Array of Point Sources

Usually the radiation pattern of a single element is relatively wide, and provides low values of relative gain. In many applications it is necessary to design antennas with very directive characteristics (very high gain) to meet the demands of long-distance communication. This can only be accomplished by increasing the electrical size of the antenna. Another way is to form an assembly of radiating elements in an electrical and geometrical configuration: ARRAY.
Antennas & Propagation
1

In most cases, the elements of an array are identical (though this is not necessary). The elements may be of any form e.g. wires, apertures etc. The total field of the array is determined by the vector addition of the fields radiated by the individual elements. This assumes that the current in each element is the same as that of the isolated elements. To provide very directive patterns, it is necessary that the fields from individual elements interfere constructively (add) in the desired direction and interfere destructively (cancel each other) in the remaining space.
Antennas & Propagation
2

EEL 338 20082008-09

EEL 338 20082008-09

In an array of identical elements there are five factors that can be used to shape the overall pattern of the antenna, viz.:
1. 2. 3. 4. 5. The geometrical configuration of the overall array (linear, circular, rectangular, spherical etc.) The relative displacement between the elements The excitation amplitude of the individual elements The excitation phase of the individual elements The relative pattern of the individual elements

TwoTwo-Element Array
Let us represent the electric fields in the far-zone of the array elements in the form

The simplest and one of the most practical arrays is formed by placing the elements along a line: Linear Array
Antennas & Propagation Antennas & Propagation

EEL 338 20082008-09

EEL 338 20082008-09

The far-field approximation of the two-element array problem:

Then, the total field is:

Assumptions: The array elements are Identical, i.e., Oriented in the same way in space (they have identical polarization), i.e., excitation is of the same amplitude, i.e., The total field of the array is equal to the product of the field created by a single element located at the origin and the Array factor, AF.

EEL 338 20082008-09

Antennas & Propagation

EEL 338 20082008-09

Antennas & Propagation

The normalized AF,

PATTERN MULTIPLICATION The field pattern of an array of nonnon-isotropic but similar point sources is the product of the pattern of the individual source and the pattern of an array of isotropic point sources having the same locations, relative amplitudes and phase as the nonnon-isotropic sources. The total field pattern of an array of non-isotropic but similar point sources is the product of the pattern of the individual source and the pattern of the array of isotropic point sources having the same locations, relative amplitudes and phase, while the total phase pattern is the sum of the phase patterns of the individual source and the array of isotropic sources.

Using the normalized field pattern of a single element, En( ,), the normalized field pattern of the array is expressed as:

Since, the array factor does not depend on the directional characteristics of the individual elements, it can be formulated by replacing the actual elements with isotropic (point) sources assuming that each point source has amplitude, phase and location of the corresponding element it is replacing.

EEL 338 20082008-09

Antennas & Propagation

EEL 338 20082008-09

Antennas & Propagation

Example 1: An array consists of two horizontal infinitesimal dipoles located at a distance d = / 4 from each other. Find the nulls of the total field, if the excitation magnitudes are the same and the phase difference is: a) = 0; b) =/2; c) = /2

The element factor En(,) does not depend on , and it produces the same null in all three cases. Since En(,) =|cos|, the null is at 1 = / 2. The AF depends on and produces different results in the 3 cases:

a) = 0

A solution with a real-valued angle does not exist. In this case, the total field pattern has only 1 null at =90.

EEL 338 20082008-09

Antennas & Propagation

EEL 338 20082008-09

Antennas & Propagation

10

Array factor for various values of d (=0):

90 1.0 0.8 0.6 0.4 0.2 0.0 180 0.2 0.4 0.6 0.8 1.0 240 270 300 210 330 0 150 30 120 60

d=/2 d=/4 d=/8

EEL 338 20082008-09

Antennas & Propagation

11

EEL 338 20082008-09

Antennas & Propagation

12

b) = /2

The equation

does not have a solution. The total field pattern has 2 nulls: 1 = 90 and 2 = 0

EEL 338 20082008-09

Antennas & Propagation

13

EEL 338 20082008-09

Antennas & Propagation

14

c) = /2 Array factor for various values of d (=/2):

d=/2 d=/4 d=/8 30

90 1.0 0.8 0.6 0.4 0.2 0.0 180 0.2 0.4 0.6 0.8 1.0 240 270 300 210 330 0 150 120 60

The total field pattern has 2 nulls: 1 = 90 and at 2 =180.

EEL 338 20082008-09

Antennas & Propagation

15

EEL 338 20082008-09

Antennas & Propagation

16

Example 2: Consider a 2-element array of identical (infinitesimal) dipoles oriented along the y-axis. Find the angles of observation, where the nulls of the pattern occur, as a function of the distance between the dipoles, d, and the phase difference, .

The array factor leads to the following solution:

In order to find the nulls:

The element factor, produces one null at 1 =/2

EEL 338 20082008-09

Antennas & Propagation

17

EEL 338 20082008-09

Antennas & Propagation

18

N-Element Linear Array: Uniform Amplitude and Spacing

N rN

The Array Factor (AF) is given by:

+ j kd cos + ) + j 2 kd cos + ) + j N 1)( kd cos + ) AF = 1 + e ( +e ( + KK + e ( j n 1 kd cos + ) j n 1 AF = e ( )( = e ( ) n =1 n =1 N N

All elements have identical amplitudes but each succeeding element has a progressive phase-lead current excitation equal to relative to the preceding one.
r3

= kd cos +

..(1)

3 d 2 d 1

r2

An array of identical elements all of identical magnitude and each with a progressive phase is referred to as a Uniform Array

Phase terms of the partial fields:

r1

Since the total array factor for a uniform array is a sum of exponentials, it can be represented by the vector sum of N phasors each of unit amplitude and progressive phase = relative to the previous one.

