h
t,
w
it
h
a
n
o
n
li
n
ea
r
r
elatio
n
s
h
ip
co
n
s
is
ten
t
w
it
h
t
h
e
ch
ar
ac
ter
i
s
tic
o
f
F
SR
,
w
h
ic
h
i
s
lo
w
s
e
n
s
it
iv
it
y
to
f
o
r
ce
an
d
s
at
u
r
ates
w
it
h
i
n
cr
ea
s
i
n
g
f
o
r
ce
.
T
h
e
s
en
s
o
r
s
ca
n
d
etec
t f
o
r
ce
s
at
d
if
f
er
en
t p
o
s
itio
n
s
an
d
co
n
v
er
t th
e
r
ea
d
o
u
ts
in
to
1
2
-
b
it a
n
alo
g
s
ig
n
al
s
r
an
g
i
n
g
f
r
o
m
0
to
4
0
9
5
.
A
t
th
e
p
o
s
itio
n
w
it
h
t
h
e
h
i
g
h
est
f
o
r
ce
,
th
e
an
a
lo
g
s
i
g
n
al
v
al
u
e
is
t
h
e
lo
w
est,
w
h
ile
at
t
h
e
p
o
s
itio
n
w
it
h
o
u
t
f
o
r
ce
,
th
e
v
al
u
e
is
clo
s
e
to
th
e
h
ig
h
est
(
4
0
9
5
)
.
T
h
is
b
eh
av
io
r
r
ef
lect
s
th
e
li
m
itatio
n
s
o
f
th
e
s
tab
ilit
y
a
n
d
ac
cu
r
ac
y
o
f
a
s
i
n
g
le
s
en
s
o
r
in
esti
m
at
in
g
t
h
e
ac
tu
a
l
f
o
r
ce
.
T
h
er
ef
o
r
e,
r
esear
ch
o
n
SF
M
u
s
es
m
u
ltip
le
m
ea
s
u
r
e
m
e
n
t
p
o
in
ts
to
g
e
th
er
t
o
g
en
er
ate
f
o
r
ce
d
is
tr
ib
u
tio
n
p
atter
n
s
in
s
tead
o
f
r
ely
in
g
o
n
a
s
in
g
le
A
DC
v
alu
e.
T
h
is
d
ata
s
u
p
p
o
r
ts
ed
g
e
-
A
I
p
r
o
ce
s
s
in
g
ap
p
r
o
ac
h
es
th
at
u
s
e
s
p
atial
an
d
te
m
p
o
r
al
ch
ar
ac
ter
is
tics
to
in
cr
ea
s
e
th
e
ac
cu
r
ac
y
o
f
d
etec
ti
n
g
ev
e
n
ts
s
u
ch
a
s
f
all
s
.
2
.
2
.
M
a
chine le
a
rning
m
o
de
ls
dev
elo
p
m
ent
a
nd
i
m
ple
m
e
nta
t
io
n
In
th
is
r
esear
c
h
,
r
esear
ch
er
s
c
h
o
s
e
to
d
ev
elo
p
a
m
u
l
ticlas
s
S
VM
s
m
o
d
el
f
o
r
an
al
y
zi
n
g
th
e
m
o
v
e
m
e
n
t
b
eh
av
io
r
o
f
th
e
eld
er
l
y
.
T
h
is
is
b
ec
au
s
e
SVMs
ar
e
m
o
r
e
s
u
it
ab
le
th
an
o
th
er
m
ac
h
in
e
lear
n
i
n
g
m
o
d
el
s
b
ec
au
s
e
th
e
y
clea
r
l
y
s
ep
ar
ate
i
n
f
o
r
m
ati
o
n
an
d
s
u
p
p
o
r
t
h
i
g
h
-
di
m
e
n
s
io
n
al
d
ata.
T
h
e
y
also
p
er
f
o
r
m
e
f
f
icie
n
tl
y
e
v
en
w
it
h
a
s
m
a
ll
a
m
o
u
n
t
o
f
tr
ain
i
n
g
d
ata,
w
h
ic
h
is
co
n
s
i
s
te
n
t
w
i
th
th
e
li
m
ita
tio
n
s
o
f
d
i
f
f
icu
l
t
d
ata
co
llectio
n
.
Usi
n
g
SVMs
al
lo
w
s
f
o
r
class
if
y
i
n
g
d
if
f
er
en
t
p
o
s
t
u
r
es
in
a
s
i
n
g
le
m
o
d
el
w
it
h
o
u
t
th
e
n
ee
d
to
cr
ea
t
e
m
u
ltip
le
s
ep
ar
ate
m
o
d
el
s
,
s
u
c
h
as
th
e
o
n
e
-
vs
-
o
n
e
o
r
o
n
e
-
vs
-
r
est
m
e
th
o
d
s
in
s
o
m
e
al
g
o
r
ith
m
s
.
