Antenas N Wave Propagation

Naval Postgraduate School Distance Learning
Antennas & Propagation

LECTURE NOTES
VOLUME II
BASIC ANTENNA PARAMETERS AND WIRE
ANTENNAS
by Professor David Jenn
PROPAGATION
DIRECTION
H
r
E
r

k
ANTENNA
(ver 1.3)
1
Naval Postgraduate School Antennas & Propagation Distance Learning
Antennas: Introductory Comments
Classification of antennas by size:
Let

l be the antenna dimension:
1. electrically small,

l << : primarily used at low frequencies where the wavelength
is long
2. resonant antennas,

l / 2: most efficient; examples are slots, dipoles, patches
3. electrically large,

l >> : can be composed of many individual resonant
antennas; good for radar applications (high gain, narrow beam, low sidelobes)
Classification of antennas by type:
1. reflectors
2. lenses
3. arrays
Other designations: wire antennas, aperture antennas, broadband antennas
2
Radiation Integrals (1)
Consider a perfect electric conductor (PEC) with an electric surface current flowing on S.
In the case where the conductor is part of an antenna (a dipole), the current may be caused
by an applied voltage, or by an incident field from another source (a reflector). The
observation point is denoted by P and is given in terms of unprimed coordinate variables.
Quantities associated with source points are designated by primes. We can use any
coordinate system that is convenient for the particular problem at hand.
) , , ( z y x J
s

r
x
y
z
) , , ( or ) , , ( r P z y x P
r
r
r
r
r r R
r r
r

PEC WITH
SURFACE CURRENT
OBSERVATION
POINT
( ) ( ) ( )
2 2 2
z z y y x x R R + +
r
S
O
The medium is almost always free space (
o o
, ), but we continue to use ( , ) to cover
more general problems. If the currents are known, then the field due to the currents can be
determined by integration over the surface.
3
The vector wave equation for the electric field can be obtained by taking the curl of
Maxwells first equation:
s
J j E k E
r r r

2
A solution for E
r
in terms of the magnetic vector potential ) (r A
r
r
is given by
( )
j
r A
r A j r E
) (
) ( ) (
r
r
r
r
r
r

+ (1)
where ) (r
r
is a shorthand notation for (x,y,z) and s d e
R
J
r A
jkR
S
s

r
r
r
4
) (
We are particularly interested in the
case were the observation point is in the
far zone of the antenna ( P ). As P
recedes to infinity, the vectors r
r
and R
r
become parallel.
x
y
z
r
r
r
r
OBSERVATION
POINT
( ) r r r r R
r r
r
r r
r
4
In the expression for ) (r A
r
r
we use the approximation r R / 1 / 1 in the denominator and
( ) [ ] r r r r r R r
r r
r
in the exponent. Equation (2) becomes
( ) [ ] ( )
s d e J e
r
s d e J
r
r A
r r jk
S
s
jkr r r r r r jk
S
s

4 4
) (
r r r
r r
r
r
When this is inserted into equation (1), the del operations on the second term lead to
2
/ 1 r
and
3
/ 1 r terms, which can be neglected in comparison to the A j
r
term, which depends
only on r / 1 . Therefore, in the far field,
( )
s d e J e
r
j
r E
r r jk
S
s
jkr
4
) (
r
r
r
r
(discard the
r
E component) (3)
Explicitly removing the r component gives,
( ) [ ]
( )
s d e J r J e
r
jk
r E
r r jk
S
r
s s
jkr

4
) (
r
r r
r
r
The radial component of current does not contribute to the field in the far zone.
5
Notice that the fields have a spherical wave behavior in the far zone:
r
e
E
jkr
~
r
. The
spherical components of the field can be found by the appropriate dot products with E
r
.
More general forms of the radiation integrals that include magnetic surface currents (
ms
J
r
)
are:

