I
nte
rna
t
io
na
l J
o
urna
l o
f
Adv
a
nces in Applie
d Science
s
(
I
J
AAS)
Vo
l.
10
,
No
.
1
,
Ma
r
ch
2
0
2
1
,
p
p
.
28
~
38
I
SS
N:
2
2
5
2
-
8
8
1
4
,
DOI
: 1
0
.
1
1
5
9
1
/ijaas.v
1
0
.
i
1
.
p
p
28
-
38
28
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
a
s
.
ia
esco
r
e.
co
m
A
3
-
dimens
io
na
l
f
a
st ma
chine learn
ing
alg
o
rithm f
o
r
mo
bile
unma
nned aeria
l
v
ehicle bas
e statio
ns
Wa
s
s
wa
Sh
a
f
ik
1
,
S.
M
o
j
t
a
ba
M
a
t
ink
ha
h
2
,
So
la
g
ba
de
Sa
heed
Af
o
la
bi
3
,
M
a
m
m
a
n Nur
Sa
nd
a
4
1,
2
De
p
a
rtme
n
t
of
Co
m
p
u
ter E
n
g
i
n
e
e
rin
g
,
Ya
z
d
Un
i
v
e
rsity
,
Ira
n
3
De
p
a
rtme
n
t
o
f
M
i
n
e
ra
l
Ch
e
m
istr
y
,
Ya
z
d
Un
i
v
e
rsity
,
Ira
n
4
De
p
a
rtme
n
t
o
f
Co
n
d
e
n
se
d
M
a
tt
e
r
P
h
y
sic
s,
Ya
z
d
Un
i
v
e
rsity
,
Ira
n
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l 1
0
,
2
0
2
0
R
ev
is
ed
Sep
2
,
2
0
2
0
Acc
ep
ted
Sep
29
,
2
0
2
0
Th
e
5
G
tec
h
n
o
lo
g
y
is
p
re
d
icte
d
t
o
a
c
h
iev
e
th
e
u
n
o
p
ti
m
ize
d
m
il
li
m
e
ter
Wav
e
(m
m
Wav
e
)
o
f
3
0
t
o
3
0
0
G
Hz
b
a
n
d
s.
T
h
is
u
n
o
p
ti
m
ize
d
b
a
n
d
is
b
e
c
a
u
se
o
f
t
h
e
lo
ss
o
f
m
m
-
Wav
e
b
a
n
d
s,
li
k
e
p
a
th
a
tt
e
n
u
a
ti
o
n
a
n
d
p
ro
p
a
g
a
t
io
n
lo
ss
e
s.
No
n
e
th
e
les
s,
b
e
c
a
u
se
:
(i)
Dire
c
ti
o
n
a
l
tran
sm
iss
io
n
p
a
v
in
g
wa
y
fo
r
b
e
a
m
fo
rm
in
g
to
re
c
o
m
p
e
n
se
fo
r
th
e
p
a
th
a
tt
e
n
u
a
ti
o
n
;
a
n
d
(ii
)
S
o
p
h
isti
c
a
ted
p
lac
e
m
e
n
t
c
o
n
c
re
ten
e
ss
o
f
t
h
e
b
a
se
sta
ti
o
n
s
(BS
)
is
th
e
b
e
st
a
lt
e
rn
a
ti
v
e
f
o
r
a
rra
y
wire
les
s
c
o
m
m
u
n
ica
ti
o
n
s
i
n
m
m
Wav
e
b
a
n
d
s
(
th
a
t
is
t
o
sa
y
1
0
0
t
o
1
5
0
m
).
Th
e
a
d
v
a
n
c
e
in
tec
h
n
o
lo
g
y
a
n
d
i
n
n
o
v
a
ti
o
n
o
f
u
n
m
a
n
n
e
d
a
e
ria
l
v
e
h
icle
s
(UA
Vs
)
n
e
c
e
ss
it
a
tes
m
a
n
y
o
p
p
o
rtu
n
it
ies
a
n
d
u
n
c
e
rtain
t
ies
.
UA
V
s
a
re
a
g
il
e
a
n
d
c
a
n
fl
y
a
ll
c
o
m
p
lex
it
ies
o
f
th
e
terra
in
s
m
a
k
in
g
g
ro
u
n
d
r
o
b
o
ts
u
n
su
i
tab
le.
Th
e
UA
V
m
a
y
b
e
m
a
n
a
g
e
d
e
it
h
e
r
i
n
d
e
p
e
n
d
e
n
tl
y
a
b
o
a
rd
c
o
m
p
u
ters
o
r
th
ro
u
g
h
d
istan
c
e
c
o
n
tro
ll
e
d
b
y
a
fli
g
h
t
a
tt
e
n
d
a
n
t
o
n
p
u
lv
e
rize
d
wire
les
s
c
o
m
m
u
n
ica
ti
o
n
li
n
k
s
in
o
u
r
c
a
se
5
G
.
Alth
o
u
g
h
a
fa
st
a
lg
o
ri
th
m
so
lv
e
d
t
h
e
p
ro
b
lem
a
ti
c
a
sp
e
c
t
o
f
b
e
a
m
se
le
c
ti
o
n
f
o
r
2
-
d
ime
n
sio
n
a
l
sc
e
n
a
rio
s.
Th
is
p
a
p
e
r
p
re
se
n
ts
3
-
d
im
e
n
si
o
n
a
l
sc
e
n
a
rio
s
fo
r
UA
Vs
.
We
m
o
d
e
led
b
e
a
m
se
lec
ti
o
n
with
e
n
v
iro
n
m
e
n
tal
re
sp
o
n
siv
e
n
e
ss
in
m
il
li
m
e
ter
Wav
e
UA
V
to
a
c
c
o
m
p
li
sh
c
lo
se
o
p
t
imu
m
a
ss
e
ss
m
e
n
ts
o
n
a
re
g
u
lar
p
e
rio
d
t
h
r
o
u
g
h
lea
rn
in
g
fro
m
t
h
e
a
v
a
il
a
b
le situ
a
ti
o
n
.
K
ey
w
o
r
d
s
:
B
ea
m
s
elec
tio
n
Mu
lti
-
ar
m
ed
b
a
n
d
it
R
ein
f
o
r
ce
m
en
t le
ar
n
i
n
g
Un
m
an
n
ed
ae
r
ial
v
eh
icle
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
:
S.
Mo
jtab
a
Ma
tin
k
h
ah
Dep
ar
tm
en
t
o
f
C
o
m
p
u
ter
E
n
g
i
n
ee
r
in
g
Yaz
d
Un
iv
er
s
ity
Yaz
d
Pro
v
in
ce
,
Saf
ae
ia
h
-
Pazh
o
o
h
esh
Stre
et,
Yaz
d
,
I
r
an
E
m
ail:
m
atin
k
ah
@
y
az
d
.
ac
.
ir
1.
I
NT
RO
D
UCT
I
O
N
Ma
ch
in
e
l
ea
r
n
in
g
(
ML
)
b
ei
n
g
a
p
ar
t
o
f
th
e
ar
tific
ial
in
tellig
en
ce
,
it
h
as
in
f
lu
en
ce
d
s
o
m
e
co
m
m
u
n
icatio
n
lik
e
in
b
r
o
ad
ca
s
t
co
m
m
u
n
icatio
n
well
d
et
ailed
in
wh
er
e
ML
ca
n
b
e
u
s
ed
in
ex
ten
d
in
g
p
ar
am
eter
s
d
u
r
in
g
co
n
n
ec
tiv
it
y
r
em
o
te
ar
ea
s
eith
er
p
o
s
itiv
e
ly
o
r
n
e
g
ativ
ely
i
n
clu
d
in
g
in
f
o
r
m
atio
n
r
etr
iev
als,
in
ter
n
et
o
f
t
h
i
n
g
s
th
at
en
tails
u
n
m
a
n
n
ed
ae
r
ial
v
eh
icles
(
UAVs
)
,
m
ed
ia
an
d
web
co
m
m
u
n
icatio
n
,
o
n
lin
e
m
ak
in
g
,
m
o
n
etiza
tio
n
,
m
in
in
g
,
q
u
ality
o
f
s
er
v
ices
in
we
b
q
u
alities
,
web
s
ec
u
r
ity
an
d
p
r
iv
ac
y
is
s
u
es
in
s
o
cial
n
etwo
r
k
s
am
o
n
g
o
th
e
r
s
[
1
]
-
[
1
1
]
.
Ma
ch
i
n
e
l
ea
r
n
in
g
(
ML
)
b
ein
g
a
p
a
r
t
o
f
th
e
a
r
tific
ial
in
tellig
en
ce
,
it
h
as
in
f
lu
en
ce
d
n
u
m
er
o
u
s
co
m
m
u
n
icatio
n
lik
e
in
b
r
o
ad
ca
s
t
co
m
m
u
n
icatio
n
well
d
etailed
in
wh
er
e
ML
ca
n
b
e
u
s
ed
in
ex
ten
d
in
g
p
ar
am
ete
r
s
d
u
r
in
g
co
n
n
ec
tiv
ity
r
em
o
te
ar
ea
s
ei
th
er
p
o
s
itiv
ely
o
r
n
eg
ativ
ely
in
clu
d
in
g
in
f
o
r
m
atio
n
r
etr
iev
als,
in
ter
n
e
t
o
f
T
h
in
g
s
th
at
en
tails
UAVs,
m
ed
ia
an
d
web
c
o
m
m
u
n
icatio
n
,
o
n
lin
e
m
ak
in
g
,
m
o
n
etiza
tio
n
,
m
in
in
g
,
q
u
ality
o
f
s
er
v
ices
i
n
web
q
u
alities
,
web
s
ec
u
r
ity
an
d
p
r
iv
ac
y
is
s
u
es
in
s
o
cial
n
etwo
r
k
s
as
we
ea
r
ly
d
ep
icted
in
th
e
s
o
cial
web
o
f
t
h
in
g
s
an
d
o
th
er
co
m
p
u
tin
g
p
ar
a
d
ig
m
s
[
1
2
]
,
[
13
]
.
Div
er
s
e
en
g
in
ee
r
in
g
co
r
p
o
r
atio
n
s
h
av
e
o
n
tr
ac
k
to
d
etr
im
en
t
f
r
o
m
u
n
m
an
n
ed
ae
r
ial
v
eh
icles,
s
u
ch
as
ag
r
o
in
d
u
s
tr
y
f
o
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2
2
5
2
-
8
8
1
4
A
3
-
d
imen
s
io
n
a
l fa
s
t m
a
ch
in
e
lea
r
n
in
g
a
lg
o
r
ith
m
fo
r
mo
b
ile
u
n
ma
n
n
e
d
a
eria
l v
eh
icle
…
(
Wa
s
s
w
a
S
h
a
fik
)
29
ex
ac
tn
ess
ag
r
o
n
o
m
y
p
r
o
ce
d
u
r
es,
m
o
v
ie
p
r
o
d
u
ctio
n
s
,
a
n
d
r
e
s
cu
e
b
r
an
c
h
es
f
o
r
ar
tific
ial
d
i
s
aster
ass
ig
n
m
en
ts
.
Nev
er
th
eless
,
th
e
m
en
tio
n
ed
co
n
tem
p
o
r
ar
y
u
n
m
an
n
e
d
ae
r
i
al
v
eh
icles
p
r
o
f
itab
ly
ap
p
licat
io
n
s
n
ec
ess
itie
s
an
u
n
m
an
n
ed
ae
r
ial
v
eh
icles
o
p
er
ativ
e
th
at
av
iato
r
s
th
e
au
to
m
o
b
ile
ex
p
lo
it
ten
u
o
u
s
ly
co
n
tr
o
l
p
o
licies.
As
an
alter
n
ativ
e,
ca
r
g
o
lo
g
is
tics
an
d
s
ca
tter
in
g
m
a
n
u
f
ac
tu
r
in
g
ar
en
as
ar
e
co
n
ce
r
n
e
d
in
t
h
e
r
e
q
u
est
o
f
s
elf
-
d
ir
ec
te
d
u
n
m
an
n
ed
ae
r
ial
v
e
h
icles f
o
r
t
h
e
m
an
ip
u
latio
n
o
f
p
r
o
ce
s
s
es o
th
er
way
s
m
ig
h
t
o
r
ig
in
f
ar
-
r
e
ac
h
in
g
v
a
r
iatio
n
s
in
eq
u
al
d
e
p
o
r
tm
e
n
ts
o
f
ca
r
g
o
lo
g
is
tics
an
d
d
is
tr
ib
u
tio
n
tas
k
s
an
d
cities.
C
o
n
s
eq
u
en
tly
,
it
is
s
ig
n
if
ican
t
to
ad
d
r
ess
th
e
d
ev
o
tio
n
to
co
n
f
id
en
t
u
n
m
an
n
e
d
ae
r
ial
v
eh
icles'
tech
n
o
lo
g
ical
tr
aits
wh
o
s
e
co
n
s
id
er
ate
ca
n
aid
in
an
ev
alu
atio
n
o
f
th
e
tan
g
ib
le
p
r
o
s
p
ec
t
o
f
s
u
ch
UAVs
in
ex
ec
u
tin
g
d
is
tr
ib
u
tio
n
s
an
d
c
o
n
n
ec
tiv
ity
in
th
e
c
ity
co
n
tex
t
d
e
m
o
n
s
tr
ated
i
n
Fig
u
r
e
1
e
n
v
ir
o
n
m
en
t.