I
n
ad
d
itio
n
,
SVMs
ar
e
m
o
r
e
r
esis
ta
n
t
to
o
v
er
f
itti
n
g
t
h
an
m
o
d
els
s
u
ch
a
s
k
-
NN
o
r
d
ec
is
io
n
tr
ee
s
w
h
e
n
t
h
e
d
ata
i
s
n
o
is
y
o
r
co
m
p
le
x
,
e
n
s
u
r
in
g
ac
cu
r
ate
an
d
r
eliab
le
b
eh
a
v
io
r
d
etec
tio
n
in
r
ea
l
-
w
o
r
ld
en
v
ir
o
n
m
e
n
t
s
.
F
u
r
th
er
m
o
r
e,
SV
Ms
a
r
e
less
co
m
p
le
x
an
d
r
eq
u
ir
e
less
co
m
p
u
tin
g
r
eso
u
r
ce
s
,
m
a
k
in
g
t
h
e
m
s
u
itab
le
f
o
r
ap
p
licatio
n
o
n
m
icr
o
co
n
tr
o
ll
er
b
o
ar
d
s
.
Fig
u
r
e
6
s
h
o
w
s
t
h
e
p
r
o
ce
s
s
o
f
d
ev
elo
p
in
g
a
m
u
lt
iclas
s
SVMs
m
o
d
el
to
class
if
y
th
r
ee
t
y
p
e
s
o
f
m
o
v
e
m
en
t
b
eh
av
io
r
s
i
n
th
e
eld
er
l
y
: s
tan
d
i
n
g
,
s
itti
n
g
,
a
n
d
f
alli
n
g
,
b
ased
o
n
th
e
v
a
lu
e
s
m
ea
s
u
r
ed
b
y
t
h
e
F
SR
s
e
n
s
o
r
.
Fro
m
Fig
u
r
e
6
,
it
s
tar
ts
w
it
h
th
e
d
ata
co
llectio
n
p
r
o
ce
s
s
to
co
llect
d
ata
an
d
p
r
e
p
ar
e
th
e
d
ata
to
b
e
tr
an
s
f
o
r
m
ed
in
to
a
f
o
r
m
at
s
u
itab
le
f
o
r
tr
ain
in
g
th
e
SVMs
m
o
d
el.
T
h
en
,
th
e
r
esu
lt
s
ar
e
p
u
t
in
to
th
e
d
ata
tr
an
s
f
o
r
m
atio
n
p
r
o
ce
s
s
to
co
n
s
id
er
th
e
r
elatio
n
s
h
ip
b
et
w
ee
n
th
e
v
al
u
es
m
ea
s
u
r
ed
b
y
t
h
e
s
en
s
o
r
s
an
d
t
h
e
t
y
p
e
o
f
m
o
v
e
m
e
n
t.
T
h
e
r
elatio
n
s
h
i
p
o
b
tain
ed
f
r
o
m
t
h
is
p
r
o
ce
s
s
w
il
l
b
e
u
s
ed
as
d
ata
to
cr
ea
te
an
ML
m
o
d
el
f
o
r
p
r
ed
ictin
g
m
o
v
e
m
e
n
t
i
n
th
e
n
ex
t
SVMs
m
o
d
ellin
g
s
tep
.
T
h
e
r
esu
lti
n
g
SVM
s
m
o
d
el
w
ill
b
e
em
b
ed
d
ed
o
n
th
e
E
SP
3
2
m
icr
o
co
n
tr
o
ller
b
o
ar
d
f
o
r
p
r
o
ce
s
s
in
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4864
A
lo
w
-
co
s
t e
d
g
e
-
A
I
s
ma
r
t flo
o
r
ma
t u
s
in
g
mu
lti
-
p
o
in
t fo
r
ce
s
en
s
o
r
s
fo
r
r
ea
l
-
time
…
(
S
a
h
a
p
o
n
g
S
o
mw
o
n
g
)
355
Fig
u
r
e
6
.
Mu
lti
-
clas
s
SVM
s
m
o
d
elin
g
2
.
2
.
1
.
Da
t
a
c
o
llect
io
n
I
n
th
is
s
tep
,
d
ata
w
a
s
co
llected
to
ev
alu
ate
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
s
e
n
s
o
r
s
a
n
d
r
ec
o
r
d
m
o
v
e
m
en
ts
.
T
h
e
th
r
ee
t
y
p
es
o
f
m
o
v
e
m
e
n
t
u
s
ed
to
co
llect
d
ata:
s
tan
d
in
g
,
s
itti
n
g
,
a
n
d
f
alli
n
g
.
E
ac
h
ca
te
g
o
r
y
h
as
4
0
s
a
m
p
les,
f
o
r
a
to
tal
o
f
1
2
0
s
a
m
p
les
a
s
s
h
o
w
n
i
n
Fi
g
u
r
e
7
.
T
h
e
r
ed
s
q
u
ar
es
in
d
icate
t
h
e
lo
ca
tio
n
s
o
f
th
e
n
i
n
e
FS
R
s
e
n
s
o
r
s
e
m
b
ed
d
ed
w
it
h
i
n
t
h
e
S
m
ar
t
c
ar
p
et.