1
]
1
S
r r k j ms
s
jkr
s d e
J
J e
r
k j
r E
4
) , , (
r
r
r

1
]
1
S
r r k j ms
s
jkr
s d e
J
J e
r
k j
r E
4
) , , (
r
r
r
The radiation integrals apply to an unbounded medium. For antenna problems

the following process is used:
1. find the current on the antenna surface, S,
2. remove the antenna materials and assume that the currents are suspended
in the unbounded medium, and
3. apply the radiation integrals.
6
Hertzian Dipole (1)
Perhaps the simplest application of the radiation integral is the calculation of the fields of
an infinitesimally short dipole (also called a Hertzian dipole). Note that the criterion for
short means much less than a wavelength, which is not necessarily physically short.
z J z J
o s
) , , (
r
a 2
l
z
x
y
a
For a thin dipole (radius, << a ) the

surface current distribution is independent
of . The current crossing a ring around
the antenna is
{
a J I
s
2
A/m
r
For a thin short dipole ( << l ) we assume

that the current is constant and flows along
the center of the wire; it is a filament of
zero diameter. The two-dimensional
integral over S becomes a one-dimensional
integral over the length,

S L
s
d I a s d J l
r r
2
7
Hertzian Dipole (2)
Using z z r
r
and cos sin sin cos sin z y x r + + gives cos z r r
r
. The
radiation integral for the electric field becomes
( )
{

l l
r
r
r
0 0
cos
4
4
) , , ( z d e
r
z I jk
z d z I e
r
jk
r E
z jk
e
jkr
e
jkr
l d
r r jk

However, because l is very short, 0 z k and 1
cos
z jk
e . Therefore,
( )
jkr jkr
e
r
z I jk
z d e
r
z I jk
r E

1
4
) , , (
0
l
r
l
leading to the spherical field components
0
4
sin
4
E E
e
r
I jk
e
r
z I jk
E E
jkr
jkr
r
l
l
r
x
y
z
l
z
SHORT CURRENT
FILAMENT
8
Hertzian Dipole (3)
Note that the electric field has only a 1/r dependence. The absence of higher order terms
is due to the fact that the dipole is infinitesimal, and therefore 0
ff
r . The field is a
spherical wave and hence the TEM relationship can be used to find the magnetic field
intensity
jkr
e
r
jkI E r E k
H

4
sin

l
r
The time-averaged Poynting vector is
{ } { } r
r
I k
r H E H E W
32
sin
2
1
2
1
2 2
2 2 2 2
* *
av

l
r r

The power flow is outward from the source, as expected for a spherical wave. The
average power flowing through the surface of a sphere of radius r surrounding the source
is

12
sin sin
32
2 2 2
3 / 8
2
2
0 0
2
2 2
2 2 2 2
0 0
av rad
l
4 4 4 4 4 3 4 4 4 4 4 2 1
l
r
I k
d d r r r
r
I k
ds n W P

W
9
Solid Angles and Steradians
Plane angles: s R, if s R then =1 radian
R
ARC LENGTH
s
Solid angles: A / R
2
, if A R
2
, then =1 steradian
SURFACE
AREA
A R
A / R
2
10
Directivity and Gain (1)
The radiation intensity is defined as
av
2
av
2
rad
) , ( W r W r r
d
dP
U
r r

and has units of Watts/steradian (W/sr). The directivity function or directive gain is
defined as
rad
2
rad
rad
av
4
) 4 /(
/
angle solid unit per radiated power average
angle solid unit per radiated power
) , (
P
W r
P
d dP
D
r

For the Hertzian dipole,

2
2
2
2
2 2
2 2
2
2
2
rad
2
sin
2
3
12
32
sin
4 4 ) , (
av

l
l
r
I k
r
I k
r
P
W r
D
The directivity is the maximum value of the directive gain
2
3
) , ( ) , (
max max max
D D D
o
11
Dipole Polar Radiation Plots
Half of the radiation pattern of the dipole is plotted below for a fixed value of . The half-
power beamwidth (HPBW) is the angular width between the half power points (1/ 2
below the maximum on the voltage plot, or 3dB below the maximum on the decibel plot).
0.5
1
1.5
30
210
60
240
90
270
120
300
150
330
180 0
-20
-10
0
10
30
210
60
240
90
270
120
300
150
330
180 0
-20
-10
0
10
30
210
60
240
90
270
120
300
150
330
180 0
-20
-10
0
10
30
210
60
240
90
270
120
300
150
330
180 0
FIELD (VOLTAGE) PLOT DECIBEL PLOT

The half power beamwidth of the Hertzian dipole,
B
:
( )
o o
90 2 45 707 . 0 sin sin ~
HP HP HP norm