I
s
s
u
es
o
f
d
r
o
n
es
ca
n
b
e
class
if
ied
in
d
if
f
er
en
t
way
s
lik
e
m
o
r
ally
,
eth
ically
,
an
d
leg
all
y
th
r
o
u
g
h
wh
ich
n
u
m
e
r
o
u
s
s
tu
d
ies
h
av
e
b
ee
n
d
o
n
e
lik
e
C
h
atto
p
ad
h
y
ay
[
14
]
pr
o
p
o
s
es
n
o
n
-
o
r
t
h
o
g
o
n
al
m
u
ltip
le
ac
ce
s
s
s
ch
em
es
aim
ed
at
th
e
5
G
UAV
co
m
m
u
n
ic
atio
n
s
.
T
h
e
au
th
o
r
d
ep
icted
th
at
th
e
s
y
s
tem
s
with
a
n
o
n
-
o
r
th
o
g
o
n
al
m
u
ltip
le
ac
ce
s
s
es
o
v
er
tak
e
o
th
er
c
o
m
p
o
u
n
d
ad
m
ittan
ce
s
ch
em
es
in
s
tan
d
in
g
s
o
f
q
u
an
tity
v
o
lu
m
e,
d
y
n
am
is
m
ef
f
ec
tiv
en
es
s
,
an
d
Sp
ec
tr
al
ef
f
ec
tiv
en
ess
.
No
tab
ly
,
th
e
p
r
o
p
o
s
ed
m
u
ltip
le
ac
ce
s
s
s
ch
em
es
ac
co
m
p
lis
h
en
h
an
ce
d
s
u
m
-
r
ate
at
a
s
u
b
o
r
d
in
at
e
elev
atio
n
wh
ich
d
im
in
is
h
es th
e
co
m
p
lete
d
y
n
a
m
is
m
o
u
tf
lo
w
o
f
th
e
u
n
m
an
n
e
d
ae
r
ial
v
eh
icles.
El
-
Say
ed
et
a
l
.
[
15
]
d
ep
lo
y
e
d
UAVs
to
ac
t
as
m
o
v
ea
b
le
ed
g
es
f
o
llo
win
g
cir
cu
latio
n
p
r
o
ce
d
u
r
es
an
d
o
v
er
cr
o
wd
in
g
e
n
v
ir
o
n
m
en
ts
.
I
n
th
eir
s
tu
d
y
,
th
e
p
r
o
p
o
s
ed
a
tr
af
f
ic
-
awa
r
e
a
p
p
r
o
ac
h
f
o
r
p
e
r
m
itti
n
g
th
e
p
lace
m
en
t
o
f
u
n
m
an
n
e
d
ae
r
ial
v
eh
icles
in
v
e
h
icu
lar
s
itu
atio
n
s
.
T
h
e
r
ec
r
ea
tio
n
a
l
g
r
ad
es
d
e
p
icted
th
e
p
r
o
jecte
d
tech
n
iq
u
e
ca
n
ac
c
o
m
p
lis
h
o
c
cu
p
ied
s
y
s
tem
atten
tio
n
u
n
d
er
d
iv
er
s
e
ci
r
cu
m
s
tan
ce
s
d
ep
r
iv
ed
o
f
th
e
ex
tr
a
an
n
o
u
n
ce
m
en
t
ab
o
v
e
o
r
in
ter
r
u
p
tio
n
.
Xu
et
a
l
.
[
16
]
co
llecti
v
ely
p
r
o
p
o
s
ed
a
m
o
d
el
to
h
a
n
d
le
th
e
g
ain
s
o
f
th
e
f
if
th
g
e
n
er
atio
n
m
m
-
W
av
e
d
etec
to
r
an
d
ML
tech
n
iq
u
es
to
r
e
co
g
n
ize
th
e
"b
lack
f
ly
in
g
"
UA
V.
T
h
e
s
tu
d
y
m
o
r
e
s
till
d
eliv
er
s
an
o
p
e
r
ativ
e
elu
c
id
atio
n
to
r
eso
lv
e
th
e
tr
ick
y
o
f
d
is
co
v
er
in
g
an
d
r
ec
o
g
n
izin
g
UAV
th
r
o
u
g
h
a
5
G
mm
-
wav
e
d
etec
to
r
in
th
e
I
o
T
th
at
is
h
ig
h
ly
co
n
cr
ete
s
u
b
m
i
s
s
io
n
.
T
h
is
is
a
f
ir
s
t
p
r
o
ject
in
th
e
UAV
r
esear
ch
f
ield
p
r
esen
tin
g
a
s
et
o
f
p
o
s
s
ib
le
b
etter
b
ea
m
s
elec
tio
n
s
ce
n
ar
io
s
in
3
D.
I
t
h
as
n
o
t
em
er
g
ed
d
u
e
to
th
e
tech
n
o
lo
g
ical
in
a
d
eq
u
a
cy
o
f
4
G
an
d
its
ass
o
ciate
d
lim
itatio
n
.
Su
p
e
r
f
icially
,
UAVs
wer
e
p
u
t
in
to
ex
is
t
s
to
s
u
p
p
o
r
t
m
ilit
ar
y
m
is
s
io
n
s
.
Ho
wev
er
,
it
h
as
cr
o
s
s
ed
to
o
th
er
ar
ea
s
lik
e
ag
r
icu
ltu
r
e,
b
u
s
in
ess
g
o
o
d
s
tr
an
s
p
o
r
tatio
n
s
ea
r
ch
,
a
n
d
r
esc
u
e
o
p
er
atio
n
am
o
n
g
o
th
e
r
s
.
T
h
e
UAV
in
th
is
r
esear
ch
is
u
s
ed
as
a
b
ase
s
tatio
n
s
in
ce
th
e
cu
r
r
en
t
s
itu
atio
n
i
s
f
ix
ed
with
a
m
ajo
r
p
r
o
b
le
m
o
f
in
ac
ce
s
s
ib
ilit
y
o
f
s
ig
n
al
to
r
ec
ei
v
er
s
.
I
n
ap
p
r
ec
iatio
n
t
o
th
e
cu
r
r
e
n
t
UAV
s
tu
d
ies,
co
n
d
u
cted
s
tu
d
ies
s
h
o
w
th
at
th
er
e
is
s
till
a
n
ee
d
t
o
ex
p
licitly
in
v
esti
g
ate
UAVs
'
co
m
m
u
n
ic
atio
n
,
s
ec
u
r
ity
,
p
r
iv
ac
y
,
d
e
v
ice
to
d
ev
ice
(
UAV
to
o
th
e
r
d
e
v
ices
)
f
ac
to
r
s
.
T
h
is
p
r
o
ject
m
ain
ly
f
o
cu
s
ed
o
n
th
e
UAV
co
m
m
u
n
icatio
n
ef
f
icie
n
cy
in
3
D
wh
er
e
UAV
is
co
n
s
id
er
ed
to
b
e
a
b
ase
s
tatio
n
as
co
m
p
ar
e
d
to
o
th
er
alter
n
ativ
e
a
p
p
r
o
ac
h
es
to
s
e
r
v
ice
d
eliv
er
y
.
Pěn
ič
k
a
et
a
l
.
[
17
]
p
r
o
p
o
s
ed
a
v
ar
iab
le
n
eig
h
b
o
r
h
o
o
d
s
ea
r
ch
with
th
e
ap
p
licab
le
o
p
tim
u
m
s
am
p
lin
g
-
b
ased
p
r
o
b
ab
le
p
r
o
ce
d
u
r
es
attitu
d
e
f
o
r
r
ec
o
r
d
s
g
ath
er
in
g
f
o
r
ec
asti
n
g
f
o
r
UAVs
in
a
city
atm
o
s
p
h
er
e
with
co
m
p
licatio
n
s
lik
e
b
lo
ck
ag
es
am
o
n
g
o
th
er
s
.
Vlad
u
ta
et
a
l
.
[
18
]
u
s
ed
UAVs
f
o
r
d
ata
g
ath
er
in
g
f
r
o
m
wir
eless
s
en
s
o
r
s
.
T
h
ey
p
r
o
p
o
s
ed
a
s
elf
-
m
o
tiv
ated
er
ec
tio
n
o
f
th
e
p
a
th
(
r
o
u
te)
th
at
UAVs
u
s
e
in
in
f
o
r
m
ati
o
n
co
llectio
n
u
s
in
g
wir
eless
s
en
s
o
r
s
o
n
th
eir
lo
ca
tio
n
s
.
T
h
e
m
ain
co
n
tr
i
b
u
tio
n
s
o
f
t
h
e
p
ap
er
ar
e
s
u
m
m
ar
ized
:
(
i
)
Mo
d
elin
g
o
f
th
e
b
ea
m
s
elec
tio
n
o
f
m
illi
m
eter
W
av
e
b
ase
s
tatio
n
s
b
y
way
o
f
a
MA
B
ap
p
r
o
ac
h
.
T
h
e
ar
ch
et
y
p
al
m
o
d
el
is
g
eo
m
etr
ic,
wh
er
e
th
e
m
o
d
el
is
m
o
d
estl
y
ad
j
u
s
tab
le
to
d
iv
er
s
e
s
itu
atio
n
s
;
(
ii
)
Av
ai
lin
g
an
d
p
r
o
v
is
io
n
o
f
th
e
o
r
i
g
in
al
cir
cu
m
s
tan
tial
o
n
lin
e
alg
o
r
ith
m
f
o
r
b
ea
m
s
elec
tio
n
in
m
illi
m
eter
W
av
e
B
S
.
T
h
e
s
y
s
tem
e
m
p
o
wer
s
th
e
B
Ss
to
in
d
ep
en
d
en
tly
cr
am
ev
er
y
r
esp
ec
tiv
e
b
ea
m
'
s
r
ec
o
r
d
s
,
d
ep
r
iv
ed
o
f
n
ec
e
s
s
itatin
g
a
tr
ain
in
g
s
tag
e
;
(
iii
)
Pro
v
id
in
g
a
n
d
co
n
tr
ib
u
tin
g
to
th
e
s
y
s
tem
atic
s
u
p
er
io
r
b
o
u
n
d
o
n
th
e
r
eg
r
et
t
h
at
d
em
o
n
s
tr
ates th
e
co
n
ju
n
ctio
n
o
f
Fas
t M
ac
h
in
e
L
ea
r
n
in
g
to
t
h
e
o
p
tim
u
m
s
elec
tio
n
s
;
an
d
(
iv
)
E
s
tab
lis
h
m
en
t
o
f
u
s
in
g
th
e
wid
e
-
r
an
g
in
g
m
o
d
el
f
o
r
in
s
tan
ce
with
an
im
ate
an
d
em
b
lem
ati
c
tr
af
f
ic
co
n
f
ig
u
r
atio
n
s
g
o
tte
n
f
r
o
m
Go
o
g
le
Ma
p
s
o
f
th
e
d
esire
d
lo
ca
tio
n
s
.
Nu
m
er
o
u
s
m
o
d
els
cu
r
r
en
tly
h
av
e
b
ee
n
p
r
o
p
o
s
ed
in
m
a
n
ag
in
g
b
ea
m
s
an
d
s
elec
tio
n
.
Den
g
et
a
l
.
[
1
9
]
co
n
s
id
er
ed
r
elay
s
elec
tio
n
a
n
d
b
ea
m
s
elec
tio
n
in
two
-
h
o
p
D2
D
r
elay
in
g
5
G
n
etwo
r
k
s
.
T
h
e
au
th
o
r
s
r
ec
y
cled
u
n
s
cr
u
p
u
l
o
u
s
d
is
p
atch
ass
o
r
tm
en
t
an
d
m
illi
m
eter
W
av
e
eq
u
iv
alen
t
b
ea
m
f
o
r
m
in
g
to
f
r
o
n
t
ier
th
e
g
esti
cu
latin
g
o
v
er
h
ea
d
.
Gio
r
d
a
n
i
et
a
l
.
[
2
0
]
p
r
o
p
o
s
e
d
en
h
a
n
ce
m
en
ts
f
o
r
th
e
b
ea
m
ad
m
i
n
is
tr
atio
n
m
ea
s
u
r
es
en
co
m
p
ass
ed
in
th
e
n
ew
r
ad
io
ac
ce
s
s
tech
n
o
lo
g
y
s
tip
u
latio
n
s
estab
lis
h
ed
o
n
e
q
u
ally
im
p
ar
tial
a
n
d
an
o
th
er
m
an
n
er
.
T
h
e
au
th
o
r
s
also
r
ec
o
m
m
en
d
s
ch
em
e
co
n
s
tr
ain
ts
in
ter
r
elate
d
t
o
b
ea
m
co
n
tr
o
llin
g
co
n
tex
ts
.
W
an
g
et
a
l
.
[
2
1
]
wo
r
k
ed
o
n
s
atellite
co
n
s
tr
u
cted
I
o
T
.
T
h
ey
p
r
o
p
o
s
ed
m
u
ltip
le
b
ea
m
f
o
r
m
in
g
f
o
r
n
u
m
er
o
u
s
m
an
ag
er
s
in
th
e
v
iewp
o
in
t
s
p
h
er
e.
