Chapter 6 : Antenna Arrays

• Introduction
• Two-Element Array
• N-element Linear Array: Uniform
Amplitude and Spacing
• N-element Linear Array: Directivity
• N-element Linear Array: Uniform spacing,
Non-uniform Amplitude
• Planar Array

1
 Antenna Array: Introduction
• Array is an assembly of antenna elements
arranged in an orderly fashion. The
elements are usually identical.
• Why array? When high gain and/or
narrow beam are required:
– Single element -> Wide beam (low directivity)
– Increasing size -> difficult to build and
expensive
– Useful especially when the element gain is low.

2
 Antenna Array: Introduction (2)
• Advantages
– Higher directivity
– Narrower beam
– Lower sidelobes
– Electronic steerable beam
• Types
– Fix direction
– Steerable : Mechanical or Electronic (phased
arrays)

3
 Antenna Array: Introduction (3)
• In an array of identical elements, there
are in general five controls that can be
used to shape the overall pattern of the
antenna:
1. Geometrical configuration (linear, circular, etc.)
2. Relative displacement between elements
3. Excitation amplitude of individual elements
4. Excitation phase of individual elements
5. Relative pattern of individual elements

4
The link ed image cannot be display ed. The file may hav e been mov ed, renamed, or deleted. Verify that the link points to the correct file and location.

Examples
Very Large Antenna (VLA)

Airborne Warning and

Control System (AWACS)

5
 Two-element Array
• Consider two-element array of horizontal
infinitesimal dipoles (assume no coupling between
elements)

Two infinitesimal dipoles Far-field observation

6
 Two-element Array (2)
Recall the far-zone electric field of horizontal infinitesimal
dipole in the y-z plane
 jkr
e
E  jkI 0l
ˆ cos 
4r
Thus the total electric field becomes:

 jkr1  jkr2
 e e 
E  E1  E 2  jkl  I1
ˆ cos 1  I 2 cos  2 
 4r1 4r2 

7
 Two-Element Array (3)
Using the far-field approximation
d d
r1  r2  r r1  r  cos  r2  r  cos 
2 2
The total field becomes:
 jkr
E  ˆjkl cos 
e
4r

I1e jkd cos / 2  I 2 e  jkd cos / 2 
If I1  I 0 e j / 2 ; I 2  I 0 e  j / 2 ,  : phase difference
 jkr
e d 
E  jkI 0l cos 
ˆ 2 cos(k cos   )
4r 2 2
total field  (element factor)  (array factor(AF))
8
 Electric Field Pattern
  0, d   / 4

9
 Electric Field Pattern (2)
   / 2, d   / 4

10
 Electric Field Pattern (2)
   / 2, d   / 4

11
 Quiz
• Find the far-zone electric field of a two-
element array of infinitesimal circular
loops. Assume that the loops are parallel
to the x-y plane and the two elements are
aligned along the z axis. Also, the
magnitude of the current are I1=I0, I2=-I0.

12
 N-element Linear Array: Uniform
amplitude & spacing
If the amplitude and spacing are both uniform, the array factor
becomes AF  1  e j ( kd cos   )  e j 2 ( kd cos   )
 L  e j ( N 1)( kd cos   )
N
  e j ( n 1)
n 1

where  kd cos   

13
 N-element Linear Array: Uniform
amplitude & spacing (2)
N N
thus N 1 j  j 
e jN  1 j  e 2
e 2
AF  j e 2
 
e 1 j j
e 2
e 2

N 
N 1 sin   
e
j
2

2 

sin
2
If the reference point is the physical center of the array
N  N 
sin    sin   
AF   2 
  2 
  :small 
sin
2 2
14
 N-element Linear Array: Uniform
amplitude & spacing (3)
Normalized AF N  N 
sin    sin   
 2   2  N 
(AF)n    sinc  
  :small  2 
N sin N
2 2
N  N
sin     0       n
Nulls 2  2 n

  
1 2n  
  n  cos       
 2d  N 
n  1,2,3, K; n  N ,2 N ,3 N ,K

   
Maxima   m   m   m  cos     2m 
1

2  2d 
m  0,1,2,K
15
 Normalized Array Factor

16
 Sinc function plot
|sin(x)/x|
1

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0
0 5 10 15
17
 N-element Linear Array: Uniform
amplitude & spacing (4)
N 
sin   
3-dB point  2   1  N
  h  1.391
 2 2
sin
2
   2.782 
  h  cos 1      
 2d  N 
    2.782 
  sin 1      
2  2d  N 
HPBW (symmetrical case) h  2m  h
N  N 2s  1
Secondary sin      s  1     s   
Maxima 2  2 2
   2 s  1 
  s  cos 1       ; s  1,2,3, K
 2d  N 
18
 N-element Linear Array: Uniform
amplitude & spacing (5)
First sidelobe N 3
   s   
2 2
   3 
  s  cos 1       
 2d  N 
First sidelobe level