EEL 338 20082008-09

Antennas & Propagation

19

EEL 338 20082008-09

Antennas & Propagation

20

The array factor can be expressed in more recognizable form:

1 ( N 1)
13 # 4 3
AF

It is apparent from the phasor diagram that the amplitude and phase of the AF can be controlled in uniform arrays by properly selecting the relative phase between the elements. In non-uniform arrays, the amplitude as well as the phase can be used to control the formation and distribution of the total array factor

j n 1 AF = e ( ) n =1

..(1)

= kd cos +
Multiply both sides of (1) by e j , subtract the original equation from the resulting equation and rearranging,

12 #3 2
o

10 #1

1 # 2

e jN 1 j ( N 1) AF = =e j e 1 AF
j ( N 1) 2 =e

e j[ N 2] e j[ N 2] j[1 2] j 1 2 e [ ] e
..(2)

N sin 2 1 sin 2
Antennas & Propagation

EEL 338 20082008-09

Antennas & Propagation

21

EEL 338 20082008-09

22

Here, N shows the location of the last element with respect to the reference point in steps with length d. The phase factor exp[ j(N 1) / 2] represents the phase shift of the arrays phase centre relative to the origin, and it would be one if the origin coincides with the array centre. Neglecting the phase factor gives (taking physical center of the array as phase reference:

To normalize equation (3) or (4), we need the maximum of the AF. Re-write equation (3) as:

f ( x) =

N sin ( x )

sin ( Nx )

N sin 2 AF = 1 sin 2
For small values of

..(3)

N sin 2 ( AF ) = N 1 N sin 2

The function f(x) has its maximum at x = 0, ,, and the value of this maximum is fmax =1.

, :

N sin 2 AF = 2
Antennas & Propagation

( AF )max = N
..(4)

EEL 338 20082008-09

23

EEL 338 20082008-09

Antennas & Propagation

24

The maximum value of equation 3 or 4 is equal to N. To normalize the array factors so that the maximum value is equal to unity, equations 3 and 4 are written in normalized form as:

Nulls of the Array: Array: Equations 5 and 6 are set equal to zero. That is,

( AF )n
and

N sin 1 2 = N 1 sin 2 N sin 1 2 = N 2

..(5)

N N sin = 0 = n 2 2 2n n = cos 1 n = 1, 2,3,.... N 2 d

..(7)

with equation 5, n N , 2 N ,3 N ,....because for these values of n, equation 5 attains its maximum value as it reduces to sin ( 0 ) 0 form.
..(6)

( AF )n

The values of n determine the order of the nulls (first, second, etc.). For a zero to exist, the argument of arccosine must be between 1 and +1. Thus the number of nulls that can exist will be a functions of element separation and phase excitation difference.
25

EEL 338 20082008-09

Antennas & Propagation

EEL 338 20082008-09

Antennas & Propagation

26

The maximum of equation 5 occurs when,

The 3-dB point for the array factor of equation 6: 6:

1 ( kd cos + ) | =m = m 2 2 m = cos 1 ( 2m ) 2 d m = 0,1, 2,K =

N N = ( kd cos + ) | = h = 1.391 2 2 2.782 h = cos 1 N 2 d

..(8)

which can also be written as

Equation 6 has only one maximum and occurs when, That is, the observation angle that makes = 0

m=0

2.782 sin 1 2 N 2 d For large values of d ( d >> ) it reduces to:

h =

m = cos 2 d
1

..(9)

2.782 2 2 d N
..(11)

..(10)

The half power beamwidth for a symmetrical pattern is:

h = 2 m h
EEL 338 20082008-09 Antennas & Propagation
27

EEL 338 20082008-09

Antennas & Propagation

28

Broadside Array
Maximum radiation of an array directed normal to the axis of the array. ( =90 in the present case) Maximum of the array factor occurs when (equations 5 and 6):

Example: N=10

d=/4

d=

= kd cos + = 0
For broadside array,

= kd cos + | =90 = 0
o

=0
For broadside pattern, all elements should have same phase and amplitude excitation. To ensure that there are no maxima in other directions, which are referred to as grating lobes, the separation between the elements should not be equal to multiples of a wavelength when = 0

EEL 338 20082008-09

Antennas & Propagation

29

EEL 338 20082008-09

Antennas & Propagation

30

Parameters for Broadside Array

Ordinary End-Fire Array

An end-fire array is an array, which has its maximum radiation along the axis of the array ( =0, 180 in the present case). To direct the maximum toward =0: Half-power points

n = cos 1
Nulls

n N d n = 1, 2,3,... n N , 2 N ,3 N ,...

1.391 h cos 1 Nd d << 1

= kd cos + | =0 = kd
o

To direct the maximum toward =180:

Maxima

m = cos 1

m d m = 0,1, 2,...

s cos 1
s = 1, 2,3... d << 1

2d N

2s + 1

= kd cos + | =180 = kd
o

Minor-lobe maxima

If the element separation is a multiple of a wavelength (d=n, n=1,2,3,), then there exists a maxima in the broadside directions.