T
h
e
q
u
an
ti
ty
o
f
r
esp
ec
tiv
e
b
ea
m
f
o
r
m
in
g
b
ee
n
n
o
m
i
n
ated
f
o
r
th
e
c
o
n
d
u
it
d
im
en
s
io
n
is
cu
s
h
io
n
ed
,
a
n
d
a
d
ap
t
s
th
e
p
r
o
s
p
ec
t o
f
th
e
a
r
b
itra
r
y
co
llectio
n
o
f
g
r
i
n
s
.
T
h
e
s
elec
tio
n
o
f
th
e
b
est
b
ea
m
f
r
o
m
a
d
is
cr
ete
co
d
e
b
o
o
k
in
wh
ich
d
ig
ital
p
r
ec
o
d
in
g
a
n
d
an
alo
g
b
ea
m
f
o
r
m
in
g
ar
e
co
m
b
in
ed
.
T
h
ey
d
esig
n
e
d
an
iter
ativ
e
s
y
s
tem
to
m
o
d
e
r
n
ize
th
e
s
h
a
f
t
o
f
th
e
b
ea
m
f
o
r
au
x
iliar
y
en
ac
tm
en
t
en
h
a
n
ce
m
en
t.
Pal
et
a
l.
[
2
2
]
p
r
o
p
o
s
e
a
h
eu
r
is
tic
g
r
ee
d
y
alg
o
r
ith
m
f
o
r
b
ea
m
s
elec
tio
n
in
wh
ich
u
s
er
s
f
u
n
ctio
n
in
g
at
m
m
-
W
av
e
in
cid
en
ce
s
an
d
f
o
r
tif
ied
with
a
h
ef
ty
q
u
an
tity
o
f
ten
tacle
r
u
d
im
en
ts
at
th
e
ad
m
ittan
ce
f
ac
t.
T
h
ey
also
p
r
o
p
o
s
e
an
o
th
er
ap
p
r
o
ac
h
e
s
tab
lis
h
ed
o
n
th
e
Ku
h
n
-
Mu
n
k
r
es
p
r
o
ce
d
u
r
e
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
10
,
No
.
1
,
Ma
r
c
h
2
0
2
1
:
28
–
38
30
f
am
iliar
ize
an
in
n
o
v
ativ
e
o
f
f
ic
ial
co
n
tex
t
f
o
r
th
e
b
ea
m
.
Fu
r
th
er
m
o
r
e
,
an
ac
ce
s
s
p
o
in
t
with
a
g
iv
en
n
u
m
b
er
o
f
p
o
r
tio
n
ten
tacle
ass
o
r
tm
en
ts
i
n
ter
co
n
n
ec
ts
with
a
g
i
v
en
n
u
m
b
er
o
f
u
s
er
s
r
esp
ec
tiv
ely
o
b
li
g
atin
g
a
d
is
tin
ct
p
r
o
tu
b
e
r
an
ce
.
T
h
eir
d
elin
q
u
en
t
is
ch
o
o
s
in
g
a
g
iv
en
b
ea
m
with
th
e
e
n
tire
ty
r
ate
as
th
e
en
ac
tm
en
t
m
etr
ic.
T
h
e
m
o
d
eled
r
ay
m
is
ce
llan
y
as
a
b
ilater
al
id
en
tical
am
o
n
g
s
t
th
e
two
s
ets
o
f
p
lay
e
r
s
an
d
co
n
tem
p
late
d
u
alis
tic
d
iv
er
s
e
tech
n
i
q
u
es
o
f
d
e
m
o
n
s
tr
atin
g
th
e
p
lay
er
'
s
p
r
ed
ilect
io
n
s
.
Me
n
g
et
a
l
.
[
2
3
]
p
r
o
p
o
s
ed
B
ea
m
R
aster
,
a
p
r
ec
o
d
in
g
s
ch
em
e
,
f
o
r
m
ass
iv
e
m
u
ltip
le
u
s
er
s
’
m
u
ltip
le
i
n
p
u
ts
an
d
m
u
ltip
le
h
a
r
v
est
c
lass
if
icatio
n
s
.
T
h
e
s
tr
u
ctu
r
e
p
r
e
-
s
u
b
t
r
ac
ts
a
cu
s
to
m
ar
y
o
f
p
er
s
p
ec
tiv
e
-
d
o
m
ain
b
ea
m
s
tr
ain
er
s
wh
ich
r
if
ts
th
e
f
r
eq
u
en
c
y
h
o
o
k
ed
o
n
m
a
n
eu
v
e
r
in
g
s
u
b
s
p
ac
es.
T
h
r
o
u
g
h
a
d
is
co
r
d
an
t b
en
ch
o
f
t
h
e
tr
ajec
to
r
y
th
e
r
elatio
n
s
h
ip
a
m
o
n
g
s
t
a
s
tr
ea
m
o
f
b
ea
m
s
in
in
s
tan
tan
eo
u
s
,
an
d
v
ia
an
in
ter
f
er
en
ce
-
awa
r
e
n
ess
o
f
u
s
er
-
b
ea
m
ass
o
r
tm
en
t,
an
d
l
astl
y
th
r
o
u
g
h
a
p
r
e
-
d
is
to
r
tio
n
tech
n
iq
u
e
to
s
cr
atc
h
th
e
o
u
ts
tan
d
in
g
m
e
d
d
lin
g
d
u
e
to
th
e
s
id
e
-
lo
b
es,
B
ea
m
R
aster
m
an
ag
es
th
e
cr
o
s
s
-
in
ter
f
er
en
ce
.
T
h
e
s
o
f
t
war
e
-
d
ef
in
ed
r
ad
io
p
latf
o
r
m
an
d
r
e
p
licatio
n
s
th
r
o
u
g
h
e
x
ten
s
iv
e
p
r
o
jectio
n
s
s
p
ec
tacle
th
at
ap
p
r
o
ac
h
ca
n
ac
co
m
p
lis
h
a
h
ig
h
er
ca
p
a
b
i
li
ty
ac
h
iev
em
en
t.
An
tó
n
-
Har
o
an
d
Me
s
tr
e
[
2
4
]
in
v
esti
g
ate
h
o
w
an
g
le
-
of
-
ar
r
iv
al
f
ac
ts
m
ig
h
t
b
e
s
u
b
ju
g
ated
v
ia
ex
tr
em
e
lear
n
in
g
,
d
ee
p
lea
r
n
in
g
,
o
r
an
y
o
th
e
r
ML
tactics
to
co
m
p
lete
b
ea
m
ass
o
r
tm
en
t
in
th
e
u
p
lin
k
o
f
a
m
illi
m
eter
W
av
e
co
m
m
u
n
icatio
n
s
ch
em
e.
T
wo
f
o
ld
ad
m
i
n
is
ter
ed
ML
m
eth
o
d
o
lo
g
ies
ap
p
r
o
ac
h
ed
ar
e
f
ir
s
tly
,
th
e
k
-
ad
jace
n
t
n
eig
h
b
o
r
s
an
d
s
u
s
ten
an
ce
tr
ajec
to
r
y
class
if
ier
s
;
an
d
s
ec
o
n
d
ly
,
t
h
e
f
ee
d
-
f
r
o
n
twar
d
u
n
f
ath
o
m
a
b
le
n
eu
r
al
s
y
s
tem
s
:
th
e
m
u
ltip
le
-
lay
er
p
er
ce
p
tr
o
n
’
s
.
T
h
e
y
co
m
p
ar
ed
t
h
e
s
u
m
-
r
ate
o
f
th
eir
p
r
o
p
o
s
al
with
ex
h
au
s
tiv
e
s
ea
r
ch
.
J
ian
g
an
d
J
af
ar
k
h
an
i
[
2
5
]
p
r
o
p
o
s
ed
a
b
ea
m
f
o
r
m
in
g
s
ch
e
m
e
th
at
o
p
tim
izes
th
e
im
p
r
o
v
em
en
t
an
d
in
ter
-
u
s
er
in
tr
u
s
io
n
at
a
n
id
en
tical
s
tr
etch
.
Ab
o
v
e
an
d
b
ey
o
n
d
,
th
e
au
th
o
r
s
p
r
e
m
ed
itated
a
v
ig
o
r
o
u
s
b
ea
m
f
o
r
m
in
g
s
tr
u
ct
u
r
e
to
d
el
iv
er
s
tu
r
d
in
ess
in
c
o
n
tr
a
d
ictio
n
o
f
d
e
f
ec
tiv
e
f
r
eq
u
e
n
cy
d
a
ta.
Mo
r
e
s
till
,
th
ey
d
em
o
n
s
tr
ated
th
at
th
e
an
tici
p
ated
n
o
n
-
r
o
b
u
s
t
m
u
lti
-
u
s
er
eq
u
iv
ale
n
t
b
ea
m
f
o
r
m
er
o
u
t
d
o
es
th
e
o
u
td
ate
d
eq
u
iv
alen
t
b
ea
m
f
o
r
m
in
g
tech
n
iq
u
e
af
ter
th
e
SNR
-
ef
f
ec
t
is
in
h
eig
h
t
an
d
th
e
p
lan
n
ed
v
i
g
o
r
o
u
s
b
ea
m
f
o
r
m
e
r
m
ay
d
eliv
er
o
v
er
1
1
0
% e
n
h
an
ce
m
en
t in
th
e
s
u
m
-
r
ate
ass
o
ciate
d
th
r
o
u
g
h
th
e
b
ea
m
tech
n
iq
u
es o
f
s
elec
tio
n
.
Ab
o
v
e
all
th
ese
m
o
d
els
p
r
o
p
o
s
ed
,
n
o
n
e
o
f
th
em
a
d
d
r
ess
r
em
o
te
ac
ce
s
s
co
m
m
u
n
icatio
n
lik
e
i
n
v
alley
s
b
elo
w
th
e
lin
e
o
f
s
ites
,
h
elp
in
tim
e
o
f
d
a
n
g
er
an
d
also
o
p
tim
ized
b
ea
m
s
elec
tio
n
in
a
p
p
r
o
ac
h
t
o
av
er
ag
e
d
ata
to
b
e
r
ec
ei
v
ed
,
th
e
im
p
ac
t
o
f
f
ix
ed
b
ase
s
tatio
n
s
,
th
e
im
p
ac
t
o
f
s
elec
ted
b
ea
m
s
,
im
p
ac
t
o
n
b
lo
ck
ag
es,
liv
e
d
aily
tr
af
f
ic
p
atter
n
s
am
o
n
g
o
th
er
s
.
T
h
ese
j
u
s
tify
th
e
n
ee
d
to
h
av
e
d
ee
p
s
tu
d
y
co
v
er
all
th
e
ab
o
v
e
f
ea
t
u
r
es
m
en
tio
n
ed
th
a
t
th
e
cu
r
r
e
n
t
f
ea
tu
r
es
ar
e
n
o
t
ap
p
r
o
ac
h
in
g
at
all
p
o
s
s
ib
le
ap
p
licatio
n
s
r
eq
u
ir
e
litt
le
o
r
n
o
h
u
m
a
n
in
ter
v
e
n
tio
n
an
d
th
u
s
m
u
s
t
h
av
e
th
e
ab
i
lity
to
o
p
er
ate
au
to
n
o
m
o
u
s
ly
th
r
o
u
g
h
th
e
u
s
e
o
f
g
lo
b
al
m
ap
s
lik
e
Go
o
g
le
Ma
p
s
.
T
h
ese
ju
s
tify
th
e
n
ee
d
to
h
a
v
e
an
in
-
d
ep
th
s
tu
d
y
co
v
er
all
th
e
ab
o
v
e
f
ea
t
u
r
es
m
en
tio
n
ed
th
at
th
e
cu
r
r
en
t
f
ea
tu
r
es
ar
e
n
o
t
a
p
p
r
o
ac
h
in
g
at
al
l
p
o
s
s
ib
le
ap
p
licatio
n
s
r
eq
u
i
r
e
litt
le
o
r
n
o
h
u
m
a
n
in
ter
v
en
tio
n
a
n
d
th
u
s
m
u
s
t
h
av
e
th
e
ab
ilit
y
to
o
p
er
ate
a
u
to
n
o
m
o
u
s
ly
th
r
o
u
g
h
th
e
u
s
e
o
f
a
g
lo
b
al
m
ap
lik
e
p
r
esen
ted
in
F
ig
u
r
e
1
.
Fig
u
r
e
1
.
T
h
e
s
im
u
latio
n
atm
o
s
p
h
er
e
.
T
h
e
t
r
o
lley
s
with
b
u
s
m
o
m
en
tar
ily
b
lo
ck
r
ed
ca
r
with
b
lack
ca
r
co
r
r
esp
o
n
d
in
g
ly
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2
2
5
2
-
8
8
1
4
A
3
-
d
imen
s
io
n
a
l fa
s
t m
a
ch
in
e
lea
r
n
in
g
a
lg
o
r
ith
m
fo
r
mo
b
ile
u
n
ma
n
n
e
d
a
eria
l v
eh
icle
…
(
Wa
s
s
w
a
S
h
a
fik
)
31
2.
RE
S
E
ARCH
M
E
T
H
O
D
2
.
