Beamforming Antennas for

Wireless Communications
Yikun Huang, Ph.D.
ECE/CCB
Yikun@cns.montana.edu
November 24 2003

Outline
Phased Array Antennas

Vector Antennas

Beamforming antennas for WLAN
Conclusion
Introduction



Beamforming and its applications
Beamforming antennas vs. omnidirectional antennas
Direction of arrival (DOA) estimation
Beamforming
Basic configurations: fixed array and adaptive array
smart antenna systems:switched array and adaptive array
DOA and polarization
super CART
3-loop and 2-loop vector antenna array
Direction of arrival (DOA) estimation
Vector antenna vs. phased array antenna
Infrastructure mode
An indoor WLAN design
Ad hoc mode
Ad hoc WLAN for rural area
Applications

Description

RADAR

Phased array RADAR; air traffic control; synthetic aperture
RADAR

SONAR

Source location and classification

Communications

Smart antenna systems; Directional transmission and
reception; sector broadcast in satellite communications

Imaging

Ultrasonic; optical; tomographic

Geophysical Exploration

Earth crust mapping; oil exploration

Astrophysical Exploration

High resolution imaging of universe

Biomedical

Neuronal spike discrimination; fetal heart monitoring;
tissue hyperthermia; hearing aids

Source: B.D.Van Veen and K.M. Buckley, University of Michigan, Beamforming: A
Versatile approach to spatial filtering,1988
Applications of beamforming technology
Phased array RADAR
Phased array spike sorting
0.139
0.544
Ey1n t ( )
1.2 104 0 t
0.056
0.205
Ey2n t ( )
1.2 104 0 t
0.042
0.187
Ey3n t ( )
1.2 104 0 t
Sorted
Spike of
individual
neurons.
1

2

3

4

1 6

5

6

7

8

9

1 4

1 5

1 3

1 2

1 1

1 0

0.139
0.534
Rn 3 t , ( )
1.2 10
4

0 t
0.183
0.539
Rn 5 t , ( )
1.2 10
4

0 t
0.147
0.534
Rn 7 t , ( )
1.2 10
4

0 t
0.147
0.534
Rn 9 t , ( )
1.2 10
4

0 t
0.183
0.539
Rn 11 t , ( )
1.2 10
4

0 t
0.139
0.534
Rn 13 t , ( )
1.2 10
4

0 t
0.14
0.534
Rn 1 t , ( )
1.2 10
4

0 t
0.148
0.534
Rn 15 t , ( )
1.2 10
4

0 t
Neuronal
spikes
recorded by
electrode
array
P
h
a
s
e
d

a
r
r
a
y

s
p
i
k
e

s
o
r
t
i
n
g

s
y
s
t
e
m

Center for Computational Biology, MSU
Patterns, beamwidth & Gain
Isotropic dipole
t
o
p

v
i
e
w
(
h
o
r
i
z
o
n
t
a
l
)

s
i
d
e

v
i
e
w
(
v
e
r
t
i
c
a
l
)

half-wave dipole beamformer
2 1/

Half-power
beam width
Half-power
beam width
Half-power
beam width
Main lobe
side lobes
nulls
2 1/
78
Beamformers vs. omnidirectional antennas
1) Beamformers have much higher Gain than omnidirectional antennas:
Increase coverage and reduce number of antennas!
Gain:
2
1
N
G
G
N
=
0
30
60
90
120
150
180
210
240
270
300
330
6
4
2
0
6
9.961 10
7

Field 6 0 , | , ( )
Field 2 0 , | , ( )
Field 1 0 , | , ( )
|
Beamformers vs. omnidirectional antennas
2) Beamformers can reject interference while omnidirectional
antennas cant: Improve SNR and system capacity!
3) Beamformers directionally send down link information to the
users while omnidirectional antennas cant: save energy!
user
interference
user
interference null
Beamformers vs. omnidirectional antennas
user
user
null
multipath
4) Beamformers provide N-fold diversity Gain of omnidirectional antennas:
increase system capacity(SDMA)
5) Beamformers suppress delay spread:improve signal quality
DOA estimation
kd

d
k k k
+ = + = sin sin
2
phase delay
1 2 3 4 5 6 7 N N-2 N-1 N-3


d
k k
d sin =
k

Plane wave
Beamforming
phase shifters
1 2 3 4 5 6 7 N N-2 N-1 N-3


k

A
1,,k
A
2,,k
A
3,,k
A
4,,k
A
5,,k
A
6,,k
A
7,,k
A
N-3,,k
A
N-2,,k
A
N-1,,k
A
N,,k
) sin )( (
,
kd N
k k N
+ = 1
phased array (fixed/adaptive) configurations-time domain
Basic phased array configurations
Narrowband
s
N
(k)
s
2
(k)
s
1
(k)
.
.
.
w*
N
w*
2
w*
1

) (k y
broadband
s
N
(k)
s
2
(k)
s
1
(k)
.
.
.