Dif
f
er
e
n
t
m
o
v
e
m
e
n
ts
g
en
er
ate
d
if
f
er
en
t
p
r
ess
u
r
e
d
is
t
r
ib
u
tio
n
p
atter
n
s
o
n
th
e
s
en
s
o
r
ar
r
a
y
,
w
h
ich
ar
e
r
ec
o
r
d
ed
an
d
u
s
ed
as
in
p
u
t
f
o
r
tr
ai
n
in
g
a
n
d
e
v
al
u
ati
n
g
th
e
p
r
o
p
o
s
ed
ed
g
e
-
A
I
m
o
d
el
.
Fig
u
r
e
7
.
Data
co
llectio
n
g
r
o
u
p
B
y
co
llec
tin
g
f
o
r
ce
m
ea
s
u
r
e
m
en
t
d
ata
f
r
o
m
t
h
r
ee
s
en
s
o
r
s
i
n
t
h
r
ee
p
o
s
tu
r
es
f
r
o
m
120
test
er
s
,
th
e
s
p
atial
d
is
tr
ib
u
tio
n
o
f
f
o
r
ce
ac
r
o
s
s
t
h
e
n
i
n
e
F
SR
s
en
s
o
r
s
r
ev
ea
led
d
if
f
er
en
t
p
atter
n
s
f
o
r
ea
ch
p
o
s
t
u
r
e.
I
n
t
h
e
s
tan
d
i
n
g
p
o
s
tu
r
e,
t
w
o
s
en
s
o
r
s
d
etec
te
d
f
o
r
ce
,
w
it
h
a
n
a
v
er
ag
e
ADC
v
alu
e
o
f
ap
p
r
o
x
i
m
ate
l
y
1527
an
d
a
s
tan
d
ar
d
d
ev
iatio
n
o
f
ap
p
r
o
x
i
m
a
tel
y
4
9
5
.
I
n
th
e
s
ta
n
d
in
g
p
o
s
t
u
r
e,
b
o
th
s
e
n
s
o
r
s
p
r
i
m
ar
il
y
r
ec
ei
v
e
f
o
r
ce
f
r
o
m
b
o
th
f
ee
t,
r
esu
lti
n
g
in
h
i
g
h
f
o
r
ce
d
etec
tio
n
.
I
n
th
e
s
itti
n
g
p
o
s
tu
r
e,
th
r
ee
s
en
s
o
r
s
d
etec
ted
f
o
r
ce
,
w
h
ic
h
is
d
u
e
to
th
e
w
id
e
f
o
r
ce
d
is
tr
ib
u
tio
n
in
t
h
e
s
itti
n
g
p
o
s
tu
r
e.
T
h
e
av
er
ag
e
A
D
C
v
a
lu
e
f
r
o
m
t
h
e
s
en
s
o
r
s
w
a
s
ap
p
r
o
x
i
m
ate
l
y
2219
an
d
a
s
tan
d
ar
d
d
ev
iatio
n
o
f
ap
p
r
o
x
i
m
atel
y
5
3
7
.
T
h
is
in
d
icate
s
t
h
at
in
t
h
e
s
itti
n
g
p
o
s
tu
r
e,
m
o
r
e
f
o
r
ce
is
d
is
tr
ib
u
ted
to
th
e
s
e
n
s
o
r
s
,
ea
c
h
o
f
w
h
ic
h
d
etec
ts
less
f
o
r
ce
th
a
n
i
n
t
h
e
s
tan
d
in
g
p
o
s
t
u
r
e.
I
n
th
e
f
all
in
g
d
o
w
n
p
o
s
t
u
r
e,
t
h
e
f
o
r
ce
is
d
is
tr
ib
u
ted
to
m
o
r
e
th
a
n
6
s
en
s
o
r
s
s
i
m
u
l
tan
eo
u
s
l
y
,
r
es
u
lti
n
g
i
n
a
n
av
er
a
g
e
A
DC
v
al
u
e
o
f
ap
p
r
o
x
im
a
tel
y
2
8
5
6
.
T
h
is
in
d
icate
s
t
h
at
ea
ch
s
en
s
o
r
d
etec
ts
a
lo
w
f
o
r
ce
co
m
p
ar
ed
to
s
tan
d
i
n
g
an
d
s
itti
n
g
,
as
s
h
o
w
n
i
n
f
o
r
ce
s
en
s
o
r
p
r
o
f
ile
o
f
s
ta
n
d
i
n
g
,
s
itti
n
g
a
n
d
f
all
in
g
i
n
Fi
g
u
r
es
8
-
10
,
r
esp
ec
tiv
el
y
.
T
h
ese
p
h
y
s
ica
l
f
ea
t
u
r
es
p
r
o
v
i
d
e
s
tr
o
n
g
d
is
cr
i
m
in
a
tiv
e
p
r
o
p
er
ties
f
o
r
th
e
SV
M
class
if
ier
a
n
d
s
u
p
p
o
r
t
th
e
r
o
b
u
s
t
n
es
s
o
f
th
e
ed
g
e
-
AI
m
eth
o
d
f
o
r
p
o
s
e
class
i
f
icati
o
n
.
T
h
is
in
f
o
r
m
a
tio
n
,
alo
n
g
w
it
h
t
h
e
a
s
s
o
ciate
d
m
o
tio
n
a
n
n
o
tatio
n
s
,
is
u
s
ed
to
d
ev
elo
p
a
m
at
h
e
m
atica
l
m
o
d
el
u
s
in
g
t
h
e
SVM
al
g
o
r
i
th
m
f
o
r
s
u
b
s
eq
u
en
t
class
i
f
icatio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
9
-
4864
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
15
,
No
.