1
.
M
o
del descr
iptio
n
T
h
e
m
UB
S
ass
u
m
ed
a
p
r
ed
eter
m
in
ed
s
et
is
d
en
o
ted
as
|
ℏ
|
d
is
cr
ete
n
o
t
n
o
n
-
r
i
g
h
t
an
g
les
(
n
o
n
-
o
r
th
o
g
o
n
al)
UAV
b
ea
m
s
.
T
h
e
m
o
d
el
ad
o
p
ted
th
at
t
h
e
m
UB
S
m
ig
h
t
s
o
litar
y
s
elec
t
a
s
u
b
s
ec
tio
n
o
f
b
ea
m
s
co
n
cu
r
r
en
tly
n
o
tin
g
th
at
∈
ℕ
,
≺
,
r
em
ain
s
co
n
s
id
er
ed
to
b
e
an
im
m
o
b
ile
am
o
u
n
t.
So
m
e
o
f
th
e
lim
itatio
n
s
id
en
tifie
d
o
n
m
m
W
av
e
ch
an
n
el
s
p
ar
s
ity
.
T
h
e
m
ain
r
ea
s
o
n
f
o
r
th
e
m
UB
S
s
tan
d
s
to
s
elec
t
a
s
u
b
s
ec
tio
n
o
f
th
at
will
b
e
m
a
x
im
izin
g
th
e
n
u
m
b
e
r
o
f
r
ec
o
r
d
s
m
ag
n
if
ice
n
tly
e
x
p
ec
ted
b
y
th
e
ap
p
r
o
ac
h
i
n
g
C
R
in
th
e
r
ep
o
r
tag
e
zo
n
e.
T
h
e
m
o
d
el
f
u
r
th
er
a
d
o
p
te
d
th
e
f
ac
t
t
h
at
th
e
m
UB
S
is
n
o
t
ce
r
tain
o
r
n
o
n
o
th
in
g
ab
o
u
t
th
en
en
v
ir
o
n
m
e
n
t.
No
r
m
ally
,
in
th
is
s
i
tu
atio
n
,
th
e
co
m
p
licatio
n
o
f
th
e
co
m
p
lex
ap
p
licatio
n
r
ed
u
ce
s
as
th
e
o
p
er
ativ
e
n
ec
ess
ities
n
o
th
in
g
to
b
e
co
n
f
ig
u
r
e
d
lik
e
at
ea
ch
m
UB
S
ac
co
r
d
in
g
to
th
e
en
v
ir
o
n
m
e
n
t.
T
h
er
ef
o
r
e,
th
e
m
UB
S
h
as
to
lear
n
as
th
e
s
itu
atio
n
ch
an
g
es
to
s
elec
t
th
e
s
u
b
s
et
o
f
th
e
b
ea
m
s
.
I
n
th
is
way
,
th
e
UAV
will
b
e
to
ac
co
u
n
t
f
o
r
e
v
er
y
a
p
p
r
o
ac
h
in
g
C
R
in
co
n
te
x
t
to
b
ea
m
s
it
em
its
.
W
e
will
also
tak
e
in
m
i
n
d
a
d
is
tin
ct
-
in
ter
v
al
s
itu
atio
n
,
g
iv
en
th
at
th
e
m
UB
S
in
f
o
r
m
s
th
a
t
th
e
b
ea
m
in
th
e
s
tead
y
i
n
ter
v
al
s
etu
p
in
ev
er
y
a
s
etu
p
o
r
p
e
r
io
d
=
1
,
.
.
.
,
.
.
,
ℵ
,
g
iv
e
n
th
at
ℵ
∈
ℕ
also
co
n
s
id
er
ed
to
b
e
a
f
in
ite
h
o
r
izo
n
;
th
e
f
o
llo
win
g
th
r
e
e
ac
tiv
ities
ar
e
ap
p
lied
.
a.
A
s
ec
tio
n
=
{
,
}
=
1
,
.
.
.
.
,
.
.
.
is
g
iv
en
th
at
=
|
|
C
R
lik
e
au
to
m
o
b
iles
,
d
if
f
e
r
en
t
I
o
T
s
am
o
n
g
o
th
e
r
s
to
th
e
m
UB
S.
T
h
e
n
u
m
b
er
o
f
th
e
C
R
s
f
u
lf
ills
th
e
co
n
d
itio
n
th
at
≤
ma
x
co
n
s
id
er
in
g
t
h
e
∈
ℕ
will
b
e
th
e
ex
tr
em
e
s
u
m
r
ein
f
o
r
ce
d
C
R
s
with
in
th
e
co
v
er
ag
e
ar
ea
.
At
th
e
tim
e
o
f
th
e
r
eg
is
tr
atio
n
,
m
UB
S
will
b
e
h
av
in
g
th
e
ca
p
ac
ity
to
r
ec
eiv
e
d
ata
ab
o
u
t
th
e
s
itu
atio
n
it,
o
f
ev
er
y
f
o
r
t
h
co
m
in
g
au
to
m
o
b
ile
,
,
will
b
e
a
r
ea
lis
tic
tr
ajec
to
r
y
r
eser
v
ed
f
r
o
m
th
e
co
n
s
tr
ain
ed
co
v
er
ag
e
a
r
ea
=
[
0
,
1
,
2
]
.
b.
T
h
e
m
UB
S
ch
o
o
s
es
a
s
u
b
s
ec
t
io
n
o
f
bm
.
T
h
e
s
tu
d
y
also
d
en
o
tes
th
at
th
e
cu
s
to
m
er
s
o
f
p
ar
ticu
lar
b
ea
m
s
in
th
e
r
etr
o
in
th
e
in
ter
v
al
by
=
{
,
}
=
1
,
.
.
.
.
,
⊆
.
Fo
r
m
er
ly
t
h
e
C
R
in
th
e
will
b
e
in
f
o
r
m
e
d
ab
o
u
t th
e
n
o
m
in
ated
r
ay
s
th
r
o
u
g
h
C
R
b
o
u
n
d
ar
y
.
c.
At
th
e
tim
e,
in
ca
s
e
t
h
e
C
R
u
will
b
e
with
in
t
h
e
r
a
n
g
e
o
f
m
UB
S
co
v
er
ag
e
a
r
ea
ac
co
r
d
in
g
to
g
o
o
g
le
m
a
p
an
d
m
UB
S
wil
l
b
e
in
a
p
o
s
iti
o
n
to
tr
an
s
m
it
d
ata
to
an
y
C
R
with
in
th
e
co
v
er
ag
e
ar
ea
.
Ob
s
er
v
atio
n
will
b
e
co
n
s
id
er
ed
o
n
th
e
am
o
u
n
t
o
f
d
ata
,
(
,
,
)
CR
C
r
u
will
p
r
o
d
u
cti
v
ely
b
e
r
ec
eiv
e
d
b
y
s
elec
ted
b
ea
m
s
,
,
=
1
,
.
.
.
,
,
till
th
e
en
d
o
f
th
e
.
Den
o
tin
g
th
at
th
e
r
an
d
o
m
v
ar
iab
le
(
)
th
e
b
ea
m
en
ac
tm
en
t
o
f
u
n
d
er
th
e
co
n
tex
t
o
f
th
e
x.
I
t
will
b
e
m
ea
n
in
g
th
at
th
e
q
u
a
n
tity
o
f
r
ec
o
r
d
s
rb
(
x)
o
f
C
R
with
th
e
co
n
tex
t
will
b
e
r
ec
eiv
in
g
af
te
r
th
e
m
UB
S
u
s
in
g
th
e
.
T
h
e
s
tu
d
y
p
r
esu
m
ed
th
at
th
is
a
r
b
itra
r
y
a
d
ap
tab
le
is
c
o
n
s
tr
ain
ed
in
[
0
,
1
,
M
Aod
]
,
g
iv
en
M
Aod
will b
e
th
e
all
-
o
u
t su
m
o
f
r
ec
o
r
d
s
wh
ich
m
ay
b
e
ac
k
n
o
wled
g
ed
b
y
C
R
.
M
Aod
will b
e
r
estricte
d
b
y
th
e
all
-
o
u
t
d
eg
r
ee
o
f
th
e
b
r
o
ad
ca
s
t
c
h
an
n
el.
W
e
d
en
o
ted
th
e
ex
p
e
cted
v
alu
e
o
f
th
e
b
ea
m
p
er
f
o
r
m
an
ce
o
f
th
e
b
ea
m
in
th
e
s
itu
atio
n
p
w
ith
μ
b
(
x)
.
T
h
e
m
UB
S
am
b
itio
n
s
at
c
h
o
o
s
in
g
a
s
u
b
g
r
o
u
p
o
f
th
e
b
ea
m
s
th
at
m
ax
im
izes
t
h
e
p
r
ed
ictab
le
r
ec
eiv
e
r
ec
o
r
d
s
at
C
R
.
T
h
er
ef
o
r
e,
th
e
o
p
tim
al
s
u
b
s
et
in
p
e
r
io
d
t
∀
∗
(
)
=
{
∀
∗
,
(
)
}
=
1
,
.
.
.
,
.
.
.
,
⊆
.
T
h
er
ef
o
r
e
th
e
cu
s
to
m
er
∀
∗
(
)
will b
e
d
ep
en
d
in
g
o
n
X
t
={xt,i}i=
1
,
….
,
an
d
b
m
s
atis
f
y
(
1
)
.
(
)
(
)
(
)
(
)
*
1*
1
1
,
k
a
r
g
m
a
x
,
\
t
t
t
b
m
l
z
k
t
t
P
p
t
z
b
m
P
=
=
=
(
1
)
No
tin
g
th
at
j=1
,
bm
.
I
n
ca
s
e
th
e
m
UB
S
wi
ll
k
n
o
w
th
e
ex
p
ec
ted
b
ea
m
p
e
r
f
o
r
m
an
ce
μ
b
(
x)
.
Fo
r
ev
er
y
C
R
s
itu
atio
n
p
ϵ
X
an
d
ea
ch
b
ea
m
bm
ϵ
,
it
will
b
e
s
elec
ted
o
p
ti
m
u
m
s
u
b
s
ec
tio
n
o
f
th
e
b
ea
m
f
o
r
ev
er
y
r
eg
u
lar
o
f
f
o
r
th
co
m
in
g
C
R
.
T
o
o
b
tain
th
e
ex
p
ec
ted
am
o
u
n
t,
t
h
e
d
ata
will
b
e
r
ec
eiv
ed
o
v
er
th
e
s
eq
u
en
ce
f
r
o
m
1
to
th
e
tim
e
T
th
at
o
b
tain
e
d
(
2
)
.
∑
∑
∑
[
∀
,
(
)
∗
(
,
)
]
=
∑
∑
∑
∀
,
(
)
∗
(
,
)
=
1
=
1
ℵ
=
1
=
1
=
1
ℵ
=
1
(
2
)
T
h
e
m
UB
S
d
o
es
n
o
t r
ec
o
g
n
ize
th
e
co
v
e
r
ag
e
a
r
ea
;
it
will
b
e
lear
n
in
g
th
e
ex
p
ec
te
d
p
er
f
o
r
m
an
ce
μ
b
(
x)
.
Ov
er
a
g
iv
en
p
e
r
io
d
(
3
)
,
to
ac
q
u
ir
e
th
e
id
en
tifie
d
c
o
n
ce
r
n
s
,
th
e
m
UB
S
r
eq
u
ir
es
attem
p
tin
g
d
iv
er
s
e
b
ea
m
s
o
f
d
iv
er
s
e
C
R
s
itu
atio
n
co
m
p
leted
s
tr
etch
s
im
ilar
ly
en
s
u
r
in
g
th
at
th
e
b
ea
m
s
will
b
e
p
r
o
v
ed
in
b
ein
g
g
o
o
d
.
L
astl
y
,
th
e
lear
n
i
n
g
alg
o
r
ith
m
(
Alg
o
r
ith
m
1
)
will
b
e
d
o
n
e
s
o
m
etim
es
f
o
r
C
R
in
th
e
c
o
v
er
ag
e
ar
ea
in
th
e
co
n
tex
t o
f
P
t
,
s
elec
tin
g
th
e
St
o
f
bm
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
10
,
No
.
1
,
Ma
r
c
h
2
0
2
1
:
28
–
38
32
Alg
o
r
ith
m
1
.
Ps
eu
d
o
co
d
e
o
f
p
r
o
p
o
s
ed
r
ein
f
o
r
ce
m
en
t a
lg
o
r
it
h
m
1.
I
n
p
u
t:
,
,
(
)
2.
I
n
itialize
co
n
tex
t p
ar
titi
o
n
:
C
r
ete
p
ar
titi
o
n
o
f
co
n
te
x
t sp
ac
e
[
0
,
1
,
2
]
in
to
(
)
h
y
p
er
cu
b
es
o
f
id
en
tical
s
i
ze
s
3.
I
n
itialize
co
u
n
ter
s
: Fo
r
all
∈
an
d
all
ℎ
∈
s
et
,
ℎ
=
0
4.
I
n
itialize
esti
m
ates: Fo
r
all
∈
an
d
all
ℎ
∈
,
s
et
′
,
ℎ
=
0
5.
f
o
r
ea
ch
=
1
,
.
.
.
do
6.