  N 
 sin  2   
(AF)n     2
  0.212  13.46 dB
 N  3
 2   , s 1
s

19
 Example
A 3-element array of z
isotropic sources has the
phase and amplitude #2 I 2   I 0
relationships shown. The d  j / 2
spacing between elements I1  I 0 e
#1 y
is d=λ/2.
d
(a)Find the array factor. #3 I 3  I 0
(b)Find all the nulls.

20
 Broadside Array
  kd cos      / 2  0

N  10, d   / 4
21
 Grating Lobes
Should avoid d=nλ because

  kd cos    d  n ,   0
n 1, 2 , 3,K

 2n cos    0 ,  2n

No grating lobes

d max  
N  10, d  
22
23
24
 Ordinary End-fire Array
  kd cos     0  kd      kd
  kd cos        kd      kd

N  10
d /4
25
 Ordinary End-fire Array (2)

N  10, d   / 4
26
 Grating Lobes
•If d=λ/2, end-fire radiation exists simultaneously
in both directions.
•If d = nλ, also broad-side radiation.
•To avoid grating lobes,


d max 
2

27
28
29
 Phased (Scanning) Array
  kd cos      0  kd cos  0      kd cos  0

N  10, d   / 4
30
 Phased (Scanning) Array (2)
HPBW
   2.782  1    2.782 
 h  cos 1   kd cos     cos   kd cos   
 2d   2d 
0 0
N  N 
 2.782  1  2.782 
 cos 1  cos  0    cos  cos  0  
 Nkd   Nkd 

Since N = (L+d)/d,

 1   1   
 h  cos  cos  0  0.443   cos  cos  0  0.443 
 Ld   Ld 

Not valid for end-fire arrays, i.e., θ0=0,π.

31
 Hansen-Woodyard End-fire
Array
 2.92   
   kd     kd   for  0  0
 N   N
 2.92   
   kd     kd   for  0  
 N   N

Additional Conditions:

For  0  0; |  || kd cos    | 0  ; |  || kd cos    |   
N

For  0   ; |  || kd cos    |   ; |  || kd cos    | 0  
N
which yields N 1  
d 
N 4 N :large 4
32
 Hansen-Woodyard End-fire
Array (2)

33
 Hansen-Woodyard End-fire
Array (3)

34
35
36
 N-Element Array: Directivity
• Broadside Array
Recall that AF for broadside arrays is given by
N  N 
sin    sin   
(AF)n  
2   2  ;  kd cos 

  :small 
N sin N
2 2
The radiation intensity then becomes:
2
 N 
 sin  2 kd cos     sin( Z )  2
U ( ,  )  (AF) n 
2
     
N
; Z  kd cos 
kd cos  
 N   Z  2
 2 

Clearly, the maximum Umax=1 at θ=π/2

37
 N-Element Array: Directivity (2)
• Broadside Array (cont’d)
The “average” radiation intensity can be obtained from
2
Prad 1 1 2   sin( Z ) 
U0 
4 4
  U ( ,  )d 
4  
0 0  Z  sin dd
 
2
 N 
1 
sin( kd cos  ) 
   2
N  sin d
2 
0
kd cos  
 2 
N N
Using Z kd cos  ; dZ   kd sin d
2 2
2 2
1  Nkd / 2  sin( Z )  1 Nkd / 2  sin( Z ) 
U0  
Nkd Nkd / 2  Z  dZ  Nkd  Nkd / 2  Z  dZ
38
 N-Element Array: Directivity (3)
• Broadside Array (cont’d)
For a large array (Nkd/2 -> large),
2 2
1 Nkd / 2  sin( Z )  1   sin( Z ) 
U0 
Nkd  Nkd / 2  Z  dZ  Nkd   Z  dZ

2
 sin( Z )  
Since   Z  dZ   U0 
Nkd
The directivity is then given by
U Nkd d
D0  max   2N
U0  
d  Ld L
Using L=(N-1)d D0  2 N  21    2
 L ( N 1) d  d   L d 
39
 N-Element Array: Directivity (4)
• Ordinary end-fire Array
Recall that AF for ordinary end-fire arrays (θ=0) is given by
N  N 
sin    sin   
(AF)n  
2   2  ;  kd (cos  1)