EEL 338 20082008-09

Antennas & Propagation

31

EEL 338 20082008-09

Antennas & Propagation

32

AF pattern of an End Fire Array (EFA): N = 10, d = /4

Parameters for Ordinary EndEnd-Fire Array

n = cos 1 1
Nulls

n N d n = 1, 2,3,... n N , 2 N ,3 N ,...

h cos 1 1 d << 1

1.391 Nd
Half-power points

Maxima

m = cos 1 1

m = 0,1, 2,...

m d

s cos 1 1
s = 1, 2,3... d << 1

2s + 1 2d N
Minor-lobe maxima

EEL 338 20082008-09

Antennas & Propagation

33

EEL 338 20082008-09

Antennas & Propagation

34

Phased (Scanning) Array

The 0th order maximum (m=0) of AFn occurs when

Example: Values of the progressive phase shift as dependent on the direction of the main beam 0 for a uniform linear array with d = /4. 0 0 60 120 180

= kd cos + = 0
This gives the relation between the direction of the main beam 0 and the phase difference . The direction of the main beam can be controlled by the phase shift . This is the basic principle of electronic scanning for phased arrays. The scanning must be continuous. That is why the feeding system should be capable of continuously varying the progressive phase between the elements. This is accomplished by ferrite or diode phase shifters (varactors).

= kd cos0 = cos 0
2

-90 -45 +45

+90

The HPBW of a scanning array is obtained with = kdcos0:

h1,2 = cos 1
The total beamwidth is

2.782 N 2 d

HPBW = h1 h 2 2.782 2.782 1 HPBW = cos 1 kd cos 0 cos kd cos 0 + N N 2 d 2 d

EEL 338 20082008-09

Antennas & Propagation

35

EEL 338 20082008-09

Antennas & Propagation

36

k = 2

2.782 2.782 cos 1 cos 0 + HPBW = cos 1 cos 0 Nkd Nkd

If L is length of the array:

N=

L+d d

1 HPBW = cos 1 cos 0 0.443 cos cos 0 + 0.443 L + d L + d

These equations can be used to calculate the HPBW of a broadside array, too (0 =90=const ). However, they are not valid for endfire arrays.

EEL 338 20082008-09

Antennas & Propagation

37

EEL 338 20082008-09

Antennas & Propagation

38

Hansen-Woodyard End-Fire Array

End-fire arrays (EFA) have relatively broad HPBW as compared to broadside arrays Directivity is low. To enhance the directivity of an end-fire array, Hansen and Woodyard proposed that the phase shift of an ordinary EFA be increased:

Hansen-Woodyard conditions, ensure minimum beamwidth (maximum directivity) in the end-fire direction. There is, however, a trade-off in the side-lobe level, which is higher than that of the ordinary EFA. These conditions have to be complemented by additional requirements, to ensure low level of the radiation in the direction opposite to the main lobe:

= kd = + kd

= kd + N = + kd + N
2.94

2.94

For maximum in =0 For maximum in =180

Hansen-Woodyard conditions for endfire radiation (do not necessarily yield the maximum possible directivity)
EEL 338 20082008-09 Antennas & Propagation EEL 338 20082008-09 Antennas & Propagation

39

40

Maximum at = 0

N-Element Linear Array: Directivity

1. Broadside Array:

=0

In order to ensure a maximum in the =180 direction, we must have:

( AF )n
for

=0
o

N kd cos sin 1 2 = 1 N sin kd cos 2 N sin 2 kd cos = N kd cos 2

=180
o

d <<

Hence, Hansen-Woodyard conditions for maximum directivity in the = 0 direction are:

=0
o o

( AF )n

=180
EEL 338 20082008-09 Antennas & Propagation EEL 338 20082008-09

41

Antennas & Propagation

42

The radiation intensity can be written as:

The Average Radiation Intensity can be written as:

N 2 sin kd cos 2 sin ( Z ) 2 ( AF ) = U ( ) = = n N kd cos Z 2 Z=

U0 =

1 1 Prad = 4 4

U ( ) sin d d
0 0

The directivity: directivity:

N kd cos 2

N sin kd cos 1 2 sin d = 2 N 0 kd cos 2

changing the variable

D0 =

4 U max U max = Prad U0

Since the array factor is normalized, the numerator is unity and occurs at = 90o .

N kd cos 2 N dZ = kd sin d 2 Z=
EEL 338 20082008-09 Antennas & Propagation
44

EEL 338 20082008-09

Antennas & Propagation

43

The directivity is then,

U0 =

1 Nkd

Nkd 2

Nkd 2

sin Z dZ Z

D0 =
using

for a large array Nkd 2 Large the above equation can be approximated by extending the limits to infinity.

U max Nkd d = 2N U0
overall length of the array

L = ( N 1) d

U0
since

1 Nkd

+ sin Z

dZ Z

+ sin Z

dZ = Z

d L d D0 2 N 2 1 + d
For a large array,

L >> d L

U0

Nkd

D0 2

EEL 338 20082008-09

Antennas & Propagation

45

EEL 338 20082008-09

Antennas & Propagation

46

2. Ordinary End-Fire Array: For a large array,

N-Element Linear Array: ThreeThree-Dimensional Characteristics

Generally, it is assumed that the linear-array elements are located along the z-axis, which is symmetrical around the z-axis. If the array axis has an arbitrary orientation, the array factor can be expressed as

L >> d L

D0 4

3. Hansen-Woodyard End-Fire Array: For a large array,

L >> d

j n 1 kd cos + ) j n1 AF = an e ( )( = an e ( ) n =1 n=1

L D0 1.789 4

= kd cos +
is the angle subtended between the array axis and the position vector to the observation point.