Ob
s
er
v
e
v
eh
icle
co
n
tex
ts
=
{
,
}
=
1
,
.
.
.
,
7.
Fin
d
=
{
ℎ
,
}
=
1
,
.
.
.
,
s
u
ch
th
at
,
∈
,
∈
,
=
1
,
.
.
.
,
8.
C
o
m
p
u
te
th
e
s
et
o
f
u
n
d
e
r
-
ex
p
l
o
r
ed
b
ea
m
s
(
)
in
(
5
)
9.
if
(
)
≠
0
th
en
E
x
p
lo
r
atio
n
10.
=
(
(
)
)
11.
if
≥
th
en
12.
s
elec
t
,
1
,
.
.
.
,
,
r
an
d
o
m
ly
f
r
o
m
(
)
13.
else
14.
s
elec
t
,
1
,
.
.
.
,
,
as u
b
ea
m
s
f
r
o
m
(
)
15.
s
elec
t
,
,
as th
e
(
−
)
b
ea
m
s
f
r
o
m
(
6
)
1
,
m
u
,
t
(
t)
,
..
.,
b
H
(
t)
t
bH
−
16.
else if
17.
else
18.
s
elec
t
,
1
,
.
.
.
,
,
as th
e
b
ea
m
s
1
,
m
u
,
t
(
t)
,
..
.,
b
H
(
t)
t
bH
−
f
r
o
m
(
7
)
19.
en
d
if
20.
Ob
s
er
v
e
th
e
r
ec
eiv
e
d
d
ate
,
o
f
ea
ch
v
eh
icle
,
=
1
,
.
.
.
,
in
ea
ch
b
ea
m
,
,
=
1
,
.
.
.
,
21.
f
o
r
=
1
,
.
.
.
,
do
22.
f
o
r
=
1
,
.
.
.
,
do
23.
,
,
′
ℎ
,
=
,
′
,
ℎ
,
+
,
,
,
ℎ
,
+
1
an
d
,
,
,
,
ℎ
,
ℎ
,
24.
en
d
f
o
r
25.
en
d
f
o
r
26.
en
d
f
o
r
T
h
e
co
llectio
n
o
f
th
e
er
u
d
itio
n
will
b
e
d
ep
en
d
in
g
o
n
th
e
h
is
to
r
y
o
f
th
e
b
ea
m
s
to
b
e
s
elec
ted
(
4
)
.
T
h
e
an
ticip
ated
q
u
an
tity
o
f
r
ec
o
r
d
s
co
n
v
en
tio
n
al
b
y
t
h
e
v
e
h
icle
wi
ll
b
e
g
iv
e
n
as
f
o
llo
ws
in
ca
s
e
we
co
n
s
id
er
th
e
s
elec
tio
n
St,
t=
1
.
.
.
,
T
o
f
an
alg
o
r
ith
m
.
∑
∑
∑
[
,
(
,
)
]
=
1
=
1
ℵ
=
1
(
3
)
∑
∑
∑
[
,
(
)
(
,
)
−
,
(
,
)
]
=
1
=
1
ℵ
=
1
(
4
)
T
h
er
ef
o
r
e,
th
e
an
ticip
ated
m
o
d
if
icatio
n
in
th
e
q
u
an
tity
o
f
co
n
v
e
n
tio
n
al
in
f
o
r
m
atio
n
co
m
p
leted
an
d
a
n
alg
o
r
ith
m
will
b
e
th
e
"lea
r
n
in
g
r
eg
r
et"
tak
en
to
b
e
R
co
n
s
id
er
in
g
b
o
th
(
2
)
a
n
d
(
4
)
.
Fo
llo
w
in
g
th
e
f
o
r
m
u
latio
n
o
f
th
e
last
eq
u
atio
n
co
n
cl
u
s
iv
ely
lead
in
g
to
th
e
R
(
T
)
p
r
o
d
u
ce
d
b
y
th
e
(
5
)
an
d
(
6
)
r
esp
ec
tiv
e
ly
.
(
)
=
[
∑
∑
∑
(
∀
,
(
)
∗
(
,
)
)
=
1
=
1
ℵ
=
1
−
,
(
)
(
,
)
]
(
5)
(
)
=
[
∑
∑
∑
(
∀
,
(
)
∗
(
,
)
)
=
1
=
1
ℵ
=
1
−
[
,
(
)
(
,
)
]
]
(
6
)
2
.
2
.
L
ea
rning
t
ec
hn
i
qu
e
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
o
f
s
elec
tin
g
b
ea
m
in
an
m
UB
S
as
a
f
ast
3
-
d
im
en
s
io
n
s
em
i
-
o
n
lin
e
lear
n
in
g
p
r
o
b
lem
as
d
e
p
icted
in
th
e
al
g
o
r
ith
m
b
elo
w
s
in
ce
it
allo
ws
th
e
id
en
tific
atio
n
o
f
th
e
b
est
b
ea
m
s
in
d
ep
en
d
e
n
tly
in
a
p
r
ea
r
r
an
g
ed
in
ter
v
al
wh
il
e
s
ec
r
etar
ial
f
o
r
s
elf
-
mo
tiv
ated
tr
af
fic
an
d
atm
o
s
p
h
er
e
ch
a
n
g
es
d
ep
icted
in
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2
2
5
2
-
8
8
1
4
A
3
-
d
imen
s
io
n
a
l fa
s
t m
a
ch
in
e
lea
r
n
in
g
a
lg
o
r
ith
m
fo
r
mo
b
ile
u
n
ma
n
n
e
d
a
eria
l v
eh
icle
…
(
Wa
s
s
w
a
S
h
a
fik
)
33
Alg
o
r
ith
m
1
.
Hen
ce
f
o
r
th
,
th
e
m
UB
S
n
ec
ess
it
ies
to
ca
teg
o
r
ize
th
e
s
u
p
er
lativ
e
b
ea
m
th
r
o
u
g
h
p
r
u
d
e
n
tly
ch
o
o
s
in
g
s
u
b
ca
te
g
o
r
ies
o
f
b
e
am
s
o
v
er
a
ce
r
tain
in
ter
v
al.
T
h
is
tactic
ca
s
ca
d
es
b
en
ea
th
th
e
class
if
icatio
n
o
f
ap
p
r
o
p
r
iate
MA
B
d
if
f
icu
lties
.
Su
ch
is
s
u
es
f
u
r
t
h
er
m
o
r
e
co
n
tain
s
id
e
d
ata
th
at
to
u
c
h
es
th
e
r
ewa
r
d
s
o
f
th
e
m
o
v
em
en
ts
.
A
co
n
tex
tu
al
m
u
l
ti
-
ar
m
ed
b
an
d
it
ap
p
r
o
ac
h
will
ap
p
r
ec
iate
th
e
m
UB
S
d
o
es
n
o
t
m
o
d
estl
y
ab
s
o
r
b
,
o
r
in
f
er
io
r
wh
ich
b
ea
m
s
ar
e
u
n
to
u
ch
a
b
le
o
n
u
n
v
ar
y
in
g
,
th
e
n
ag
ain
in
its
p
lace
it e
x
p
lo
its
s
u
p
er
f
lu
o
u
s
ev
id
en
ce
ab
o
u
t
im
m
in
en
t
C
R
to
d
is
tin
g
u
is
h
wh
ich
b
ea
m
s
ar
e
th
e
s
u
p
er
lativ
e
u
n
d
er
n
ea
th
an
ass
u
m
ed
tr
af
fic
co
n
d
itio
n
as th
e
p
r
o
ce
d
u
r
e
clar
if
ies.
3.
NUM
E
RIC
AL
R
E
SU
L
T
S A
ND
SI
M
UL
AT
I
O
N
B
r
ief
ly
,
th
is
s
ec
tio
n
p
r
esen
ts
a
d
is
cu
s
s
io
n
o
f
th
e
p
r
o
v
id
e
d
s
im
u
latio
n
r
esu
lts
in
Fig
u
r
e
2
,
an
d
a
d
etailed
ex
p
lan
atio
n
o
f
th
e
a
lg
o
r
ith
m
p
r
esen
ted
o
b
tain
ed
f
r
o
m
th
e
s
im
u
latio
n
o
f
th
e
c
h
an
n
el
d
escr
ip
tio
n
p
r
esen
ted
in
T
a
b
le
1
an
d
f
r
o
m
p
ar
am
eter
s
in
p
u
ts
in
T
a
b
le
2
.
Fig
u
r
e
2
.
I
m
p
ac
t o
f
b
lo
c
k
ag
es
o
n
th
e
c
u
m
u
lativ
e
r
ec
ei
v
ed
d
at
a
f
o
r
a
n
ar
r
iv
al
r
ate
o
f
s
p
ec
if
ie
d
λ
an
d
tim
e
T
ab
le
1
.
C
h
an
n
el
p
ar
am
eter
s
P
a
r
a
me
t
e
r
s
V
a
l
u
e
s
a
n
d
N
o
t
a
t
i
o
n
s
C
a
r
r
i
e
r
r
e
g
u
l
a
r
i
t
y
2
8
G
i
g
a
h
e
r
t
z
B
a
n
d
w
i
d
t
h
1
G
i
g
a
h
e
r
t
z
Tr
a
n
sm
i
t
p
o
w
e
r
3
0
d
e
c
i
b
e
l
s
P
a
t
h
l
o
ss
m
o
d
e
l
(
d
B
)
32
.
4
+
17
.
3
10
(
)
+
20
lo
g
10
(
(
)
)
+
N
o
i
se
f
i
g
u
r
e
4
d
e
c
i
b
e
l
s
A
u
t
o
mo
b
i
l
e
U
A
V
’
s b
e
a
m
w
i
d
t
h
36
0
S
i
mu
l
a
t
o
r
t
o
o
l
N
S
3
Th
e
r
m
o
N
o
i
s
e
−
174
/
T
ab
le
2
.
Simu
latio
n
i
n
p
u
ts
an
d
s
p
ec
if
icatio
n
s
I
n
p
u
t
p
a
r
a
me
t
e
r
s
:
S
p
e
c
i
f
i
c
a
t
i
o
n
C
h
a
n
n
e
l
l
o
a
d
M
e
a
n
a
b
so
l
u
t
e
d
e
v
i
a
t
i
o
n
r
e
s
u
l
t
s
β
0
.
1
0
0
0
0
0
v
e
h
/
m
Ϸ
=
0
.
2
3
,
α
=
0
.
0
5
8
4
1
5
O
p
t
i
mal
1
.
6
0
λ
1
0
H
z
Ϸ
=
0
.
2
3
,
α
=
0
.
0
5
8
4
1
5
2
D
F
M
L
0
.
1
9
Pt
2
0
d
B
m
Ϸ
=
0
.
2
3
,
α
=
0
.
0
5
8
4
1
5
2
D
F
M
L
0
.
2
1
S
2
s
u
b
c
h
a
n
n
e
l
s
Ϸ
=
0
.
2
3
,
α
=
0
.
0
5
8
4
1
5
U
C
B
0
.
0
7
B
1
9
0
b
y
t
e
s
Ϸ
=
0
.
2
3
,
α
=
0
.
0
5
8
4
1
5
R
a
n
d
o
m
1
.
2
5
β
0
.
2
0
0
0
0
0
v
e
h
/
m
Ϸ
=
0
.
4
4
,
α
=
0
.
4
7
8
9
2
1
O
p
t
i
mal
0
.
9
1
λ
1
0
H
z
Ϸ
=
0
.
4
4
,
α
=
0
.
4
7
8
9
2
1
2
D
F
M
L
0
.
1
6
Pt
2
0
d
B
m
Ϸ
=
0
.
4
4
,
α
=
0
.
4
7
8
9
2
1
2
D
F
M
L
0
.
1
8
S
4
s
u
b
c
h
a
n
n
e
l
s
Ϸ
=
0
.
4
4
,
α
=
0
.
4
7
8
9
2
1
U
C
B
0
.
0
7
B
1
9
0
b
y
t
e
s
Ϸ
=
0
.
4
4
,
α
=
0
.
4
7
8
9
2
1
R
a
n
d
o
m
0
.
8
3
β
0
.
3
0
0
0
0
0
v
e
h
/
m
Ϸ
=
0
.
6
2
,
α
=
0
.
8
3
2
4
7
0
O
p
t
i
mal
0
.
9
2
λ
1
0
H
z
Ϸ
=
0
.
6
2
,
α
=
0
.
8
3
2
4
7
0
2
D
F
M
L
0
.
1
4
Pt
2
0
d
B
m
Ϸ
=
0
.
6
2
,
α
=
0
.
8
3
2
4
7
0
3
D
F
M
L
0
.
2
0
S
4
s
u
b
c
h
a
n
n
e
l
s
Ϸ
=
0
.
6
2
,
α
=
0
.
8
3
2
4
7
0
U
C
B
0
.
0
7
B
1
9
0
b
y
t
e
s
Ϸ
=
0
.
6
2
,
α
=
0
.
8
3
2
4
7
0
R
a
n
d
o
m
0
.
8
4
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
10
,
No
.
1
,
Ma
r
c
h
2
0
2
1
:
28
–
38
34
3
.
1
.