2
,
J
u
l
y
202
6
:
350
-
3
6
3
356
Fig
u
r
e
8
.
Fo
r
ce
s
en
s
o
r
p
r
o
f
ile
o
f
s
ta
n
d
in
g
Fig
u
r
e
9
.
Fo
r
ce
s
en
s
o
r
p
r
o
f
ile
o
f
s
itti
n
g
Fig
u
r
e
10
.
Fo
r
ce
s
en
s
o
r
p
r
o
f
ile
o
f
f
alli
n
g
2
.
2
.
2
.
Da
t
a
t
ra
ns
f
o
rm
a
t
io
n
I
n
th
i
s
s
tep
,
t
h
e
A
DC
s
i
g
n
al
le
v
els
o
b
tain
ed
f
r
o
m
t
h
e
s
e
n
s
o
r
s
w
er
e
an
a
l
y
ze
d
a
n
d
p
r
ep
ar
ed
t
o
d
ev
elo
p
a
m
at
h
e
m
a
tical
m
o
d
el
f
o
r
clas
s
if
ica
tio
n
tr
ai
n
i
n
g
.
I
t
w
a
s
f
o
u
n
d
th
at
t
h
e
p
h
y
s
ical
b
e
h
av
io
r
s
o
f
s
ta
n
d
in
g
,
s
it
tin
g
,
an
d
f
alli
n
g
o
n
th
e
S
FM
af
f
ec
te
d
th
e
n
u
m
b
er
o
f
f
o
r
ce
s
en
s
o
r
s
.
E
s
p
ec
iall
y
in
th
e
s
ta
n
d
in
g
p
o
s
itio
n
,
r
eg
ar
d
less
o
f
th
e
p
o
s
itio
n
o
n
th
e
SF
M,
g
e
n
e
r
all
y
o
n
l
y
1
-
2
s
e
n
s
o
r
s
co
u
ld
d
etec
t
th
e
f
o
r
ce
.
I
n
th
e
s
it
tin
g
p
o
s
itio
n
,
2
-
3
s
en
s
o
r
s
w
er
e
tr
ig
g
er
ed
,
w
h
i
le
in
th
e
ca
s
e
o
f
f
alli
n
g
,
m
o
r
e
th
a
n
3
s
en
s
o
r
s
r
esp
o
n
d
e
d
.
T
h
er
ef
o
r
e,
th
e
A
DC
v
al
u
es
r
ec
o
r
d
ed
f
r
o
m
all
s
e
n
s
o
r
s
u
n
d
er
th
ese
c
o
n
d
itio
n
s
w
er
e
an
al
y
ze
d
to
ca
lcu
late
th
e
m
ea
n
a
n
d
s
tan
d
ar
d
d
ev
iatio
n
as d
ef
i
n
ed
b
y
(
1
)
an
d
(
2
)
,
r
esp
ec
tiv
el
y
,
to
co
n
s
id
er
th
e
r
elatio
n
s
h
ip
b
et
w
ee
n
t
h
e
s
en
s
o
r
ac
tiv
itie
s
an
d
th
e
t
y
p
es
o
f
mo
v
e
m
e
n
t.
T
h
e
r
esu
lts
ar
e
s
h
o
w
n
i
n
Fi
g
u
r
es
11
an
d
12
.
̅
=
∑
=
=
(
1
)
w
h
er
e
is
th
e
n
u
m
b
er
o
f
s
en
s
o
r
s
.
σ
=
√
∑
(
−
̅
)
2
=
1
(
2
)
w
h
er
e:
is
th
e
v
al
u
e
o
f
ea
c
h
d
ata
,
̅
is
th
e
a
v
er
ag
e
o
f
all
d
ata
,
a
n
d
N
is
th
e
to
tal
n
u
m
b
er
o
f
d
at
a
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4864
A
lo
w
-
co
s
t e
d
g
e
-
A
I
s
ma
r
t flo
o
r
ma
t u
s
in
g
mu
lti
-
p
o
in
t fo
r
ce
s
en
s
o
r
s
fo
r
r
ea
l
-
time
…
(
S
a
h
a
p
o
n
g
S
o
mw
o
n
g
)
357
Fig
u
r
e
11
.
A
v
er
a
g
e
o
f
A
D
C
with
m
o
v
e
m
en
t
b
eh
av
io
r
s
Fig
u
r
e
12
.