) (k y
w*
N,0
w*
N,1
w*
N,k-1
.

.

.

Z
-1
Z
-1

w*
2,0
w*
2,1
w*
2,k-1
.

.

.

Z
-1
Z
-1

w*
1,0
w*
1,1
w*
1,k-1
.

.

.

Z
-1
Z
-1

phased array (fixed/adaptive) configuration-frequency domain
Basic phased array configurations

s
N
(k)
s
2
(k)
s
1
(k)
.
.
.
-
+
I
F
F
T
MSE
F
F
T
w*
N
w*
2
w*
1

) (k y
) (t d
F
F
T
F
F
T
F
F
T
broadband
.
.
.
Smart antenna systems
Military
networks
Cellular
communication
networks
Wireless
local area
networks
switched array
adaptive array
switched array
adaptive array
switched array
adaptive array
Wi-Fi Data rate:11Mbps
3G Data rate:100kbps
Switched array (predetermined)
top view(horizontal)
Smart antenna systems
interference
user
1
2
3
4 5
6
7
8
9
10
11
12
13
14
15
16
user 1
Interference 1
top view(horizontal)
user 2
Smart antenna systems
Interference 2
Adaptive array


Smart antenna system
www.vivato.net
12
100

In door range
(Mixed Office)

11 Mbps: up to 300m
5.5 Mbps: up to 400m
2 Mbps: up to 500m
1 Mbps: up to 600m


Out door range
(outdoor to indoor)

11 Mbps: up to 1.00km
5.5 Mbps: up to 1.25km
2 Mbps: up to 2.00km
1 Mbps: up to 2.50km


Out door range
(outdoor to outdoor)

11 Mbps: up to 4.20km
5.5 Mbps: up to 5.10km
2 Mbps: up to 6.00km
1 Mbps: up to 7.20km

Active user per switch

100

Example: Vivato 2.4 GHz indoor & outdoor Wi-Fi Switches
(EIRP=44dBm;Gain=25 dBi;3-beam)
Polarization
circular
E
u

linear
q=0
E
|
E
u

ellipse
q=45
X
Y
Z


i
E

u
j i
e E sin
E
i
cos
|
u
E
|

E
u
q=90
E
|
SuperCART
Compact array radiolocation technology
Flam&Russell,Inc.,1990
U.S. Patent No., 5,300,885;1994
Frequency range: 2 30 MHz
Super CART
3-loop
V
6

V
4

V
3

V
1

V
2

V
5

Y
X
|
L
e
Z I V ) 0 (
0
=
L
e
Z I V ) (t
t
=
i
H z
0

E
I
i
E y
0

A
I
kbs0.5
b
2-loop
H
E
S
Steering vector
(

(
(
(
(

=
|
|
|
|
|
.
|

\
|
=

cos
e sin
sin
cos cos sin
sin
cos cos sin
h
h
e
e
j
z
x
z
y
0
0
0 0
0 0
0
4

H
i
i
E
0
0
=
1
2 2 2
= + +
z y x
e e e
1
2 2 2
= + +
z y x
h h h
Blind point
Vector antennas vs. spatial array antennas
Vector antennas measure: |,u,,q, and power simultaneously,
no phase shift device, or synchronization is needed.
Phased array antennas with omnidirectional element measure:
|,u, and power
Source: Nehorai,A.,University of Illinois at Chicago
Vector antennas vs. spatial array antennas
VA
SA
VA SA
Vector antennas vs. spatial array antennas
Phased array antennas: spatial ambiguities exist
2 2 1 1
f f sin sin =
1 2 3 4 5 6 7
k

1 2 3 4 5 6 7
1

P , , , , h , h , h , e , e , e
z y x z y x

Vector antenna: no ambiguities for DOA estimation
Vector antennas Vs. phased array antennas
Disadvantages of vector antennas
Cheap?
Can use hardware and software of existing communication
systems for performance?
f=2.4GHz, =0.125m; vector antenna size: 0.0125m ~ 0.063m
Phased array:ds /2=0.063m;L=(N-1)d: 0.188m-0.69m(N=412)
f=800MHz, =0.375m; antenna size: 0.04m ~ 0.19m
Phased array:ds /2=0.19m;L=(N-1)d: 0.56m-2.06m(N=412)
Low profile?

source:M.R. Andrews et al., Nature, Vol. 409(6818), 18 Jan. 2001, pp 316-318.
Working in scattering environment
(a) 2-dipole(monopole)
Low profile antennas with polarization diversity
(c) dipole-loop
(b) 2-loop
TDD/TDMA
Packet switching
A
AP1 AP2
user
Handoff between Aps
was not standardized
at the same time as
802.11b
Packet switching: 3 beam system
top view(horizontal)
i
i i
P
P P
d
1 1 +

=
P. Sanchis, et al. 02
i
P
1 i
P
1 + i
P


( )
( )