EEL 338 20082008-09

Antennas & Propagation

47

EEL 338 20082008-09

Antennas & Propagation

48

Thus, if the array axis is along the unit vector

= sin a cos a x + sin a sin a y + cos a z a

and the position vector to the observation point is

N-Element Linear Array: Uniform Spacing, NonNon-uniform Amplitude

The most often used Broad-Side Arrays (BSAs), are classified according to the type of their excitation amplitude: 1. The uniform BSA relatively high directivity, but the sidelobe levels are high; 2. DolphTschebyscheff BSA for a given number of elements maximum directivity is next after that of the uniform BSA; side-lobe levels are the lowest in comparison with the other two types of arrays for a given directivity; 3. Binomial BSA does not have good directivity but has very low side-lobe levels (when d = /2, there are no side lobes at all).

= sin cos x + sin sin y + cos z r r = sin a cos a sin cos x + cos = a + cos a cos z sin a sin a sin sin y cos = sin a sin cos ( a ) + cos a cos
To Do: Express the array factor for elements along x-axis and y-axis.

EEL 338 20082008-09

Antennas & Propagation

49

EEL 338 20082008-09

Antennas & Propagation

50

Even number (2M) of elements, located symmetrically along the z-axis, with excitation symmetrical with respect to z = 0. For a broadside array ( =0),

( AF )

1 + j kd cos = a1e 2

+ a2

3 + j kd cos e 2

+ ..... + aM e

+j

( 2 M 1) kd cos
2 j

1 j kd cos a1e 2 M

+ a2

3 j kd cos 2 e

+ ..... + aM e

( 2 M 1) kd cos
2

( AF )e = 2 an cos
= a cos ( AF )e n n
n =1 M n =1 n =1 M

( 2n 1) 2 2

kd cos

( 2n 1)

kd cos 1 d u = kd cos = cos 2

52

= a cos ( AF )e ( 2n 1) u n n
EEL 338 20082008-09 Antennas & Propagation
51

EEL 338 20082008-09

Antennas & Propagation

Odd number (2M+1) of elements, located symmetrically along the z-axis, For a broadside array ( =0),

Binomial Array

(1 + x )m1 = 1 + ( m 1) x +
+ a3e+ j 2 kd cos + ..... + aM +1e+ jMkd cos +

( m 1)( m 2 )
2!

x2 +

( m 1)( m 2 )( m 3)
3!

x3 + ...

( AF )o = 2a1 + a2e+ jkd cos

The positive coefficients of the series expansion:

m=1 m=2 m=3 m=4 m=5 m=6

a2 e jkd cos + a3e j 2 kd cos + ..... + aM +1e jMkd cos

1 1 1 1 1 1 5 4 10 3 6 10 2 3 4 5 1 1 1 1 1

( AF )

= 2 an cos ( n 1) kd cos
n =1 M +1 n =1 M +1 n =1

M +1

( AF )n =
o

an cos ( n 1) kd cos 1 d u = kd cos = cos 2

53

= an cos ( AF )o 2 ( n 1) u n
EEL 338 20082008-09

Antennas & Propagation

EEL 338 20082008-09

Antennas & Propagation

54

An approximate closed-form expression for the HPBW with d = /2:

HPBW =

1.06 N 1

1.06 2L

1.75 L

The directivity with spacing d = /2 is

D0 = 1.77 N = 1.77

( 2L ) + 1

EEL 338 20082008-09

Antennas & Propagation

55

EEL 338 20082008-09

Antennas & Propagation

56

Array Factor for symmetric amplitude excitation:

Dolph-Tschebyscheff Array

( AF )2 M ( even ) = an cos ( 2n 1) u ( AF )2 M +1 ( odd ) = an cos 2 ( n 1) u

n =1 n =1 M +1

A compromise between uniform and binomial arrays. Excitations coefficients are related to Tschebyscheff polynomials. A DolphDolph-Tschebyscheff array with no sidelobes (or side lobes of dB) reduces to the binomial design. The excitation coefficients for this case, as obtained by both methods would be identical.

u=

d cos

Summation of M or (M+1) cosine terms. Largest harmonic of the cosine terms is one less than the total no. of elements of the array. Each cosine term, whose argument is an integer times a fundamental frequency, can be rewritten as a series of cosine functions with the fundamental frequency as the argument.

EEL 338 20082008-09

Antennas & Propagation

57

EEL 338 20082008-09

Antennas & Propagation

58

By de Moivres theorem:

jmu

= cos mu + j sin mu = ( cos u + j sin u )

m

m=0
m

cos ( mu ) = 1 cos ( mu ) = cos ( u ) cos ( mu ) = cos ( 2u ) = 2 cos 2 u 1 cos ( mu ) = cos ( 3u ) = 4 cos3 u 3cos u cos ( mu ) = cos ( 4u ) = 8cos 4 u 8cos 2 u + 1 cos ( mu ) = cos ( 5u ) = 16 cos5 u 20 cos3 u + 5cos u cos ( mu ) = cos ( 6u ) = 32 cos 6 u 48cos 4 u + 18cos 2 u 1 cos ( mu ) = cos ( 7u ) = 64 cos 7 u 112 cos5 u + 56 cos3 u 7 cos u cos ( mu ) = cos ( 8u ) = 128cos8 u 256 cos6 u + 160 cos 4 u 32 cos 2 u + 1 cos ( mu ) = cos ( 9u ) = 256 cos9 u 576 cos 7 u + 432 cos5 u 120 cos3 u + 9 cos u

m =1 m=2 m=3 m=4 m=5

On taking real parts:

cos mu = Re ( cos u + j sin u )