Stan
d
ar
d
d
ev
iatio
n
o
f
th
e
A
D
C
w
it
h
m
o
v
e
m
e
n
t b
eh
a
v
io
r
s
An
al
y
s
i
s
o
f
t
h
e
m
ea
n
an
d
s
ta
n
d
ar
d
d
ev
iatio
n
o
f
A
D
C
s
ig
n
al
le
v
els
o
b
tain
ed
f
r
o
m
s
en
s
o
r
m
ea
s
u
r
e
m
en
t
s
r
ev
ea
led
th
at
t
h
e
m
ea
n
A
D
C
l
ev
els
co
r
r
esp
o
n
d
in
g
to
f
alli
n
g
w
er
e
s
ig
n
i
f
ica
n
tl
y
lo
w
er
th
a
n
th
o
s
e
f
o
r
s
tan
d
i
n
g
an
d
s
itt
in
g
,
d
u
e
to
th
e
h
i
g
h
er
n
u
m
b
er
o
f
s
e
n
s
o
r
s
d
etec
ti
n
g
f
o
r
ce
in
t
h
e
l
y
i
n
g
p
o
s
itio
n
co
m
p
a
r
ed
to
s
tan
d
in
g
a
n
d
s
itti
n
g
.
I
n
co
n
tr
ast,
t
h
e
m
ea
n
A
D
C
le
v
els
f
o
r
s
ta
n
d
in
g
an
d
s
itti
n
g
w
er
e
r
elativ
el
y
s
i
m
i
lar
,
p
o
s
s
ib
l
y
d
u
e
to
t
h
e
s
i
m
ilar
n
u
m
b
er
o
f
s
en
s
o
r
s
d
etec
tin
g
f
o
r
ce
in
b
o
th
ca
s
es.
Fu
r
th
er
m
o
r
e,
w
h
en
ex
a
m
i
n
i
n
g
th
e
s
tan
d
ar
d
d
ev
iat
io
n
o
f
A
D
C
lev
el
s
,
s
tan
d
i
n
g
h
ad
th
e
h
i
g
h
e
s
t
s
ta
n
d
ar
d
d
ev
iatio
n
,
f
o
llo
w
ed
b
y
s
itti
n
g
,
w
h
ile
f
al
lin
g
h
ad
th
e
lo
w
e
s
t
s
tan
d
ar
d
d
ev
iatio
n
.
T
h
is
r
es
u
l
t
in
d
icate
s
a
m
o
r
e
ev
e
n
d
i
s
tr
i
b
u
tio
n
o
f
f
o
r
ce
in
s
ta
n
d
in
g
a
n
d
a
m
o
r
e
u
n
i
f
o
r
m
d
is
tr
ib
u
tio
n
d
u
r
in
g
f
a
lls
.
T
h
er
ef
o
r
e,
th
e
r
esear
ch
er
s
a
n
al
y
ze
d
th
e
m
ea
n
an
d
s
ta
n
d
ar
d
d
ev
ia
tio
n
o
f
A
D
C
v
al
u
e
s
o
b
tain
ed
f
r
o
m
all
s
e
n
s
o
r
s
to
d
eter
m
i
n
e
th
eir
r
elatio
n
s
h
ip
w
i
t
h
m
o
v
e
m
e
n
t
in
t
h
e
ca
s
es
o
f
s
t
an
d
in
g
,
s
itti
n
g
,
an
d
f
alli
n
g
.
I
t
w
a
s
f
o
u
n
d
th
a
t th
er
e
w
as a
te
n
d
en
c
y
to
s
i
g
n
i
f
ica
n
tl
y
d
is
ti
n
g
u
i
s
h
m
o
v
e
m
e
n
t,
as s
h
o
w
n
in
F
ig
u
r
e
13
.
Fig
u
r
e
1
3
.
R
elatio
n
o
f
a
v
er
ag
e
an
d
s
tan
d
ar
d
d
ev
iatio
n
w
i
th
m
o
v
e
m
e
n
t
b
eh
a
v
io
r
s
2
.
2
.
3
.
Su
pp
o
rt
v
ec
t
o
r
m
a
chi
ne
m
o
dellin
g
Fro
m
a
to
tal
o
f
1
2
0
s
a
m
p
les,
it
is
a
d
ata
s
et
f
o
r
lear
n
i
n
g
to
cr
ea
te
a
m
at
h
e
m
atica
l
m
o
d
el.
T
h
en
,
th
e
d
ata
s
et
f
o
r
lear
n
in
g
is
u
s
ed
to
cr
ea
te
a
SVMs
m
o
d
el
u
s
in
g
one
-
vs
-
o
n
e
(
Ov
O)
,
w
h
ic
h
is
u
s
ed
to
class
i
f
y
3
class
e
s
o
f
d
ata:
C
las
s
1
is
s
ta
n
d
in
g
b
eh
av
io
r
,
C
la
s
s
2
is
s
itti
n
g
b
e
h
av
io
r
,
an
d
C
las
s
3
is
f
a
lli
n
g
b
eh
av
io
r
.
Fro
m
t
h
e
SVMs
m
o
d
el,
3
lin
ea
r
eq
u
atio
n
s
ar
e
o
b
tain
ed
f
o
r
g
r
o
u
p
in
g
(
d
ec
is
io
n
b
o
u
n
d
ar
y
)
b
et
w
ee
n
ea
ch
p
air
o
f
class
es,
w
h
er
e
ea
ch
SV
Ms e
q
u
atio
n
r
ep
r
esen
ts
t
h
e
g
r
o
u
p
b
o
u
n
d
ar
y
b
et
w
ee
n
t
w
o
clas
s
es
u
s
i
n
g
th
e
l
in
ea
r
in
(
3
)
.