B
E
12
Dipole Radiation Pattern
Radiation pattern of a Hertzian dipole aligned with the z axis. D
n
is the normalized
directivity. The directivity value is proportional to the distance from the center.
13
Another formula for directive gain is
2
norm
) , (
4
) , (
E D
A
r
where
A
is the beam solid angle

d d E
A
sin ) , (
2
norm
2
0 0
r

and ) , (
norm
E
r
is the normalized magnitude of the electric field pattern (i.e., the
normalized radiation pattern)
) , , (
) , , (
) , (
max
norm

r E
r E
E r
r
r
Note that both the numerator and denominator have the same 1/r dependence, and hence
the ratio is independent of r. This approach is often more convenient because most of our
calculations will be conducted directly with the electric field. Normalization removes all
of the cumbersome constants.
14
As an illustration, we re-compute the directivity of a Hertzian dipole. Noting that the
maximum magnitude of the electric field is occurs when 2 / , the normalized electric
field intensity is simply
sin ) , (
norm
E
r
The beam solid angle is

3
8
sin 2
sin ) , (
3 / 4
0
3
2
norm
2
0 0

43 42 1
r
d
d d E
A
and from the definition of directivity,

2
2
2
norm
sin
2
3
sin
3 / 8
4
) , (
4
) , (
E D
A
r
which agrees with the previous result.
15
Example
Find the directivity of an antenna whose far-electric field is given by
2
4
6
8
10
30
210
60
240
90
270
120
300
150
330
180 0
'
o o
o o
180 90 , cos
90 0 , cos
10
) , , (

r
e
r
e
r E
jkr
jkr
The maximum electric field occurs when r E / 10 1 cos
max

r
. The normalized
electric field intensity is
'
o o
o o
180 90 , cos 1 . 0
90 0 , cos
) , (
norm

E
which gives a beam solid angle of
) 1 . 1 (
3
2
sin cos 01 . 0 sin cos
2
2
0 2 /
2
2
0
2 /
0

+

d d d d
A
and a directivity of 37 . 7 45 . 5
o
D dB.
16
Beam Solid Angle and Radiated Power
In the far field the radiated power is
rad rad
2
2
0 0
2
2
0 0
*
rad
2 sin
2
1
2
1
rad
/
2
P F d d r E ds r H E P
F
E

'

4 4 4 4 3 4 4 4 4 2 1
r r r
4 4 3 4 4 2 1
r
From the definition of beam solid angle
2
2
max rad
2
2
2
0 0
2
2
max
2
norm
2
0 0
rad
sin
1
sin ) , (
r E F d d r E
r E
d d E
A
F
A
r
4 4 4 4 3 4 4 4 4 2 1
r
r
r

Equate the expressions for
rad
F
2
2
2
max
rad
r E
P
A
r
17
Gain vs. Directivity (1)
Directivity is defined with respect to the radiated power,
rad
P . This could be less than the
power into the antenna if the antenna has losses. The gain is referenced to the power into
the antenna,
in
P .
in
P
inc
P
ref
P
a
R
l
R
ANTENNA
I
Define the following:
inc
P power incident on the antenna terminals
ref
P power reflected at the antenna input
in
P power into the antenna
loss
P power loss in the antenna (dissipated in resistor
l l
R I P R
2
2
1
loss
, )
rad
P power radiated (delivered to resistor
a a
R I P R
2
2
1
rad
, ,
a
R is the radiation
resistance)
The antenna efficiency, e, is
in rad
eP P where 1 0 e .
18
Gain vs. Directivity (2)
Gain is defined as
) , (
/
/
4
) 4 /(
/
) , (
rad
rad
in
rad

eD
e P
d dP
P
d dP
G
Most often the use of the term gain refers to the maximum value of ) , ( G .
Example: The antenna input resistance is 50 ohms, of which 40 ohms is radiation
resistance and 10 ohms is ohmic loss. The input current is 0.1 A and the directivity of the
antenna is 2.
The input power is 25 . 0 ) 50 ( 1 . 0
2
2
1
in
2
2
1
in
R I P W
The power dissipated in the antenna is 05 . 0 ) 10 ( 1 . 0
2
2
1 2
2
1
loss