Expanding as binomial series:

cos ( mu ) = cos m ( u )

m ( m 1)

+
Putting sin m, we have
2

2! m ( m 1)( m 2 )( m 3) 4!

cos m 2 ( u ) sin 2 ( u ) cos m 4 ( u ) sin 4 ( u ) ...

m=6 m=7 m=8 m=9

( u ) = 1 cos 2 ( u ) and substituting particular values of

Antennas & Propagation
59

EEL 338 20082008-09

EEL 338 20082008-09

Antennas & Propagation

60

If we let

z = cos u

above equations can be rewritten as:

m=0 m =1 m=2 m=3 m=4 m=5 m=6 m=7 m=8 m=9

cos ( mu ) = 1 = T0 ( z ) cos ( mu ) = z = T1 ( z ) cos ( mu ) = 2 z 2 1 = T2 ( z ) cos ( mu ) = 4 z 3 z = T3 ( z )

3

These relations between cosine functions and Tschebyscheff polynomials are valid only in the range: 1 z +1

Q cos ( mu ) 1 Tm ( z ) 1

for

1 z +1

cos ( mu ) = 8 z 4 8 z 2 + 1 = T4 ( z ) cos ( mu ) = 16 z 5 20 z 3 + 5 z = T5 ( z ) cos ( mu ) = 32 z 6 48 z 4 + 18 z 2 1 = T6 ( z ) cos ( mu ) = 64 z 7 112 z 5 + 56 z 3 7 z = T7 ( z ) cos ( mu ) = 128 z 8 256 z 6 + 160 z 4 32 z 2 + 1 = T8 ( z ) cos ( mu ) = 256 z 576 z + 432 z 120 z + 9 z = T9 ( z )
9 7 5 3

The recursion formula for Tschebyscheff polynomials is:

Tm ( z ) = 2 zTm 1 ( z ) Tm 2 ( z )

The polynomials can also be computed using:

1 Tm ( z ) = cos m cos ( z )

1 z +1 z < 1, z > +1

And each is related to a Tschebyscheff (Chebyshev) polynomial Tm

(z)
61

1 Tm ( z ) = cosh m cosh ( z )

EEL 338 20082008-09

Antennas & Propagation

EEL 338 20082008-09

Antennas & Propagation

62

The first seven Tschebyscheff polynomials have been plotted below:

Tm(z) 10 8 6 4 2

Properties Since, the array factor of an even or odd number of elements (summation of cosine terms) has the same form as the Tschebyscheff polynomials: The unknown coefficients of the array factor can be determined by equating the cosine series of the array factor to the appropriate Tschebyscheff polynomial. The order of the polynomial should be one less than the total number of elements of the array.

All polynomials pass through (1,1) For

1 z +1 1 Tm ( z ) +1 1 z +1

-1.5

-1.0

-0.5

0 0.0 -2 -4 -6 -8 -10

0.5

1.0

1.5

All roots occur within

All maxima and minima within this range have values 1

T2 T6 T3 T7

T0 T4

T1 T5

EEL 338 20082008-09

Antennas & Propagation

63

EEL 338 20082008-09

Antennas & Propagation

64

D-T Array Design Procedure

Statement: Design a broadside D-T array of 2 M or 2 M + 1 elements with spacing d between the elements. The side lobes are R0 dB below the maximum of the major lobe. Find the excitation coefficients and form the array factor. Procedure: Select the appropriate array factor (for even or odd no. of elements). Expand the array factor. Replace each cos appropriate series expansion.

The roots of the polynomial correspond to the nulls of the field pattern. The important property of the Tschebyscheff polynomial is that if the ratio R0 is specified, the beamwidth to the first null is minimized. The corollary also holds that if the beamwidth is specified, the ratio R0 is maximized (side-lobe level minimized). Since

( mu ) function by its

R0 > 1 , so z0 > 1 . But, z0 1 z0 +1 w= z z0 z z0

must be restricted to the range

Determine the point z = z0 such that Tm z0 = R0 (voltage ratio). z0 , R0 The point corresponds to the main-lobe maximum, while the minor lobes are confined to a maximum value of unity. The design procedure requires that the Tschebyscheff polynomial in the 1 z z1 where z1 is the null nearest to z = +1 be used to represent the minor lobes of the array. The major lobe of the pattern are formed from remaining part of the polynomial up to point z0 z1 < z z0 .

( )

So, we need to change the scale, by introducing a new abscissa:

Now substitute:

cos u = w =

EEL 338 20082008-09

Antennas & Propagation

65

EEL 338 20082008-09

Antennas & Propagation

66

This substitution is valid for maximized.

z z0

. At

z = z0

, cos

(u )

Dolph-Tschebyscheff Array: Revision

is The main goal is to approximate the desired AF with a Tschebyscheff polynomial such that the side-lobe level meets the requirements, and the main beam width is small enough. An array of N elements has an AF approximated with a Tschebyscheff polynomial of order m, which is always

Equate the array factor after substitution to a Tm z . The should be chosen one order less than the total number of elements.