(
1
,
2
)
=
1
,
1
+
2
,
2
+
=
0
(
3
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
9
-
4864
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
,
Vo
l.
15
,
No
.
2
,
J
u
l
y
202
6
:
350
-
3
6
3
358
w
h
er
e
:
1
,
2
ar
e
th
e
w
ei
g
h
t
s
o
f
t
h
e
in
d
ep
en
d
en
t
v
ar
iab
le
s
1
,
2
,
is
th
e
co
n
s
tan
t
(
B
ias)
,
t
h
e
p
o
in
t
at
w
h
ic
h
(
1
,
2
)
=0
is
th
e
g
r
o
u
p
li
n
e
,
an
d
t
h
e
v
alu
e
o
f
(
1
,
2
)
=
tells
w
h
ic
h
s
id
e
o
f
th
e
g
r
o
u
p
lin
e
t
h
e
d
ata
lies
o
n
.
F
r
o
m
th
e
lear
n
in
g
d
ata
s
et
to
c
r
ea
te
a
m
a
th
e
m
atica
l
m
o
d
el,
th
e
SVM
s
eq
u
atio
n
i
s
o
b
tain
ed
as
(
4
)
-
(
6
)
:
1
:
2
.
061
1
+
1
.
902
2
−
0
.
432
=
0
(
4
)
2
:
1
.
627
1
+
0
.
487
2
−
0
.
803
=
0
(
5
)
3
:
2
.
128
1
+
0
.
598
2
−
1
.
084
=
0
(
6
)
w
h
er
e
SVM
1
,
2
,
an
d
3
ar
e
lin
ea
r
eq
u
atio
n
s
th
at
d
iv
id
e
C
las
s
1
an
d
C
lass
2
,
C
las
s
1
a
n
d
C
lass
3
,
C
lass
2
an
d
C
las
s
3
,
r
esp
ec
tiv
el
y
.
Fro
m
SVM
1
,
SVM
2
,
an
d
SVM
3
,
w
h
e
n
th
e
v
al
u
es
o
f
1
an
d
2
ar
e
s
u
b
s
titu
ted
in
to
all
th
r
ee
SVMs
eq
u
atio
n
s
,
if
t
h
e
ca
lc
u
lated
v
al
u
es
ar
e
p
o
s
i
tiv
e
o
r
ze
r
o
,
th
e
cl
ass
i
f
icatio
n
a
n
al
y
s
is
ca
n
b
e
p
er
f
o
r
m
ed
as f
o
llo
w
s
:
I
f
SVM
1
is
g
r
ea
ter
th
an
o
r
eq
u
al
to
0
,
it
is
class
if
ied
as
C
la
s
s
1
;
if
it
is
less
th
an
0
,
it
is
class
i
f
ied
as
C
lass
2
.
T
h
e
s
a
m
e
ap
p
lies
to
th
e
v
al
u
e
s
ca
lcu
lated
f
o
r
SVM
2
an
d
SV
M
3
.
W
h
en
m
u
ltip
le
SVM
s
eq
u
atio
n
s
ar
e
u
s
ed
,
w
e
co
m
p
ar
e
th
e
v
alu
es
o
f
ea
ch
f
u
n
ctio
n
to
d
eter
m
i
n
e
w
h
ic
h
class
th
e
d
ata
f
alls
in
to
u
s
i
n
g
a
v
o
tin
g
m
eth
o
d
.
T
h
e
r
esu
lts
o
f
th
e
an
al
y
s
i
s
f
r
o
m
all
th
r
ee
eq
u
atio
n
s
ar
e
u
s
ed
to
d
eter
m
in
e
t
h
e
n
u
m
b
er
o
f
clu
s
t
er
in
g
r
es
u
lt
s
.
−
I
f
SVM
1
id
en
ti
f
ies a
s
C
la
s
s
1
,
SVM
2
id
en
tif
ie
s
as C
lass
1
,
co
n
clu
d
i
n
g
t
h
at
t
h
e
d
ata
ch
o
o
s
es C
las
s
1
.
−
I
f
SVM
1
id
en
ti
f
ies a
s
C
la
s
s
2
,
SVM
3
id
en
tif
ie
s
as C
lass
2
,
co
n
clu
d
i
n
g
t
h
at
t
h
e
d
ata
ch
o
o
s
es C
las
s
2
.
−
I
f
SVM
2
id
en
ti
f
ies a
s
C
la
s
s
3
,
SVM
3
id
en
tif
ie
s
as C
lass
3
,
co
n
clu
d
i
n
g
t
h
at
t
h
e
d
ata
ch
o
o
s
es C
las
s
3
.