l
R I P W
The power radiated into space is 2 . 0 ) 40 ( 1 . 0
2
2
1 2
2
1
rad

a
R I P W
If the directivity is 2
o
D then the gain is 6 . 1 ) 2 (
25 . 0
2 . 0
in
rad
,
_
,
_
D
P
P
eD G
19
Azimuth/Elevation Coordinate System
Radars frequently use the azimuth/elevation coordinate system: (Az,El) or (, ) or
(
a e
, ). The antenna is located at the origin of the coordinate system; the earth's surface
lies in the x-y plane. Azimuth is generally measured clockwise from a reference (like a
compass) but the spherical system azimuth angle is measured counterclockwise from the
x axis. Therefore 360 and 90 degrees.
CONSTANT
ELEVATION
x
y
z
ZENITH
HORIZON
P
r
20
Approximate Directivity Formula (1)
Assume the antenna radiation pattern is a pencil beam on the horizon. The pattern is
constant inside of the elevation and azimuth half power beamwidths (
a e
, ) respectively:
x
y
z
a
2 /
0

0
21
Approximate Directivity Formula (2)
Approximate antenna pattern
( ) ( )
'
else , 0
2 / 2 / and 2 / 2 / 2 / 2 / ,
) , (
a a e e
jkr
o
r
e E
E

r
The beam solid angle is
{
( ) ( ) [ ]
[ ]
e a e e a
e e a
A
d d
e
e
a
a

+

) 2 / ( 2 /
2 / sin 2 / sin
sin
1
2 2
2 2
2
2
This leads to an approximation for the directivity of
a e A
o
D

4 4
. Note that the

angles are in radians. This formula is often used to estimate the directivity of an omni-
directional antenna with negligible sidelobes.
22
Thin Wire Antennas (1)
Thin wire antennas satisfy the condition << a . If the length of the wire (l) is an integer
multiple of a half wavelength, we can make an educated guess at the current based on
an open circuited two-wire transmission line
OPEN
CIRCUIT
I(z)
z
/ 4
FEED POINTS
2 /
For other multiples of a half wavelength the current distribution has the following features

l / 2

l

l 3 / 2
FEED POINT
LOCATED AT
MAXIMUM
CURRENT GOES
TO ZERO AT END
23
On a half-wave dipole the current can be approximated by
) cos( ) ( kz I z I
o
for 4 / 4 / < < z
Using this current in the radiation integral
4 /
4 /
cos
4 /
4 /
cos
) cos(
4
) cos(
4
) , , (

z d e z k z e
r
I jk
z d e z k I z e
r
jk
r E
z jk jkr
o
z jk
o
jkr
r
From a table of integrals we find that
2 2
1 0
] [ ) ( sin ) cos(
) cos(
B A
z B B z B A e
z d e z B
z A
z A
+
+

t
48 47 6 48 47 6
where cos jk A and k B , so that
2 2 2 2 2 2 2
sin cos k k k B A + + . The
component requires the dot product sin
z .
24
Evaluating the limits gives
sin
cos
2
cos
2
sin
) 1 (
sin
4
2 2
cos cos 2
2 / cos 2 / cos
] [
2
,
_
,
_
jkr
o
j j
jkr
o
e
r
I j
k
e e k
e
r
I jk
E
4 4 4 4 4 8 4 4 4 4 4 7 6
The magnetic field intensity in the far field is
sin
cos
2
cos
2

,
_
jkr
o
e
r
jI
H
E E k
H
r
r
The directivity is computed from the beam solid angle, which requires the normalized
electric field intensity
( )
2
2
2
max
2
2
norm
sin
cos cos

E
E
E
r
25
( ) ( )
) 218 . 1 )( 2 (
y numericall Integrate
sin
2 / cos cos
2 sin
sin
2 / cos cos
2
2
2
0
2
2
0

4 4 4 4 3 4 4 4 4 2 1
d d
A
The directive gain is
( ) ( )

2
2
2
2
2
2
2
norm
sin
cos cos
64 . 1
sin
cos cos
) 218 . 1 )( 2 (
4
) , (
4

E D
A
r
The radiated power is
a o o
o
A
R I I
r
r I
r E
P
2 2
2
2
2
2
2
2
rad
2
1
57 . 36
) 2 ( 2
) 218 . 1 ( 2
2
max

where
a
R is the radiation resistance of the dipole. The radiated power can be viewed as
the power delivered to resistor that represents free space. For the half-wave dipole the
radiation resistance is
13 . 73 ) 57 . 36 )( 2 (
2
2
rad

o
a
I
P
R ohms
26
Numerical Integration (1)
The rectangular rule is a simple way of evaluating an integral numerically. The area under
the curve of ) ( x f is approximated by a sum of rectangular areas of width and height
) (
n
x f , where a n x
n
+ +