Acc
lim
a
t
izing
o
n str
uct
ura
l dy
na
m
ics
T
h
e
r
esu
lt
o
f
th
e
ass
es
s
m
en
t
is
r
u
m
m
ag
e
-
s
ale
to
ac
clim
atize
to
s
tr
u
ctu
r
al
d
im
in
u
en
d
o
s
lik
e
th
e
atten
d
an
ce
o
f
o
b
s
tr
u
ctio
n
s
an
d
d
e
v
iatio
n
s
in
tr
af
f
ic
co
n
f
ig
u
r
atio
n
s
.
T
h
e
p
r
o
ce
d
u
r
e
d
is
tin
g
u
is
h
es
b
lo
ck
ag
es
b
y
ass
es
s
in
g
th
e
cu
m
u
lativ
e
tr
ad
itio
n
al
r
ec
o
r
d
s
o
f
th
e
r
esp
ec
tiv
e
au
to
m
o
b
ile
f
o
r
th
e
co
r
r
esp
o
n
d
in
g
ly
n
o
m
in
ated
b
ea
m
.
Fu
r
t
h
er
m
o
r
e,
t
h
e
alg
o
r
ith
m
ac
clim
ates
to
r
o
a
d
tr
a
f
f
ic
p
atter
n
s
b
y
th
e
e
r
u
d
itio
n
o
f
th
e
ass
o
ciatio
n
am
o
n
g
s
t th
e
r
o
u
te
o
f
o
n
s
et
an
d
th
e
ac
k
n
o
wled
g
e
d
r
ec
o
r
d
s
.
As
a
r
esu
lt,
it
c
h
o
o
s
es
th
e
b
m
s
,
wh
i
ch
ex
p
l
o
it
th
e
co
m
p
lete
s
y
s
tem
ap
titu
d
e.
C
o
n
s
e
q
u
en
tly
,
it
d
eliv
er
s
th
e
s
u
p
p
lem
en
tar
y
r
e
m
ain
d
er
to
th
e
th
o
r
o
u
g
h
f
ar
es
with
s
o
p
h
is
ticated
tr
af
f
ic
an
d
h
er
ea
f
ter
,
s
er
v
es
an
o
u
ts
tan
d
in
g
q
u
an
tity
o
f
a
u
to
m
o
b
iles
.
T
h
e
m
o
d
el
th
e
d
elin
q
u
en
t a
s
th
at
MA
B
p
r
o
b
lem
wh
er
e
f
r
eq
u
e
n
t d
is
p
u
tes
in
wir
eless
in
f
r
astru
ctu
r
es
h
a
v
e
b
ee
n
tr
ea
ted
b
y
th
e
u
s
e
o
f
m
u
lti
-
ar
m
e
d
b
a
n
d
its
,
a
d
ec
i
s
io
n
-
m
ak
er
h
as
to
ch
o
o
s
e
a
s
u
b
d
i
v
is
io
n
o
f
ac
ti
v
ities
o
f
m
y
s
ter
io
u
s
r
ec
o
m
p
en
s
es
to
ac
h
iev
e
m
en
t
t
h
e
r
e
co
m
p
en
s
e
o
v
er
th
e
in
ter
v
al.
I
n
th
is
p
er
s
p
ec
tiv
e,
a
MA
B
m
eth
o
d
is
s
ig
n
if
ican
t
f
o
r
o
u
r
d
elin
q
u
en
t
s
u
b
s
eq
u
en
tl
y
an
u
m
B
S
u
s
es
an
in
ad
eq
u
ate
ac
c
u
s
to
m
ed
o
f
b
ea
m
s
in
s
tan
tan
eo
u
s
ly
.
3
.
2
.
Co
m
m
un
ica
t
io
n o
n num
er
o
us
f
a
ca
des
T
h
e
Alg
o
r
ith
m
s
m
atch
es
th
e
is
s
u
es
o
f
m
m
W
av
e
UAV
c
o
m
m
u
n
icatio
n
o
n
n
u
m
er
o
u
s
f
ac
ad
es
ar
e
cr
itically
id
en
tifie
d
f
o
r
in
s
t
an
ce
F
ig
u
r
e
2
id
en
tifie
s
e
v
er
last
in
g
o
b
s
tr
u
ctio
n
s
(
s
tr
u
ctu
r
es),
r
ec
u
r
r
en
tly
o
b
s
tr
u
cted
r
eg
io
n
s
o
win
g
to
tem
p
o
r
ar
y
lo
ck
ag
e
’
s
(
ca
s
e
in
e
x
am
p
le
au
to
m
o
b
ile
s
tatio
n
s
)
e
x
p
en
d
i
n
g
lear
n
in
g
;
F
ig
u
r
e
3
d
ep
icts
th
at
co
n
tr
o
ls
tr
af
f
ic
c
o
n
f
i
g
u
r
atio
n
s
to
h
av
e
s
ch
em
e
ap
titu
d
e
m
a
x
im
izatio
n
b
y
p
r
o
v
id
e
d
th
at
g
r
an
d
e
r
tr
ea
tm
en
t
(
lik
e
in
t
h
is
ca
s
e
ap
p
o
r
tio
n
m
en
t
o
f
s
u
p
p
lem
en
tar
y
b
ea
m
s
)
in
z
o
n
es
with
weig
h
tier
cir
cu
latio
n
im
p
o
r
tan
t
en
h
an
ce
d
co
m
m
u
n
icatio
n
f
atalities
[
2
6
]
.
W
ith
in
th
is
lear
n
in
g
s
ec
tio
n
,
th
e
s
tu
d
y
ad
o
p
ted
th
e
alg
o
r
ith
m
th
at
f
ir
s
t
to
o
k
an
attem
p
t
to
d
ev
elo
p
a
n
d
s
im
u
l
ate
f
ast
lear
n
in
g
wh
er
e
th
e
e
n
v
ir
o
n
m
e
n
t
o
f
UAVs
ar
e
s
tu
d
ied
an
d
r
ec
o
g
n
ized
b
y
th
e
UAV
in
th
e
lin
e
o
f
ar
r
iv
al
to
th
e
b
ase
s
tatio
n
,
w
h
ile
s
u
p
p
o
s
in
g
th
es
e
ar
r
iv
als
ar
e
o
b
tain
ab
le
o
f
th
e
o
cc
asio
n
o
f
th
is
d
aily
,
th
e
s
tu
d
y
an
ticip
ated
ac
tio
n
s
to
ca
teg
o
r
ize
an
d
ac
q
u
ir
e
f
r
o
m
s
u
ch
co
n
f
i
g
u
r
atio
n
s
[
2
7
]
.
Fig
u
r
e
3
.
I
m
p
ac
t o
n
th
e
r
ec
ei
v
ed
b
m
d
ata
f
o
r
ar
r
iv
al
r
ate
o
n
cu
m
u
lativ
e
r
ec
eiv
e
d
s
p
ec
if
ied
d
ata
3
.
3
.
T
he
g
en
er
a
l
t
ra
f
f
ic
pa
t
t
er
ns
T
h
is
is
s
ig
n
if
ican
t
s
in
ce
m
m
-
W
av
e
B
S
m
ig
h
t
co
m
m
u
n
icate
in
s
tan
tan
eo
u
s
ly
ab
o
v
e
a
r
estricte
d
q
u
an
tity
o
f
b
ea
m
s
.
T
h
is
lim
it
is
s
u
b
ject
to
th
e
h
ar
d
war
e
tr
aits
,
th
e
m
m
W
av
e
n
etwo
r
k
,
an
d
th
e
b
ea
m
f
o
r
m
in
g
d
r
ill
last
ly
F
ig
u
r
e
4
,
o
n
th
e
lis
t
it
co
n
jectu
r
es
tr
af
f
ic
d
es
ig
n
s
f
r
o
m
th
e
s
itu
atio
n
(
lik
e
th
e
a
u
t
o
m
o
b
ile’
s
r
o
u
te
o
f
o
n
s
et)
an
d
c
h
o
o
s
es
th
e
s
u
p
er
la
tiv
e
b
ea
m
s
.
Fig
u
r
e
2
,
T
im
e
s
tep
s
ag
ain
s
t
th
e
d
is
s
im
ilar
cu
m
u
lativ
e
r
eg
r
ets
n
o
te
wh
er
e
th
e
o
r
a
n
g
e
c
o
lo
r
d
e
p
i
cts
t
r
ad
itio
n
al
m
eth
o
d
s
,
b
r
o
wn
s
h
o
ws
a
s
in
g
le
-
d
im
en
s
io
n
al,
g
r
ee
n
co
lo
r
t
h
e
2
-
Dim
en
s
io
n
al
an
d
b
r
o
wn
with
th
e
3
-
d
im
en
s
io
n
al
alg
o
r
ith
m
ic
s
im
u
latio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2
2
5
2
-
8
8
1
4
A
3
-
d
imen
s
io
n
a
l fa
s
t m
a
ch
in
e
lea
r
n
in
g
a
lg
o
r
ith
m
fo
r
mo
b
ile
u
n
ma
n
n
e
d
a
eria
l v
eh
icle
…
(
Wa
s
s
w
a
S
h
a
fik
)
35
Fig
u
r
e
4
.
I
m
p
ac
t o
n
th
e
ar
r
iv
al
r
ate
λ
o
f
d
ata
b
ased
o
n
th
e
cu
m
u
lativ
e
r
ec
eiv
ed
d
ata
3
.
4
.
A
m
o
un
t
o
f
a
s
s
is
t
ed
a
uto
m
o
bil
es,
a
nd
t
y
pica
l le
a
rning
in
t
er
v
a
l
T
h
e
en
ac
tm
en
t
m
etr
ics
r
ec
y
c
led
in
th
e
esti
m
atio
n
s
ar
e
co
llectiv
e
an
d
ac
cu
m
u
lativ
e
c
o
n
v
en
tio
n
al
r
ec
o
r
d
s
,
t
h
e
n
u
m
b
e
r
o
f
ass
is
ted
-
to
a
u
to
m
o
b
iles
,
an
d
n
o
r
m
al
lear
n
in
g
in
ter
v
als.
T
h
e
co
llec
tiv
e
ac
k
n
o
wled
g
ed
d
ata
is
d
is
cr
ete
as
th
e
r
ec
o
r
d
s
r
ec
eiv
ed
o
v
e
r
co
m
p
lete
ly
th
e
au
to
m
o
b
iles
in
th
e
s
ch
em
e
in
r
etr
o
t.
A
co
llectiv
e
ac
k
n
o
wled
g
e
d
d
ata
is
d
is
cr
ete
as
th
e
d
ata
ac
k
n
o
wled
g
ed
co
m
p
lete
o
v
e
r
all
th
e
UAVs
in
th
e
m
eth
o
d
in
r
etr
o
t.
T
h
e
ac
k
n
o
wled
g
ed
d
ata
ar
e
d
elin
ea
ted
eq
u
ally
as
th
e
d
ata
ac
ce
p
ted
co
m
p
letely
v
ia
th
e
UAVs
i
n
th
e
s
ch
em
e
u
p
to
a
s
p
ec
if
ie
d
p
er
io
d
d
e
p
icted
in
[
2
8
]
.
3
.
5
.
Anim
a
t
e
re
g
ula
r
t
r
a
f
f
ic
a
rr
i
v
es
Usi
n
g
th
e
f
o
c
al
ap
p
r
o
ac
h
to
f
o
cu
s
o
n
c
o
n
s
cio
u
s
d
a
y
-
to
-
d
ay
tr
af
f
ic
ar
r
a
n
g
em
e
n
t:
T
h
e
p
r
ev
i
o
u
s
ass
es
s
m
en
t
s
tay
ed
co
n
s
tr
u
cte
d
o
n
th
e
ch
a
r
ac
ter
is
tic
tr
a
f
f
ic
d
esig
n
d
em
o
n
s
tr
ated
.
B
ec
au
s
e
o
f
th
e
o
p
tim
izin
g
m
ag
n
itu
d
e,
r
ep
r
esen
tativ
e
tr
af
f
ic
-
f
lo
ws
d
o
n
o
t
an
n
e
x
atio
n
th
e
in
s
tan
tan
eo
u
s
v
a
r
iatio
n
s
in
t
r
af
f
ic
ar
r
an
g
em
e
n
ts
th
at
ar
e
d
is
tin
g
u
is
h
ab
le
in
s
e
n
tien
t
cir
cu
latio
n
d
escr
ip
ti
o
n
illu
s
tr
ated
in
F
ig
u
r
e
5
.
C
o
n
s
e
q
u
en
tly
,
t
h
e
s
tu
d
y
f
u
r
th
er
n
o
ted
th
at
th
e
in
v
esti
g
atio
n
al
co
n
s
cio
u
s
tr
af
f
ic
f
lo
w
in
tellig
en
ce
o
f
Go
o
g
le
f
o
r
two
d
ay
s
u
s
in
g
h
alf
an
h
o
u
r
i
n
ter
v
al
.
T
h
e
s
tu
d
y
f
u
r
th
er
n
o
u
r
is
h
ed
th
e
g
o
tten
s
tatis
tics
to
th
e
s
im
u
lan
t
to
esti
m
ate
th
e
en
ac
tm
en
t
o
f
alg
o
r
ith
m
s
in
a
c
o
n
s
cio
u
s
s
tr
ea
m
o
f
tr
af
f
ic
s
itu
atio
n
s
as p
er
[
2
9
]
.