T
h
er
ef
o
r
e,
th
e
r
esear
ch
er
u
s
ed
th
e
eq
u
atio
n
s
an
d
co
n
d
itio
n
s
f
o
r
class
class
if
ica
tio
n
to
w
r
ite
a
p
r
o
g
r
am
o
n
E
SP
3
2
to
class
if
y
d
ata
in
r
e
al
ti
m
e
u
s
i
n
g
C
/C
++
la
n
g
u
a
g
e
(
A
r
d
u
i
n
o
I
DE
an
d
E
SP
-
I
DF)
a
n
d
u
s
ed
t
h
e
Vo
ti
n
g
Me
th
o
d
to
s
elec
t th
e
f
i
n
al
clas
s
w
h
ic
h
is
t
h
e
e
n
d
o
f
th
e
p
r
ed
ictio
n
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
is
s
ec
t
io
n
,
th
e
r
esear
ch
p
r
esen
ts
th
e
ex
p
er
i
m
en
ta
l
r
esu
lt
s
an
d
s
y
s
te
m
atica
ll
y
an
al
y
ze
s
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
s
y
s
te
m
.
I
t
s
tar
ts
w
it
h
ev
al
u
ati
n
g
t
h
e
S
VM
m
o
d
el
in
m
o
tio
n
class
if
i
ca
tio
n
to
r
ef
lect
it
s
ac
cu
r
ac
y
o
n
th
e
tr
ain
in
g
d
ata.
T
h
en
,
th
e
r
esu
lts
o
f
t
h
e
r
ea
l s
y
s
te
m
tes
tin
g
ar
e
p
r
esen
ted
,
d
iv
id
ed
in
to
tw
o
p
ar
ts
:
i
)
th
e
p
r
ed
ictio
n
p
er
f
o
r
m
an
ce
o
f
th
e
m
o
d
el
an
d
ii
)
th
e
ex
p
er
im
en
tal
r
esu
lt
s
o
f
th
e
r
ea
l
s
y
s
te
m
p
r
ed
ictio
n
,
w
h
ich
s
h
o
w
t
h
e
s
y
s
te
m
r
esp
o
n
s
e
to
a
b
n
o
r
m
al
ev
e
n
ts
a
n
d
co
llect
b
eh
av
io
r
al
d
ata
f
o
r
r
etr
o
s
p
ec
tiv
e
an
al
y
s
is
.
3
.
1
.
P
re
dict
iv
e
perf
o
r
m
a
nce
o
f
m
o
del
Af
ter
o
b
tain
i
n
g
a
SVM
s
m
o
d
el
f
o
r
class
i
f
y
i
n
g
th
e
m
o
v
e
m
en
t
b
eh
av
io
r
o
f
th
e
eld
er
l
y
at
3
lev
els,
n
a
m
e
l
y
,
s
ta
n
d
in
g
,
s
itti
n
g
an
d
f
alli
n
g
,
i
n
th
i
s
m
et
h
o
d
,
th
e
ef
f
icie
n
c
y
o
f
t
h
e
m
o
d
el
is
test
e
d
w
i
th
4
5
s
a
m
p
les,
d
iv
id
ed
in
to
1
5
s
a
m
p
les
p
er
l
ev
el,
w
it
h
th
e
test
r
esu
lts
a
s
s
h
o
w
n
in
Fi
g
u
r
e
14
a
n
d
t
h
e
r
e
s
u
lt
s
o
f
t
h
e
m
o
d
el
ef
f
icien
c
y
a
n
al
y
s
is
u
s
in
g
co
n
f
u
s
io
n
m
atr
i
x
as
s
h
o
w
n
in
T
ab
le
1
.
Fig
u
r
e
1
4
.
P
r
e
d
ictiv
e
p
er
f
o
r
m
an
ce
o
f
SVM
s
m
u
lt
i
-
cla
s
s
m
o
d
el
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
R
ec
o
n
f
i
g
u
r
ab
le
&
E
m
b
ed
d
ed
Sy
s
t
I
SS
N:
2089
-
4864
A
lo
w
-
co
s
t e
d
g
e
-
A
I
s
ma
r
t flo
o
r
ma
t u
s
in
g
mu
lti
-
p
o
in
t fo
r
ce
s
en
s
o
r
s
fo
r
r
ea
l
-
time
…
(
S
a
h
a
p
o
n
g
S
o
mw
o
n
g
)
359
T
ab
le
1
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
SVMs
m
u
lti
-
clas
s
m
o
d
el
A
c
t
u
a
l
\
p
r
e
d
i
c
t
e
d
S
t
a
n
d
i
n
g
S
i
t
t
i
n
g
F
a
l
l
i
n
g
S
t
a
n
d
i
n
g
13
2
0
S
i
t
t
i
n
g
1
14
0
F
a
l
l
i
n
g
0
0
15
Fro
m
th
is
m
atr
ix
,
a
co
m
p
ar
is
o
n
b
et
w
ee
n
t
h
e
ac
tu
al
an
d
p
r
ed
icted
class
es
o
f
th
e
m
o
d
el
is
s
h
o
w
n
to
ev
alu
a
te
its
ab
ilit
y
to
class
i
f
y
t
h
e
th
r
ee
ca
teg
o
r
ies,
n
a
m
el
y
,
s
tan
d
i
n
g
,
s
it
tin
g
,
an
d
f
allin
g
.
T
h
e
m
o
d
el
d
em
o
n
s
tr
ate
s
h
i
g
h
p
er
f
o
r
m
a
n
ce
w
it
h
an
o
v
er
all
ac
cu
r
ac
y
o
f
ap
p
r
o
x
im
ate
l
y
9
3
.
3
3
%.