) 1 (
2
is the center of the interval nth interval. Therefore, if
all of the rectangles are of equal width

b
a
n
b
a
x f dx x f ) ( ) (
Clearly the approximation can be made as close to the exact value as desired by reducing
the width of the triangles as necessary. However, to keep computation time to a minimum,
only the smallest number of rectangles that provides a converged solution should be used.
x
f ( x)
b
a

L
1
2 3
4
N
1
x
2
x
N
x

27
Numerical Integration (2)
Example: Matlab programs to integrate
( )
d
sin
2 / cos cos
2
0
Sample Matlab code for the rectangular rule

% integrate dipole pattern using the rectangular rule
clear
rad=pi/180;
% avoid 0 by changing the limits slightly
a=.001; b=pi-.001;
N=5
delta=(b-a)/N;
sum=0;
for n=1:N
theta=delta/2+(n-1)*delta;
sum=sum+cos(pi*cos(theta)/2)^2/sin(theta);
end
I=sum*delta
Convergence: N=5, 1.2175; N=10, 1.2187; N=50, 1.2188. Sample Matlab code using the
quad8 function
% integrate to find half wave dipole solid angle
clear
I=quad8('cint',0.0001,pi-.0001,.00001);
disp(['cint integral, I: ',num2str(I)])
function P=cint(T)
% function to be integrated
P=(cos(pi*cos(T)/2).^2)./sin(T);
28
Thin Wires of Arbitrary Length
For a thin-wire antenna of length l along the z axis, the electric field intensity is
1
1
1
1
]
1
,
_
,
_
sin
2
cos cos
2
cos
2
l l k k
e
r
I j
E
jkr
Example: 5 . 1 l (left: voltage plot; right: decibel plot)
1
2
30
210
60
240
90
270
120
300
150
330
180 0
-20
-10
0
10
30
210
60
240
90
270
120
300
150
330
180 0
-20
-10
0
10
30
210
60
240
90
270
120
300
150
330
180 0
-20
-10
0
10
30
210
60
240
90
270
120
300
150
330
180 0
29
Feeding and Tuning Wire Antennas (1)
When an antenna terminates a transmission line, as shown below, the antenna impedance
(
a
Z ) should be matched to the transmission line impedance (
o
Z ) to maximize the power
delivered to the antenna
o
Z
a
Z
o a
o a
Z Z
Z Z
+
and

+
1
1
VSWR
The antennas input impedance is generally a complex quantity, ( )
a a a
jX R R Z + +
l
.
The approach for matching the antenna and increasing its efficiency is
1. minimize the ohmic loss, 0
l
R
2. tune out the reactance by adjusting the antenna geometry or adding lumped
elements, 0
a
X (resonance occurs when
a
Z is real)
3. match the radiation resistance to the characteristic impedance of the line by
adjusting the antenna parameters or using a transformer section,
o a
Z R
30
Example: A half-wave dipole is fed by a 50 ohm line
1870 . 0
50 73
50 73