Fig
u
r
e
5
.
Ag
g
r
eg
ate
r
ec
eiv
ed
d
ata
f
o
r
a
r
r
iv
al
r
ate
with
λ
an
d
β
3
.
6
.
T
he
im
pa
ct
o
f
blo
cka
g
es
T
h
e
I
m
p
ac
t o
f
b
lo
c
k
ag
es
was
f
u
r
th
er
i
n
v
esti
g
ated
p
er
t
h
e
cu
m
u
lativ
e
r
ec
eiv
ed
d
ata.
T
o
c
o
m
p
u
te
it
in
p
er
ce
n
tag
e,
a
n
y
p
er
ce
n
tag
e
o
f
p
er
m
an
e
n
t
b
lo
c
k
ag
es,
th
e
p
r
o
p
o
s
ed
m
o
d
el
o
u
tp
er
f
o
r
m
ed
all
o
th
er
cu
r
r
en
cies
th
at
ar
e
n
o
t
o
p
tim
u
m
p
r
o
ce
d
u
r
es.
I
n
c
r
ea
s
in
g
ac
k
n
o
wled
g
ed
d
ata
ac
c
o
m
p
lis
h
ed
b
y
o
u
r
alg
o
r
ith
m
s
lies
am
o
n
g
s
t
1
4
.
5
9
%
an
d
1
9
.
4
9
%
co
m
p
lex
th
an
th
at
ac
c
o
m
p
lis
h
ed
b
y
th
e
s
u
b
s
eq
u
e
n
t
-
s
u
p
er
lati
v
e
alg
o
r
ith
m
u
p
p
e
r
co
n
f
id
en
ce
b
o
u
n
d
m
o
d
el
p
o
r
tr
ay
ed
b
elo
w
th
e
o
u
tm
o
d
ed
r
ep
r
o
d
u
ctio
n
s
.
Fu
r
th
er
m
o
r
e,
th
e
m
o
d
el
f
in
alize
d
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
10
,
No
.
1
,
Ma
r
c
h
2
0
2
1
:
28
–
38
36
r
esu
lts
d
iv
er
g
e
f
r
o
m
th
at
o
f
o
p
tim
al
m
er
ely
b
y
at
m
o
s
t 3
.
9
1
%.
L
astl
y
o
n
th
e
li
s
t a
r
e
th
e
v
a
r
iatio
n
s
in
th
e
g
r
ap
h
d
o
m
in
o
in
f
lu
e
n
ce
f
r
o
m
t
h
e
q
u
an
tity
o
f
UAVs
in
th
e
s
ch
e
m
e,
th
e
c
o
llectiv
e
ac
k
n
o
wled
g
ed
d
ata
escalatio
n
s
with
th
e
n
u
m
b
er
o
f
UAVs
d
ep
ictin
g
th
e
av
er
ag
e
d
ata
r
ec
eiv
ed
th
u
s
p
r
o
p
er
ad
m
is
s
io
n
[
3
0
]
.
T
h
e
I
m
p
ac
t
o
f
b
lo
ck
ag
es
was
f
u
r
t
h
er
in
v
esti
g
ated
p
er
t
h
e
cu
m
u
lativ
e
r
e
ce
iv
ed
d
ata.
T
o
co
m
p
u
te
it
in
p
er
ce
n
ta
g
e,
an
y
p
er
ce
n
tag
e
o
f
p
e
r
m
an
e
n
t
b
l
o
ck
ag
es,
th
e
p
r
o
p
o
s
ed
m
o
d
el
o
u
tp
er
f
o
r
m
ed
all
o
th
er
c
u
r
r
en
t
n
o
n
-
o
p
tim
al
p
r
o
ce
d
u
r
es
F
ig
u
r
e
6
.
T
h
e
in
cr
ea
s
in
g
r
ec
eiv
ed
d
ata
ac
co
m
p
li
s
h
ed
b
y
o
u
r
alg
o
r
it
h
m
s
lies
a
m
o
n
g
s
t
1
4
:5
9
%
an
d
1
9
:4
9
%
to
UC
B
m
o
d
el
d
e
p
icted
u
n
d
er
th
e
tr
a
d
itio
n
al
m
o
d
e
ls
in
F
ig
u
r
e
7
.
Fu
r
t
h
er
m
o
r
e,
t
h
e
m
o
d
el
f
in
alize
d
th
e
r
esu
lts
d
iv
er
g
e
f
r
o
m
t
h
at
o
f
o
p
tim
al
m
er
el
y
b
y
at
m
o
s
t 3
:
9
1
%.
Fig
u
r
e
6
.
Data
f
o
r
1
7
2
8
0
0
m
i
n
u
tes o
f
th
e
liv
e
d
aily
tr
af
f
ic
p
a
tter
n
Fig
u
r
e
7
.
I
m
p
ac
t o
n
n
u
m
b
er
o
f
v
eh
icles p
er
tim
e
4.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
p
r
esen
te
d
p
er
f
o
r
m
an
ce
ev
alu
atio
n
b
ased
o
n
t
h
e
em
b
lem
atic
tr
af
f
ic
p
atter
n
s
,
d
u
e
to
th
e
av
er
ag
in
g
co
n
s
eq
u
en
ce
,
ad
d
r
e
s
s
ed
a
ch
allen
g
e
o
f
b
ea
m
s
elec
tio
n
at
m
illi
m
eter
W
av
e
B
S
u
s
in
g
m
u
lti
-
ar
m
e
d
B
an
d
its
wh
er
e
th
e
af
ter
m
at
h
o
f
th
e
ass
o
r
tm
en
t
is
tr
e
m
en
d
o
u
s
ly
r
elian
t
o
n
th
e
tr
af
f
ic
-
f
lo
ws
an
d
th
e
o
b
s
tr
u
ctio
n
s
lik
e
s
tr
u
ctu
r
es
in
th
e
s
y
s
tem
.
Ob
jectiv
ely
,
th
e
s
t
u
d
y
p
r
o
p
o
s
ed
a
Fas
t
Ma
ch
in
e
L
ea
r
n
in
g
,
a
s
em
i
-
alg
o
r
ith
m
p
er
t
h
e
m
u
lti
-
ar
m
ed
b
an
d
its
th
at
o
p
e
r
ate
o
n
n
e
g
lig
ib
le
b
ac
k
g
r
o
u
n
d
n
etwo
r
k
f
ac
ts
o
n
lin
e
lear
n
in
g
as
elu
cid
ated
.
T
h
is
s
tu
d
y
d
id
n
o
t
f
o
cu
s
o
n
av
iatio
n
p
ath
s
s
tan
d
ar
d
s
ag
ain
s
t
UAVs,
UAV
in
ter
n
atio
n
al
p
o
licies
to
o
,
ass
u
m
p
tio
n
s
,
an
d
r
estrictio
n
o
n
UAVs
in
th
e
g
eo
m
etr
ic
p
r
o
b
lem
a
p
p
r
o
ac
h
.
C
o
m
b
in
i
n
g
UAV
lo
ca
lizatio
n
an
d
b
ea
m
s
elec
tio
n
t
o
m
ax
im
ize
E
f
f
icien
cy
h
as
n
o
t
b
ee
n
c
o
n
s
id
er
ed
.
T
h
is
ar
ticle
h
as
en
th
u
s
ed
u
s
m
o
r
e
t
o
f
u
r
th
er
len
g
th
en
th
is
r
esear
ch
to
in
v
esti
g
ate
th
e
d
ee
p
an
aly
s
is
o
f
UAV
s
wh
en
u
s
ed
as
a
b
ase
s
tatio
n
an
d
its
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2
2
5
2
-
8
8
1
4
A
3
-
d
imen
s
io
n
a
l fa
s
t m
a
ch
in
e
lea
r
n
in
g
a
lg
o
r
ith
m
fo
r
mo
b
ile
u
n
ma
n
n
e
d
a
eria
l v
eh
icle
…
(
Wa
s
s
w
a
S
h
a
fik
)
37
ass
o
ciate
d
is
s
u
es.
I
n
th
e
f
u
tu
r
e,
th
e
p
r
o
p
o
s
ed
m
o
d
el
is
to
b
e
s
u
b
jecte
d
to
o
th
e
r
alg
o
r
ith
m
s
b
esid
es
th
o
s
e
th
at
h
av
e
b
ee
n
u
s
ed
in
p
er
s
p
ec
ti
v
e
to
b
ea
m
o
r
th
o
g
o
n
ality
,
l
o
ca
tio
n
r
ec
o
r
d
in
g
,
an
d
s
itu
a
tio
n
o
f
c
o
-
lo
ca
ted
au
to
m
o
b
iles
,
in
ter
f
e
r
en
ce
a
n
d
n
u
m
b
er
o
f
s
elec
ted
b
ea
m
s
.
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
d
ep
ar
tm
en
t
m
em
b
er
s
o
f
C
o
m
p
u
ter
E
n
g
in
ee
r
in
g
,
Yaz
d
Un
iv
er
s
ity
f
o
r
s
p
o
n
s
o
r
in
g
th
is
wo
r
k
a
n
d
e
d
i
tin
g
f
o
r
u
s
to
h
av
e
th
is
p
a
p
er
q
u
ality
h
elp
e
d
in
UAV
ch
a
n
n
el
m
ea
s
u
r
em
e
n
ts
an
d
p
er
f
ec
tio
n
.
RE
F
E
R
E
NC
E
S
[1
]
P
.
C.
S
a
ib
a
b
u
,
H.
S
a
i
,
S
.
Ya
d
a
v
,
a
n
d
C.
R.
S
rin
iv
a
sa
n
,
“
S
y
n
th
e
sis
o
f
a
m
o
d
e
l
p
re
d
ictiv
e
c
o
n
tr
o
ll
e
r
f
o
r
a
n
id
e
n
ti
fie
d
m
o
d
e
l
o
f
M
IM
O
p
ro
c
e
ss
,
”
In
t
.
J
.
El
e
c
tr.
En
g
.
Co
m
p
u
t
.
S
c
i
.
,
v
o
l.
1
7
,
n
o
.
2
,
p
p
.
9
5
0
–
9
5
6
,
2
0
2
0
.
[2
]
W.
S
h
a
fi
k
a
n
d
S
.
A.
M
o
sta
fa
v
i,
“
Kn
o
wle
d
g
e
En
g
in
e
e
rin
g
o
n
In
ter
n
e
t
o
f
T
h
in
g
s
th
r
o
u
g
h
Re
i
n
fo
rc
e
m
e
n
t
Lea
rn
in
g
,
”
In
t
J
Co
m
p
u
t
Ap
p
l
,
v
o
l.
9
7
5
,
p
p
.
8
8
8
7
,
2
0
2
0
.
[3
]
Z.
Qin
,
Q.
Yu
,
a
n
d
M
.
J.
B
u
e
h
ler,
“
M
a
c
h
in
e
lea
rn
in
g
m
o
d
e
l
fo
r
fa
st
p
re
d
icti
o
n
o
f
t
h
e
n
a
t
u
ra
l
fre
q
u
e
n
c
ies
o
f
p
ro
tei
n
m
o
lec
u
les
,
”
RS
C
A
d
v
.
,
v
o
l
.
1
0
,
n
o
.
2
8
,
p
p
.
1
6
6
0
7
–
1
6
6
1
5
,
2
0
2
0
.
[4
]
S
.
M
o
sta
fa
v
i
a
n
d
W.
S
h
a
fik
,
“
F
o
g
C
o
m
p
u
ti
n
g
Arc
h
it
e
c
t
u
re
s,
P
ri
v
a
c
y
a
n
d
S
e
c
u
ri
ty
S
o
lu
ti
o
n
s,”
J
.
C
o
mm
u
n
.
T
e
c
h
n
o
l.
El
e
c
tro
n
.
C
o
mp
u
t.
S
c
i.
,
v
o
l
.
2
4
,
p
p
.
1
–
1
4
,
2
0
1
9
.
[5
]
K.
Wan
g
,
J.
L
u
o
,
Y.
Wei,
K.
W
u
,
J.
Li
,
a
n
d
Z.
C
h
e
n
,
“
P
ra
c
ti
c
a
l
a
p
p
li
c
a
ti
o
n
o
f
m
a
c
h
in
e
lea
rn
i
n
g
o
n
fa
st
p
h
a
se
e
q
u
il
i
b
riu
m
c
a
lcu
lati
o
n
s i
n
c
o
m
p
o
siti
o
n
a
l
re
se
rv
o
ir
sim
u
latio
n
s,”
J
.
Co
mp
u
t.
Ph
y
s.
,
v
o
l.
4
0
1
,
p
p
.
1
0
9
0
1
3
,
2
0
2
0
.
[6
]
W.
S
h
a
fik
,
M
.
M
a
ti
n
k
h
a
h
,
a
n
d
M
.
N.
S
a
n
d
a
,
“
Ne
two
rk
Re
so
u
rc
e
M
a
n
a
g
e
m
e
n
t
Driv
e
s
M
a
c
h
i
n
e
Le
a
rn
in
g
:
A
S
u
rv
e
y
a
n
d
F
u
tu
re
Re
se
a
rc
h
Dire
c
ti
o
n
,
”
J
.
Co
mm
u
n
.