T
h
e
an
al
y
s
i
s
o
f
th
e
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
o
f
th
e
co
n
f
u
s
io
n
m
atr
i
x
is
s
h
o
w
n
i
n
T
ab
le
2
.
T
ab
le
2
.
R
esu
lts
o
f
co
n
f
u
s
io
n
m
atr
i
x
an
al
y
s
i
s
C
l
a
ss
P
r
e
c
i
si
o
n
R
e
c
a
l
l
F1
-
sco
r
e
S
t
a
n
d
i
n
g
0
.
9
2
8
6
0
.
8
6
6
7
0
.
8
9
6
8
S
i
t
t
i
n
g
0
.
8
7
5
0
0
.
9
3
3
3
0
.
9
0
3
5
F
a
l
l
i
n
g
1
.
0
0
0
0
1
.
0
0
0
0
1
.
0
0
0
0
Fro
m
th
e
tab
le,
th
e
m
o
d
el
'
s
F
1
-
s
co
r
e
in
all
clas
s
es
e
x
ce
ed
s
0
.
8
9
,
w
h
ic
h
i
s
co
n
s
id
er
ed
an
ac
ce
p
tab
le
lev
el
f
o
r
ac
ti
v
it
y
r
ec
o
g
n
itio
n
r
esear
ch
.
T
h
e
d
ep
r
ess
ed
clas
s
h
as
p
er
f
ec
t
p
r
ec
is
io
n
a
n
d
r
ec
all
o
f
1
.
0
,
w
h
ic
h
is
o
f
g
r
ea
t
ac
ad
e
m
ic
s
i
g
n
if
ican
ce
i
n
th
e
ca
s
e
o
f
d
etec
tin
g
cr
itical
ev
en
ts
(
e.
g
.
,
f
al
ls
)
.
T
h
e
s
tan
d
in
g
class
an
d
t
h
e
s
itti
n
g
class
h
a
v
e
s
i
m
ilar
le
v
els
o
f
p
r
ec
is
io
n
an
d
r
ec
all,
in
d
icati
n
g
th
at
th
e
m
o
d
el
i
s
n
o
t
b
iased
to
w
ar
d
s
an
y
o
f
t
h
e
class
es.
3
.
2
.
E
x
peri
m
ent
a
l r
esu
lt
o
n pre
dict
ing
o
f
i
m
ple
m
e
nte
d s
y
s
t
e
m
3
.
2
.
1
.
P
re
dict
iv
e
perf
o
rm
a
nc
e
o
f
i
m
ple
m
ent
a
t
io
n sy
s
t
e
m
I
n
th
i
s
s
tep
,
th
e
m
u
l
ti
-
c
lass
S
VM
s
m
at
h
e
m
atica
l
m
o
d
el
is
em
b
ed
d
ed
in
t
h
e
E
SP
3
2
m
icr
o
co
n
tr
o
ller
to
p
r
o
ce
s
s
th
e
m
o
tio
n
an
al
y
s
is
i
n
r
ea
l
-
w
o
r
ld
ap
p
licatio
n
s
.
T
h
en
,
its
p
er
f
o
r
m
a
n
ce
i
n
g
es
tu
r
e
cla
s
s
if
ica
tio
n
is
te
s
ted
w
it
h
a
to
tal
o
f
6
0
s
a
m
p
les,
d
i
v
id
ed
in
to
2
0
s
a
m
p
les
p
er
clas
s
.
T
h
e
r
esu
lts
o
f
t
h
e
m
o
tio
n
cl
ass
i
f
icatio
n
test
ar
e
s
h
o
w
n
in
F
ig
u
r
e
1
5
,
an
d
th
e
p
er
f
o
r
m
an
ce
a
n
al
y
s
i
s
r
esu
lts
u
s
in
g
co
n
f
u
s
io
n
m
atr
ix
ar
e
s
h
o
wn
in
T
ab
le
3
.
Fig
u
r
e
1
5
.
P
r
e
d
ictin
g
s
y
s
te
m
p
er
f
o
r
m
an
ce
T
ab
le
3
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
i
m
p
le
m
e
n
tat
io
n
s
y
s
te
m
A
c
t
u
a
l
l
y
\
p
r
e
d
i
c
t
e
d
S
t
a
n
d
i
n
g
S
i
t
t
i
n
g
F
a
l
l
i
n
g
S
t
a
n
d
i
n
g
16
4
0
S
i
t
t
i
n
g
3
17
0
F
a
l
l
i
n
g
0
0
20
An
al
y
s
i
s
o
f
th
e
m
o
d
el
'
s
clas
s
if
icatio
n
r
es
u
lt
s
u
s
i
n
g
th
e
co
n
f
u
s
io
n
m
atr
ix
r
e
v
ea
led
t
h
at
t
h
e
m
o
d
el
co
u
ld
ef
f
icien
tl
y
c
lass
if
y
all
t
h
r
ee
b
eh
av
io
r
s
,
n
a
m
el
y
,
s
ta
n
d
i
n
g
,
s
itti
n
g
,
a
n
d
f
al
lin
g
,
w
it
h
a
n
o
v
er
all
ac
cu
r
ac
y
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.