( )

Tm ( z )

Write the array factor using the coefficients obtained above. This is the optimum design for the specified side-lobe level.

m = N 1
where N= 2M, if N is even; and N= 2M+1, if N is odd. In general, for a given side-lobe level, the higher the order m of the polynomial, the narrower the beamwidth. However, for m > 10, the difference is not substantial.

EEL 338 20082008-09

Antennas & Propagation

67

EEL 338 20082008-09

Antennas & Propagation

68

 The AF of an N-element array is identical with a Tschebyscheff polynomial if

Then, the maximum of TN1 is fixed at an argument z0, where

max TN 1 ( z0 ) = R0

TN 1 ( z )

n =1

an cos 2 ( n 1) u , N = 2 M + 1, odd

n =1 M

an cos ( 2n 1) u , N = 2 M , even

The main beam-to-side lobe ratio R0 is the value of the array factor at the main beam maximum, since the side-lobe level magnitude is unity. This corresponds to

AF ( u ) = AF max ( u0 )
Obviously, z0 must satisfy the condition

1 d u = kd cos = cos 2
Let the side-lobe level (voltage ratio) be

z0 > 1
Then, the portion of AF(u), which corresponds to TN1(z) for |z|<1, will have levels lower or equal to the specified side-lobe level R0. This corresponds to the out-of-main-beam radiation pattern, i.e., the side lobes.
69

R0 =

Emax 1 1 = = Esl Esl Emax AFsl

Antennas & Propagation

EEL 338 20082008-09

EEL 338 20082008-09

Antennas & Propagation

70

The AF is a polynomial of cosu , and the TN1(z) is a polynomial of z where the limits for z are

Example of DD-T distribution for an array for Ten sources

1 z z0
Since

1 cos u 1

An array of 2M=10 in-phase isotropic sources, with spacing between the elements d is to have a side-lobe level of 26 dB below the main lobe maximum. Find: (a) the amplitude distribution fulfilling this requirement that produces the minimum beamwidth between the first nulls. Solution: Solution: The array factor is given by (even case):

the relation between z and cos(u) must be normalized as

cos u =

z z0

z0 > 1

( AF )2 M = an cos ( 2n 1) u ( AF )10 = a1 cos ( u ) + a2 cos ( 3u ) + a3 cos ( 5u ) + a4 cos ( 7u ) + a5 cos ( 9u )

u=
n =1

M =5

d cos
Antennas & Propagation
72

EEL 338 20082008-09

Antennas & Propagation

71

EEL 338 20082008-09

Replace cosine terms with their expansions:

Using
1 Tm ( z ) = cos m cos ( z ) 1 Tm ( z ) = cosh m cosh ( z )

3 5 3 ( AF )10 = a1 cos ( u ) + a2 4 cos ( u ) 3cos ( u ) + a3 16 cos ( u ) 20 cos ( u ) + 5cos ( u ) 7 5 3 + a4 64 cos ( u ) 112 cos ( u ) + 56 cos ( u ) 7 cos ( u ) 9 7 5 3 256 cos u 576 cos u 432 cos u 120 cos + a5 + ( ) ( ) ( ) ( u ) + 9 cos ( u )

1 z +1 z < 1, z > +1

After rearranging we get,

( AF )10 = cos ( u ) [ a1 3a2 + 5a3 7a4 + 9a5 ] + cos ( u ) [ 4a2 20a3 + 56a4 120a5 ] + cos5 ( u ) [16a3 112a4 + 432a5 ] + cos 7 ( u ) [ 64a4 576a5 ] + cos9 ( u ) [ 265a5 ]
3

R0 = 20 = T9 ( z0 ) = cosh 9cosh 1 ( z0 ) 1 z0 = cosh cosh 1 ( 20 ) = 1.0851 9

Determine Zo by equating R0 to T9(z):

Substitute:

R 0 (dB) = 26 = 20log10 ( R0 ) R0 ( voltage ratio ) = 20

EEL 338 20082008-09 Antennas & Propagation

cos ( u ) =

z z = z0 1.0851

73

EEL 338 20082008-09

Antennas & Propagation

74

Equate this array factor to T9(z):

( AF )10 =

z [ a1 3a2 + 5a3 7a4 + 9a5 ] + 1.0851

3 5

( AF )10 =

z [ a1 3a2 + 5a3 7a4 + 9a5 ] 1.0851

3

z z [ 4a2 20a3 + 56a4 120a5 ] + [16a3 112a4 + 432a5 ] + 1.0851 1.0851 z z [ 64a4 576a5 ] + [ 265a5 ] 1.0851 1.0851
7 9

z + [ 4a2 20a3 + 56a4 120a5 ] 1.0851 z + [16a3 112a4 + 432a5 ] 1.0851 z + [ 64a4 576a5 ] 1.0851 z + [ 265a5 ] 1.0851 = 256 z 9 576 z 7 + 432 z 5 120 z 3 + 9 z = T9 ( z )
9 7 5

EEL 338 20082008-09

Antennas & Propagation

75

EEL 338 20082008-09

Antennas & Propagation

76

Match similar terms to get the coefficients ans:

The array factor can be written as:

a5 = 2.0860 a3 = 4.1184

a4 = 2.8308 a2 = 5.2073 a1 = 5.8377

Normalization 1: With respect to amplitude of the elements at the edge Normalization 2: With respect to amplitude of the element at the centre