  :small 
N sin N
2 2
The radiation intensity then becomes:
2
 N 
 2
sin kd (cos   1)  2
   sin( Z )  N
U ( ,  )  (AF) n     ; Z  kd (cos   1)
2
kd (cos   1) 
 N   Z  2
 2 

Clearly, the maximum Umax=1 at θ=0

40
 N-Element Array: Directivity (5)
• Ordinary end-fire Array (cont’d)
The “average” radiation intensity can be obtained from
2
Prad 1 1 2   sin( Z ) 
U0 
4 4
  U ( ,  )d 
4   0 0  Z  sin dd
 
2
 N 
1 
sin( kd (cos   1)) 
   2 sin d

2  N kd (cos   1) 
0

 2 
N N
Using Z  kd (cos   1); dZ   kd sin d
2 2
2 2
1  Nkd  sin( Z )  1 Nkd  sin( Z ) 
U0  
Nkd 0  Z  dZ  Nkd 
0  Z  dZ
41
 N-Element Array: Directivity (6)
• Ordinary end-fire Array (cont’d)
For a large array (Nkd -> large),
2 2
1 Nkd  sin( Z )  1   sin( Z ) 
U0 
Nkd 0  Z  dZ  Nkd
 
0  Z  dZ
 

Thus U0 
2 Nkd
The directivity is then given by
U 2 Nkd d
D0  max   4N
U0  
d  Ld L
Using L=(N-1)d D0  4 N  41    4
 L ( N 1) d  d   L d 
42
 N-Element Array: Directivity (7)
• Hansen-Woodyard end-fire Array
For a large array (Nkd -> large),
2
1     2  0.871 
U0       1.8515   0.554
Nkd  2   2   Nkd 2 Nkd
The directivity is then given by

U max 1 2 Nkd  d
D0    1.805 4 N 
U0 0.554   
Using L=(N-1)d
  Ld  L
D0  1.80541     1.8054 
L  ( N 1) d
  d    L d  
43
44
 Example
• Design an 18-element uniform linear
array with a spacing of λ/4 between
elements. Assume that the array is
aligned along the z-axis.
a) Find the array factor for the broadside array
case.
b) Find the first null and sidelobe locations of a).
c) Find the phase shift such that the maximum of
the array factor is at θ0=45°.
d) Find the first null and sidelobe locations of c).

45
 18-element AF (broadside array) 18−element array factor (Broadside scan
0

−5

−10

−15

−20
|(AF)n| [dB]

−25

−30

−35

−40

−45

−50
0 20 40 60 80 100 120 140 160 180
θ [Degree]

46
 18-element AF (scan array)
18−element array factor
0

−5

−10

−15

−20
|(AF) | [dB]

−25
n

−30

−35

−40

−45

−50
0 20 40 60 80 100 120 140 160 180
θ [Degree]
47
 Quiz
Find the array factor of the 3-element array of isotropic
sources shown below. The spacing between elements
is d=λ/4 and I1 = 1, I2 = -j2, I3= -1.

48
 N-element Array: Non-uniform
amplitude, uniform spacing
• Uniform amplitude -> High sidelobe
• Two popular distributions:
– Binomial (maximally flat)
– Tschebysheff (equiripple)
• HPBW: Uniform<Tschebysheff<Binomial
• Sidelobe level:
Binomial<Tschebysheff<Uniform

49
50
51
52
53
54
55
56
57
58
 Planar Array
• Linear Array = one-dimensional array,
i.e., can scan the beam only in one plane.
• In order to be able to scan the beam in
any direction, two-dimensional arrays are
needed. Geometries can be planar, circle,
cylindrical, spherical and so on.

59
 Planar Array (2)

60
 Array Factor
AF for each linear array along x-axis:
M
AF   I m1e j ( m 1)( kd x sin cos   x )
m 1
AF for the entire planar array:
N
M j ( m 1)( kd x sin  cos   x )  j ( n 1)( kd y sin  sin    y )
AF   I1n  I m1e e
n 1  m 1 
 S xm S yn
M  N 
sin   x  sin   y 
For uniform excitation, (AF)n   2   2 ;
i.e., |Im1I1n|=I0, x y
M sin N sin
2 2
 x  kd x sin  cos    x ;
 y  kd y sin  sin    y
61
 Planar Array Example

d x  d y   / 2;  x   y   /(2 2 );
 0   / 6, 0   / 4; M  N  5

62