>
< +
< +
=
1 2 2 1
1 2
1 2 2 1
d d
d d
d d

i
i
i
DOA
), / ( /
), / (
), / ( /

max
max
max
An indoor WLAN design
A 4-story office building (including basement), high 30 m, wide 60m and long 100m. We plan
to install a Vivato switched array on the 3rd floor.
L=100m
h=30m
w=60m
Switched array
3
2
1
Basement
An indoor WLAN design
Data rate

1Mbps, 2Mbps, 5.5Mbps, 11Mbps
APs EIEP

44dBm
APs antenna Gain G
A
25 dBi
PC antenna Gain G
P
0 dBi
Shadowing 8dB

APs antenna receiving sensitivity S
min
-95dBm ,-92dBm, ,-89dBm, -86dBm

APs Noise floor -178dBm/Hz
Body/orientation loss 2dB
Soft partition attenuate factor (p= number)

p1.39 dB
Concrete-wall attenuate factor(q= number)

q2.38 dB
Average floor attenuation(floor number) 14.0dB(1),19.0dB(2),23.0dB(3),26.0dB(4)
Frequency 2.4GHz
Reference pathloss PL
0
(LOS/NLS, r=1m) 45.9dB/ 50.3dB
Pathloss exponent (LOS/NLS, r=1m) 2.1/3.0
Pathloss standard deviation o (LOS/NLS) 2.3dB/4.1dB
Average floor attenuation(floor number) 14.0dB(1),19.0dB(2),23.0dB(3),26.0dB(4)
Data of APs antenna is from www.vivato.net
An indoor WLAN design
Mean pathloss with s
min
:
P
G S EIRP L + =
min
o sd fl sm w allowable
L L L L L L PL =
Path loss model:
) log( ) (
0
0
10
r
r
PL r PL + =
al
PL r PL = ) (
The coverage ranges are:r=36m,29m,23m and 18m for date rate at 1Mbps, 2Mbps,
5.5Mbps and 11Mbps respectively
Allowable pathloss:
Case 1: user is on the 3
rd
floor: 3 concrete walls, 3 soft partitions
The coverage ranges are: r=176m,140m,111m and 88m for date rate at 1Mbps,
2Mbps, 5.5Mbps and 11Mbps respectively .
Case 2: user is in the basement : 3 floors; 2 concrete walls, 3 soft partitions
Beamforming antennas in ad hoc networks
P.Gupta and P.R. Kumar,00
t
h
r
o
u
g
h
p
u
t

o
b
t
a
i
n
e
d

b
y

e
a
c
h

n
o
d
e

|
|
.
|

\
|
n nlog
W
~
Beam-
forming
antennas
?
new
routing
protocol
new
channel
access
scheme
Beamforming antennas in ad hoc networks
interference
target
Phased patch
antenna
D.Lu and D.Rutledge,Caltech,02
Z
0
=50O
Z
0
=50O,L~/2 Z
0
=25O,L~/2
Series resonant patch array
Phased patch array
Beamforming antennas in ad hoc networks
Medium Access Control Protocol(CSMA/CA)
CSMA/CA:carrier sense multiple access/collision avoidance
( for omnidirectional antennas)
(Scheduled/On-demand) Packet routing
Neighbor discovery
No standard MAC protocols for directional antenna
Ad hoc networks may achieve better performance in some cases
using beamforming antennas.
No obvious improvement for throughput using beamforming antennas
Neighbor discovery become more complex using beamforming antennas.
Beamforming antennas can significantly increasing node and
network lifetime in ad hoc networks.
1) traditional exposed node
problem for omnidirectional
antennas
Channel access
Source:Y Ko et al., 00
A B C D E
RTS
CTS
DATA
ACK
RTS
CTS
DATA
DATA
DATA
ACK
A B C D E
RTS
CTS CTS
DATA
DATA
ACK
RTS
CTS
CTS
DATA
DATA
ACK
1) No coverage change. May save power.
2) B may not know the location of C.
The nodes
are
prohibit to
transmit or
receive
signals
The node
is free to
transmit or
receive
signals
The node is
blocked to
communicat
e with C
2) Omnidirectional and
directional antennas solve
the exposed node problem
Channel access
A B C D E
RTS
CTS
CTS
DATA
RTS
collision
deaf collision
A B C D E
RTS
CTS
DATA
DATA
RTS
3) beamforming antennas create new problems
Neighbor discovery
A
B
C
D
E
A
t
Nt
Hello
AP

Neighbors

A

B,C

B

A,C

C

A,B,E

D

E

E

C,D



Ad hoc WLAN for rural area
Conclusion
Beamforming antenna systems improve wireless
network performance
-increase system capacity
-improve signal quality
-suppress interference and noise
-save power
Beamforming antennas improve infrastructure
networks performance. They may improve ad hoc
networks performance. New MAC protocol
standards are needed.
Vector antennas may replace spatial arrays to
further improve beamforming performance