( AF )10 = 2.798cos ( u ) + 2.496 cos ( 3u ) + 1.974cos ( 5u ) + 1.357 cos ( 7u ) + cos ( 9u )

u=

d cos

a5 = 1 a4 = 1.357 a3 = 1.974 a2 = 2.496 a1 = 2.798

EEL 338 20082008-09

a5 = 0.357 a4 = 0.485 a3 = 0.706 a2 = 0.890 a1 = 1

Antennas & Propagation
77

EEL 338 20082008-09

Antennas & Propagation

78

d d z = z0 cos u = z0 cos cos = 1.0851cos cos

EEL 338 20082008-09

Antennas & Propagation

79

EEL 338 20082008-09

Antennas & Propagation

80

Examples of AFs of arrays of non-uniform amplitude distribution

(N=5, d=/2, 0 = 90) Uniform amplitude distribution

Triangular (1:2:3:2:1) amplitude distribution

EEL 338 20082008-09

Antennas & Propagation

81

EEL 338 20082008-09

Antennas & Propagation

82

Binomial (1:4:6:4:1) amplitude distribution

Dolph-Tschebyschev (1:1.61:1.94:1.61:1) amplitude distribution

EEL 338 20082008-09

Antennas & Propagation

83

EEL 338 20082008-09

Antennas & Propagation

84

Dolph-Tschebyschev (1:2.41:3.14:2.41:1) amplitude distribution

Beamwidth of D-T Array

For large D-T arrays, scanned near-broadside and side lobes in the -20 to -60 dB range: Calculate beamwidth of a uniform array with same no. of elements and element spacing:

1 HPBW = cos 1 cos 0 0.443 cos cos 0 + 0.443 L + d L + d

multiply it by beam-broadening factor f

2 f = 1 + 0.636 cosh R0
EEL 338 20082008-09 Antennas & Propagation EEL 338 20082008-09

( cosh

R0

85

Antennas & Propagation

86

Directivity of D-T Array

D0 =

1 + R0 2 1 f

2 R0 2

(L + d )

Example: For a D-T array of 10 elements with spacing d=/2 and side lobe level of 26 dB, calculate HPBW and the directivity. R0 = 26 dB R0 = 20 (voltage ratio) The beam broadening factor f=1.079 Beamwidth for uniform BSA with L+d=5 = 10.17 Beamwidth for DT array = 10.17f = 10.97

The directivity of D-T array, with a given side-lobe level, increases as the array size or no. of elements increases. For a given array length (or no. of elements in the array), the directivity does not necessarily increase as side lobe level decreases.

Directivity = 9.18 (dimensionless) = 9.63 dB

EEL 338 20082008-09

Antennas & Propagation

87

EEL 338 20082008-09

Antennas & Propagation

88

PLANAR ARRAYS

Planar arrays provide directional beams, symmetrical patterns with low side lobes, much higher directivity (narrow main beam) than that of their individual element. In principle, they can point the main beam toward any direction. Applications tracking communications, etc. radars, remote sensing,

EEL 338 20082008-09

Antennas & Propagation

89

EEL 338 20082008-09

Antennas & Propagation

90

The AF of a linear array of M elements along the x-axis is

AF =

m =1

I m1e (

j m 1)( kd x sin cos + x )

If N such arrays are placed next to each other in the y direction, a rectangular array is formed. We assume again that they are equispaced at a distance dy and there is a progressive phase shift along each row of y. We also assume that the normalized current distribution along each of the x-directed arrays is the same but the absolute values correspond to a factor of I1n (n=1,...,N).

sin cos = cos x

directional cosine with respect to x-axis.

All elements are equispaced with an interval of dx and a progressive shift x. Im denotes the excitation amplitude of the element at the point with coordinates: x=(m-1)dx, y=0. This is the element of the m-th row and the 1st column of the array matrix.

EEL 338 20082008-09

Antennas & Propagation

91

EEL 338 20082008-09

Antennas & Propagation

92

Then, the AF of the entire MN array is

N M j m 1 kd sin cos + x ) j ( n 1)( kd y sin sin + y ) AF = I1n I m1e ( )( x e n =1 m=1

The pattern of a rectangular array is the product of the array factors of the linear arrays in the x and y directions. In the case of a uniform planar (rectangular) array, all elements have the same excitation amplitudes: I m1 = I1n
M N

= I0

AF = S xM S yN

j ( n 1)( kd y sin sin + y ) j m 1 kd sin cos + x ) AF = I 0 e ( )( x e m =1 n =1

S xM = AFx1 =

S yMN = AF1 y = I1n e

n =1

m =1 N

I m1e (

j m 1)( kd x sin cos + x )

The normalized array factor is obtained as:

j ( n 1) kd y sin sin + y

r = cos x sin cos = x r = cos y sin sin = y

EEL 338 20082008-09 Antennas & Propagation
93

M N 1 sin 2 x 1 sin 2 y AFn ( , ) = 1 1 M N sin x sin y 2 2 x = kd x sin cos + x

y = kd y sin sin + y
EEL 338 20082008-09 Antennas & Propagation
94

S xM

M 1 sin 2 x = 1 M sin x 2

S yN

N 1 sin 2 y = 1 N sin y 2

If it is required that the main beams of SxM and SyN intersect (which is usually the case), then the common main beam is in the direction:

The major lobe (principal maximum) and grating lobes of are located at angles such that

= 0

= 0

m=n=0

kd x sin m cos m + x = 2m kd y sin n sin n + y = 2n

m = 0,1,... n = 0,1,...