o a
o a
Z Z
Z Z
46 . 1
1
1
VSWR

+
The loss due to reflection at the antenna terminals is

965 . 0 1
2 2

( ) 155 . 0 log 10
2
dB
which is stated as 0.155 dB of reflection loss (the negative sign is implied by using the
word loss).
31
The antenna impedance is affected by
1. length
2. thickness
3. shape
4. feed point (location and method of feeding)
5. end loading
Although all of these parameters affect both the real and imaginary parts of
a
Z , they are
generally used to remove the reactive part. The remaining real part can be matched using
a transformer section.
Another problem is encountered when matching a balanced radiating structure like a
dipole to an unbalanced transmission line structure like a coax.
g
V
+
/ 2
COAX
DIPOLE
g
V
+
/ 2
TWO-WIRE LINE
DIPOLE
UNBALANCED FEED
BALANCED FEED
32
If the two structures are not balanced, a return current can flow on the outside of the
coaxial cable. These currents will radiate and modify the pattern of the antenna. The
unbalanced currents can be eliminated using a balun (balanced-to-unbalanced transformer)
UNBALANCED FEED BALANCED FEED
b a
V V
b a
V V
) (z I
. . . .
b a a b
Baluns frequently incorporate chokes, which are circuits designed to choke off current
by presenting an open circuit to current waves.
33
An example of a balun employing a choke is the sleeve or bazooka balun
4 /
o
Z
o
Z
1
I
2
I
1
I
2
I
HIGH IMPEDANCE (OPEN CIRCUIT)
PREVENTS CURRENT FROM FLOWING
ON OUTSIDE
SHORT CIRCUIT
The choke prevents current from flowing on the exterior of the coax. All current is
confined to the inside of surfaces of the coax, and therefore the current flow in the two
directions is equal (balanced) and does not radiate. The integrity of a short circuit is easier
to control than that of an open circuit, thus short circuits are used whenever possible.
Originally balun referred exclusively to these types of wire feeding circuits, but the term
has evolved to refer to any feed point matching circuit.
34
Calculation of Antenna Impedance (1)
The antenna impedance must be matched to that of the feed line. The impedance of an
antenna can be measured or computed. Usually measurements are more time consuming
(and therefore expensive) relative to computer simulations. However, for a simulation to
accurately include the effect of all of the antennas geometrical and electrical parameters
on
a
Z , a fairly complicated analytical model must be used. The resulting equations must
be solved numerically in most cases.
One popular technique is the method of moments (MM) solution of an integral equation
(IE) for the current.
g
V
+
g
E
a 2
2 / l
2 / l
z
2 / b
2 / b
1. ) (z I is the unknown current distribution on the
wire
2. Find the z component of the electric field in
terms of ) (z I from the radiation integral
3. Apply the boundary condition

E
z
( a)
0, b / 2 z l / 2
E
g
, b / 2 z

'

in order to obtain an integral equation for ) (z I
35
One special form of the integral equation for thin wires is Pocklingtons equation

k
2
+

2
z
2

_
,
I( z )
e
jkR
4R
l / 2
l / 2
d z
0, b/ 2 z l / 2
jE
g
, b / 2 z

'

where
2 2
) ( z z a R + . This is called an integral equation because the unknown
quantity ) (z I appears in the integrand.
4. Solve the integral equation using the method of moments (MM). First approximate the
current by a series with unknown expansion coefficients { }
n
I
I( z ) I
n
n
( z )
n1
N
The basis functions or expansion functions { }

n
are known and selected to suit the
particular problem. We would like to use as few basis functions as possible for
computational efficiency, yet enough must be used to insure convergence.
36
Example: a step approximation to the current using a series of pulses. Each segment is
called a subdomain. Problem: there will be discontinuities between steps.
z
I( z )

l
2

l
2
I
1
1
I
2
2
INN

L

L
CURRENT STEP
APPROXIMATION
z
1
z
2
z
N
0

A better basis function is the overlapping piecewise sinusoid
z
I( z )

l
2

l
2
I
1
1
I
2
2
INN
L

L
CURRENT
PIECEWISE
SINUSOID
z
1
z
2
z
N
0

37
A piecewise sinusoid extends over two segments (each of length ) and has a maximum at
the point between the two segment
z
n
( z )
z
n
) (
1 +
+
n
n
z
z
) (
1

n
n
z
z
n
( z )
z z
n1
z
n
z
n1
, z
n1
z z
n
z
n+1
z
z
n+1
z
n
, z
n
z z
n+1
0, elsewhere

'

Entire domain functions are also
possible. Each entire domain
basis function extends over the
entire wire. Examples are
sinusoids.
z

l
2

l
2
n
( z )
38
Solving the integral equation: (1) insert the series back into the integral equation

k
2
+

2
z
2

_
,
I
n
n
( z )
n1
N

_
,

e
jkR
4R
l / 2
l / 2
d z
0, b / 2 z l / 2
jE
g
, b/ 2 z

'

Note that the derivative is with respect to z (not z ) and therefore the differential operates
only on R. For convenience we define new functions f and g:
4 4 4 4 4 3 4 4 4 4 4 2 1
l
4 4 4 4 4 4 4 3 4 4 4 4 4 4 4 2 1
l
l
g
g
N
n
z f f
z z jkR
n n
z b E j
z b
z d
z z R
e
z
z
k I
n n

'
'
,
_
+
2 / ,
2 / 2 / , 0
) , ( 4
) (
1
) ( ) (
2 /
2 /
) , (
2
2
2
Once
n
is defined, the integral can be evaluated numerically. The result will still be a
function of z, hence the notation ) (z f
n
.
(2) Choose a set of N testing (or weighting) functions
m
{ }. Multiply both sides of the
equation by each testing function and integrate over the domain of each function
m
to
obtain N equations of the form
dz z g z dz z f I z
m m
m
N
n
n n m