T
e
c
h
n
o
l
.
El
e
c
tro
n
.
C
o
mp
u
t.
S
c
i.
,
v
o
l.
3
0
,
p
p
.
1
–
1
5
,
2
0
2
0
.
[7
]
W.
S
h
a
fi
k
,
M
.
M
a
ti
n
k
h
a
h
,
M
.
A
sa
d
i,
Z.
Ah
m
a
d
i,
a
n
d
Z.
Ha
d
iy
a
n
,
“
A
S
tu
d
y
o
n
I
n
tern
e
t
o
f
Th
i
n
g
s
P
e
rfo
rm
a
n
c
e
Ev
a
lu
a
ti
o
n
,
”
J
.
C
o
mm
u
n
.
T
e
c
h
n
o
l
.
El
e
c
tro
n
.
Co
m
p
u
t.
S
c
i
.
,
v
o
l.
2
8
,
p
p
.
1
–
1
9
,
2
0
2
0
.
[8
]
W.
S
h
a
fi
k
,
S
.
M
.
M
a
ti
n
k
h
a
h
,
a
n
d
M
.
G
h
a
se
m
a
z
a
d
e
,
“
F
o
g
-
M
o
b
il
e
Ed
g
e
P
e
rfo
rm
a
n
c
e
E
v
a
lu
a
ti
o
n
a
n
d
An
a
l
y
s
is
o
n
In
tern
e
t
o
f
Th
in
g
s,”
J
.
A
d
v
.
Res
.
M
o
b
.
Co
m
p
u
t
.
,
v
o
l
.
1
,
n
o
.
3
,
2
0
1
9
.
[9
]
W.
S
h
a
fik
,
S
.
M
.
M
a
ti
n
k
h
a
h
,
a
n
d
M
.
G
h
a
se
m
z
a
d
e
h
,
“
In
tern
e
t
o
f
Th
in
g
s
-
Ba
se
d
En
e
r
g
y
M
a
n
a
g
e
m
e
n
t,
C
h
a
ll
e
n
g
e
s,
a
n
d
S
o
lu
ti
o
n
s in
S
m
a
rt
Cit
ies
,
”
J
.
Co
mm
u
n
.
T
e
c
h
n
o
l.
El
e
c
tro
n
.
Co
mp
u
t.
S
c
i
.
,
v
o
l.
2
7
,
p
p
.
1
–
1
1
,
2
0
2
0
.
[1
0
]
A.
P
a
tra
a
n
d
A.
S
u
re
sh
,
“
B
LAZE
:
Blaz
in
g
F
a
st
P
ri
v
a
c
y
-
P
re
se
rv
in
g
M
a
c
h
in
e
Lea
rn
i
n
g
,
”
ArXi
v
Pre
p
r.
ArXi
v
2
0
0
5
0
9
0
4
2
,
2
0
2
0
.
[1
1
]
W.
S
h
a
fi
k
,
S
.
M
.
M
a
ti
n
k
h
a
h
,
a
n
d
M
.
G
h
a
se
m
z
a
d
e
h
,
“
A
F
a
st
M
a
c
h
in
e
Lea
rn
in
g
fo
r
5
G
Be
a
m
S
e
lec
ti
o
n
fo
r
Un
m
a
n
n
e
d
Ae
ria
l
Ve
h
icle
Ap
p
li
c
a
ti
o
n
s,”
I
n
f.
S
y
st.
T
e
lec
o
mm
u
n
.
,
p
p
.
2
6
2
,
2
0
2
0
.
[1
2
]
W.
S
h
a
fik
a
n
d
S
.
M
.
M
a
ti
n
k
h
a
h
,
“
P
riv
a
c
y
Iss
u
e
s
in
S
o
c
ial
Web
o
f
Th
i
n
g
s,”
in
2
0
1
9
5
th
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
W
e
b
Res
e
a
rc
h
(ICW
R),
2
0
1
9
,
p
p
.
2
0
8
–
2
1
4
.
[1
3
]
L.
Jia
n
g
a
n
d
H
.
Ja
fa
rk
h
a
n
i,
“
M
u
l
ti
-
Us
e
r
An
a
lo
g
Be
a
m
fo
rm
in
g
i
n
M
il
li
m
e
ter
Wav
e
M
IM
O
S
y
ste
m
s
Ba
se
d
o
n
P
a
t
h
An
g
le In
f
o
rm
a
ti
o
n
,
”
IEE
E
T
ra
n
s.
W
ire
l.
Co
mm
u
n
.
,
v
o
l.
1
8
,
n
o
.
1
,
p
p
.
6
0
8
–
6
1
9
,
Ja
n
2
0
1
9
.
[1
4
]
S
.
CHA
TT
OPADH
YA
Y,
“
No
n
-
o
rth
o
g
o
n
a
l
m
u
lt
ip
le
a
c
c
e
ss
(NO
M
A)
fo
r
5
G
n
e
two
rk
s
a
n
d
i
ts
a
p
p
li
c
a
ti
o
n
i
n
u
n
m
a
n
n
e
d
a
e
rial
v
e
h
icle
a
ss
isted
c
o
m
m
u
n
ica
ti
o
n
,
”
1
6
-
Ap
r
-
2
0
1
9
.
[On
li
n
e
]
.
Av
a
il
a
b
le:
h
tt
p
s:/
/www
.
p
o
li
tes
i.
p
o
li
m
i.
it
/h
a
n
d
le/1
0
5
8
9
/
1
4
7
3
2
9
.
[Ac
c
e
ss
e
d
:
0
6
-
Ju
l
-
2
0
1
9
].
[1
5
]
H.
El
-
S
a
y
e
d
,
M
.
C
h
a
q
fa
,
S
.
Zea
d
a
ll
y
,
a
n
d
D.
P
u
tk
a
l
,
“
A
Traffic
-
A
wa
re
Ap
p
ro
a
c
h
fo
r
En
a
b
li
n
g
Un
m
a
n
n
e
d
Ae
rial
Ve
h
icle
s
(UA
Vs
)
in
S
m
a
rt
Cit
y
S
c
e
n
a
rio
s
,”
IEE
E
Acc
e
ss
,
v
o
l.
7
,
p
p
.
8
6
2
9
7
-
8
6
3
0
5
,
2
0
1
9
.
[1
6
]
F
.
Xu
,
T
.
Ho
n
g
,
J.
Zh
a
o
,
a
n
d
T.
Ya
n
g
,
“
De
tec
ti
o
n
a
n
d
i
d
e
n
ti
fica
ti
o
n
tec
h
n
o
l
o
g
y
o
f
ro
t
o
r
u
n
m
a
n
n
e
d
a
e
rial
v
e
h
icle
s
in
5
G
sc
e
n
e
,
”
In
t.
J
.
Distrib
.
S
e
n
s.
Ne
tw
.
,
v
o
l.
1
5
,
n
o
.
6
,
J
u
n
.
2
0
1
9
.
[1
7
]
R.
P
ě
n
ičk
a
,
J.
F
a
ig
l,
a
n
d
M
.
S
a
sk
a
,
“
P
h
y
sic
a
l
Orie
n
tee
rin
g
P
ro
b
le
m
fo
r
Un
m
a
n
n
e
d
Ae
rial
Ve
h
icle
Da
ta
Co
ll
e
c
ti
o
n
P
lan
n
i
n
g
in
En
v
iro
n
m
e
n
ts W
it
h
Ob
sta
c
les
,
”
IEE
E
Ro
b
o
t
.
Au
to
m.
L
e
tt
.
,
v
o
l.
4
,
n
o
.
3
,
p
p
.
3
0
0
5
–
3
0
1
2
,
Ju
l.
2
0
1
9
.
[1
8
]
M
e
n
g
,
H.,
S
h
a
fi
k
,
W.
,
M
a
ti
n
k
h
a
h
,
S
.
M
.
a
n
d
Ah
m
a
d
,
Z.
,
"
A
5
G
Be
a
m
S
e
lec
ti
o
n
M
a
c
h
i
n
e
Lea
rn
i
n
g
Al
g
o
rit
h
m
f
o
r
Un
m
a
n
n
e
d
Ae
rial
Ve
h
icle
Ap
p
li
c
a
ti
o
n
s
,"
W
ire
les
s Co
mm
u
n
ica
ti
o
n
s a
n
d
M
o
b
i
le Co
mp
u
ti
n
g
,
v
o
l.
2
0
2
0
,
2
0
2
0
.
[1
9
]
C.
An
tó
n
-
Ha
ro
a
n
d
X.
M
e
stre
,
“
Lea
rn
in
g
a
n
d
Da
ta
-
Driv
e
n
Be
a
m
S
e
lec
ti
o
n
f
o
r
m
m
Wav
e
Co
m
m
u
n
ica
ti
o
n
s:
An
An
g
le o
f
Arriv
a
l
-
Ba
se
d
Ap
p
ro
a
c
h
,
”
IEE
E
Acc
e
ss
,
v
o
l.
7
,
p
p
.
2
0
4
0
4
–
2
0
4
1
5
,
2
0
1
9
.
[2
0
]
M
.
G
io
rd
a
n
i,
M
.
P
o
les
e
,
A.
Ro
y
,
D.
Ca
sto
r,
a
n
d
M
.
Z
o
rz
i,
“
S
tan
d
a
l
o
n
e
a
n
d
No
n
-
S
tan
d
a
lo
n
e
Be
a
m
M
a
n
a
g
e
m
e
n
t
fo
r
3
G
P
P
NR at mm
Wav
e
s,”
IEE
E
C
o
mm
u
n
.
M
a
g
.
,
v
o
l.
5
7
,
n
o
.
4
,
p
p
.
1
2
3
–
1
2
9
,
Ap
r
.
2
0
1
9
.
[2
1
]
Y.
Wan
g
,
Q.
Li
,
J.
Jia
o
,
S
.
Wu
,
a
n
d
Q.
Z
h
a
n
g
,
“
ARM:
Ad
a
p
t
iv
e
R
a
n
d
o
m
-
S
e
lec
ted
M
u
lt
i
-
Be
a
m
fo
rm
in
g
Esti
m
a
ti
o
n
S
c
h
e
m
e
fo
r
S
a
telli
te
-
Ba
se
d
In
ter
n
e
t
o
f
T
h
in
g
s,”
IEE
E
Acc
e
ss
,
v
o
l.
7
,
p
p
.
6
3
2
6
4
–
6
3
2
7
6
,
2
0
1
9
.
[2
2
]
Y.
Wan
g
,
Q.
Li
,
J.
Jia
o
,
S
.
Wu
,
a
n
d
Q.
Z
h
a
n
g
,
“
ARM:
Ad
a
p
t
i
v
e
R
a
n
d
o
m
-
S
e
lec
ted
M
u
lt
i
-
Be
a
m
fo
rm
in
g
Esti
m
a
ti
o
n
S
c
h
e
m
e
fo
r
S
a
telli
te
-
Ba
se
d
In
ter
n
e
t
o
f
T
h
in
g
s,”
IEE
E
Acc
e
ss
,
v
o
l.
7
,
p
p
.
6
3
2
6
4
–
6
3
2
7
6
,
2
0
1
9
.
[2
3
]
R.
P
a
l,
A.
K.
C
h
a
it
a
n
y
a
,
a
n
d
K.
V.
S
rin
i
v
a
s,
"
L
o
w
-
Co
m
p
le
x
it
y
Be
a
m
S
e
lec
ti
o
n
Alg
o
ri
th
m
s
fo
r
M
il
li
m
e
ter
-
Wav
e
Be
a
m
sp
a
c
e
M
IM
O S
y
ste
m
s,"
IE
EE
Co
mm
u
n
.
L
e
tt
.
,
v
o
l
.
2
3
,
n
o
.
4
,
p
p
.
7
6
8
–
7
7
1
,
Ap
r.
2
0
1
9
.
[2
4
]
S
.
M
.
M
a
ti
n
k
h
a
h
,
W.
S
h
a
fik
,
a
n
d
M
.
G
h
a
se
m
z
a
d
e
h
,
“
Eme
rg
in
g
Artifi
c
ial
In
telli
g
e
n
c
e
Ap
p
li
c
a
ti
o
n
:
Re
in
fo
rc
e
m
e
n
t
Lea
rn
in
g
Iss
u
e
s
o
n
C
u
rre
n
t
In
tern
e
t
o
f
Th
i
n
g
s,”
in
1
6
t
h
in
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
in
in
fo
rm
a
ti
o
n
k
n
o
wled
g
e
a
n
d
T
e
c
h
n
o
l
o
g
y
(ikt
2
0
1
9
)
,
2
0
1
9
.
[2
5
]
M
.
M
e
n
g
e
t
a
l
.
,
“
Be
a
m
Ra
ste
r:
A
P
ra
c
ti
c
a
l
F
a
st
M
a
ss
iv
e
M
U
-
M
IM
O
S
y
ste
m
Wi
t
h
P
re
-
C
o
m
p
u
te
d
P
r
e
c
o
d
e
rs,”
IEE
E
T
ra
n
s.
M
o
b
.
C
o
mp
u
t
.
,
v
o
l.
1
8
,
n
o
.
5
,
p
p
.
1
0
1
4
–
1
0
2
7
,
M
a
y
2
0
1
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.