If the principal maximum is specified by (0,0) , then the progressive phases x and y must satisfy

The principal maxima correspond to m=0, n=0. In general, x and y are independent from each other.

x = kd x sin 0 cos 0 y = kd y sin 0 sin 0

EEL 338 20082008-09

Antennas & Propagation

95

EEL 338 20082008-09

Antennas & Propagation

96

 When x and y are specified, the main beam direction can be found by:

3-D PATTERN OF A 5-ELEMENT SQUARE PLANAR UNIFORM ARRAY WITHOUT GRATING LOBES (d=/4, x= y=0 )

ydx tan 0 = xd y

2 x y sin 0 = + kd x kd y

The grating lobes can be located by

tan mn =

sin 0 sin 0 n

sin 0 cos 0 m sin 0 cos 0 m cos mn sin 0 sin 0 n sin mn

dy dx dx

In order a true grating lobe to occur, these equations must have a real solution (mn,mn). To avoid grating lobes, the spacing between the elements must be less than (dx< and dy<).

sin mn =

dy

EEL 338 20082008-09

Antennas & Propagation

97

EEL 338 20082008-09

Antennas & Propagation

98

3-D PATTERN OF A 5-ELEMENT SQUARE PLANAR UNIFORM ARRAY WITHOUT GRATING LOBES (d=/2, x= y=0 ):

The beamwidth of a planar array

For a large array, whose maximum is near the broad side, the elevation plane HPBW is approximately:

h =

1 cos
2

2 cos 2 0 0 x0

2 2 + y 0 sin 0

The HPBW in the plane, which is orthogonal to the =0 plane and contains the maximum, is

h =

1
2 sin 2 0 x0 2 2 + y 0 cos 0

(0,0) specifies the main-beam direction; x0 is the HPBW of a linear broadside array of M elements; y0 is the HPBW of a linear broadside array of N elements;
EEL 338 20082008-09 Antennas & Propagation
99

EEL 338 20082008-09

Antennas & Propagation

100

For a square array (M=N, x0=y0) with amplitude distributions of the same type along the x and y axes,

Directivity of Planar Array

The general expression for the calculation of the directivity of an array is:

h = x 0 sec 0 = y 0 sec 0 h = x0 = y 0
The beam solid angle of the planar array can be approximated by

D0 = 4

AF (0 , 0 )
2

0 0

AF (0 , 0 ) sin d d

A = h h A = x 0 y 0 sec 0 y 02 2 cos sin 2 0 + 0 2 x 0

12

For large planar arrays, which are nearly broadside:

D0 = cos 0 Dx D y 2 sin 2 0 + x 0 2 cos 2 0 y0

12
Dx is the directivity of the respective linear BSA, x-axis; Dy is the directivity of the respective linear BSA, y-axis.

Approximate directivity of a nearly broadside planar array:

D0
EEL 338 20082008-09

A rads

2
2

32, 400 A deg 2

)
102

EEL 338 20082008-09

Antennas & Propagation

101

Antennas & Propagation

ARRAYS
9 The main beam direction is controlled through the phase shifts, x and y. 9 The beamwidth and side-lobe levels are controlled through the amplitude distribution.

Example: Compute the HPBWs, beam solid angle and directivity of a planar square array of 100 isotropic elements (10 10). Assume a D-T distribution, /2 spacing between elements, -26 dB side lobe level, and the maximum oriented along 0 = 30, 0 = 45.

EEL 338 20082008-09

Antennas & Propagation

103

EEL 338 20082008-09

Antennas & Propagation

104

Solution 6.10: Problem 6.10: Design an ordinary end-fire uniform linear array with only one maximum so that its directivity is 20 dBi (above isotropic). The spacing between the elements is d = /4, and its length is much greater than the spacing. Determine the: 1. number of elements, 2. overall length of the array (in wavelengths), 3. approximate half-power beamwidth (in degrees), 4. amplitude level (compared to the maximum of the major lobe) of the first minor lobe (in dB) 5. progressive phase between the elements (in degrees).

d D0 = 4 N D0 = 20dB = 100 L = ( N 1) d
1.391 N d

N = 100 L = 24.75

h = cos 1 1

h = 10.8o

HPBW = 2 h = 21.6o

EEL 338 20082008-09

Antennas & Propagation

105

EEL 338 20082008-09

Antennas & Propagation

106

Solution 6.15: Problem 6.15: Show that in order for a uniform array of N elements not to have any minor lobes, the spacing and the progressive phase shift between the elements must be: For =0: 1. d=/N, =0 for a broadside array, 2. d=/(2N), =kd for an ordinary end-fire array.

N sin 1 2 ( AF )n = N 1 sin 2

= kd cos +

N sin kd cos 1 2 ( AF )n = N 1 sin kd cos 2

In order for the array not to have any minor lobes, we can assume that its first null occurs at = 0 or 180. Thus,

N kd sin 1 2 = 0 ( AF )n = N 1 sin kd 2
EEL 338 20082008-09 Antennas & Propagation
107

N kd = 2

d=

EEL 338 20082008-09

Antennas & Propagation

108

For =kd, maxima occurs at =180:

N kd ( cos + 1) sin 1 2 ( AF )n = N 1 sin kd ( cos + 1) 2

In order for the array not to have any minor lobes, we can assume that its first null occurs at = 0. Thus,

( AF )n =

1 sin ( Nkd ) =0 N sin ( kd )

Nkd =

d=

2N

EEL 338 20082008-09

Antennas & Propagation

109