) ( ) ( ) ( ) (
1
, m 1, 2,..., N
39
Interchange the summation and integration operations and define new impedance and
voltage quantities
4 4 3 4 4 2 1
4 4 4 3 4 4 4 2 1
m
m m
V
m
mn
mn m
N
n
n
dz z g z
Z
dz z f z I

'
) ( ) ( ) ( ) (
1
This can be cast into the form of a matrix equation and solved using standard matrix
methods
I
n
n1
N
Z
mn
V
m
Z [ ] I [ ] V [ ] I [ ] Z [ ]
1
V [ ]
[ ] Z is a square impedance matrix that depends only on the geometry and material
characteristics of the dipole. Physically, it is a measure of the interaction between the
currents on segments m and n. [ ] V is the excitation vector. It depends on the field in the
gap and the chosen basis functions. [ ] I is the unknown current coefficient vector.
After [ ] I has been determined, the resulting current series can be inserted in the radiation
integral, and the far fields computed by integration of the current.
40
Self-Impedance of a Wire Antenna
The method of moments current allows calculation of the self impedance of the antenna by
taking the ratio Z
self
V
g
/ I
o
RESISTANCE REACTANCE
41
The Fourier Series Analog to MM
The method of moments is a general solution method that is widely used in all of
engineering. A Fourier series approximation to a periodic time function has the same
solution process as the MM solution for current. Let ) (t f be the time waveform
( ) ( ) [ ]
+ +

1
sin cos
2
) (
n
n n n n
o
t b t a
T T
a
t f
For simplicity, assume that there is no DC component and that only cosines are necessary
to represent ) (t f (true if the waveform has the right symmetry characteristics)
( )

1
cos
2
) (
n
n n
t a
T
t f
The constants are obtained by multiplying each side by the testing function cos(
m
t) and
integrating over a period
( ) ( ) ( )
'

,
_
n m a
n m
dt t t a
T
dt t t f
n
T
T
m
n
n n m
T
T
,
, 0
cos cos
2
cos ) (
2 /
2 /
1
2 /
2 /

This is analogous to MM when ) ( ) ( z I t f ,
n n
I a , ) cos( t
n n
, and
) cos( t
m m
. (Since ) (t f is not in an integral equation, a second variable t is not
required.) The selection of the testing functions to be the complex conjugates of the
expansion functions is referred to as Galerkins method.
42
Reciprocity (1)
When two antennas are in close proximity to each other, there is a strong interaction
between them. The radiation from one affects the current distribution of the other, which
in turn modifies the current distribution of the first one.
V
1
+
I
1
Z
1
I
2
Z
2
V
21
E
g
ANTENNA 1
(SOURCE)
ANTENNA 2
(RECEIVER)
Consider two situations (depicted on the following page) where the geometrical
relationship between two antennas does not change.
1. A voltage is applied to antenna 1 and the current induced at the terminals of antenna 2 is
measured.
2. The situation is reversed: a voltage is applied to antenna 2 and the current induced at the
terminals of antenna 1 is measured.
43
Reciprocity (2)
Case 1:
V
1
+
ANTENNA 1
ANTENNA 2
I
2
ENERGY FLOW
CURRENT
METER
Case 2:
ANTENNA 1
ENERGY FLOW
CURRENT
METER
I
1
V
2
ANTENNA 2
+
44
Mutual Impedance (1)
Define the mutual impedance or transfer impedance as
Z
12

V
2
I
1
and Z
21

V
1
I
2
The first index on Z refers to the receiving antenna (observer) and the second index to the
source antenna.
Reciprocity Theorem: If the antennas and medium are linear, passive and isotropic, then
the response of a system to a source is unchanged if the source and observer (measurer)
are interchanged.
With regard to mutual impedance:
This implies that Z
21
Z
12
The receiving and transmitting patterns of an antenna are the same if it is
constructed of linear, passive, and isotropic materials and devices.
In general, the input impedance of antenna number n in the presence of other antennas is
obtained by integrating the total field in the gap (gap width b
n
)

Z
n

V
n
I
n
1
I
n
r
E
n

b
n
d
r
l
45

r
E
n
is the total field in the gap of antenna n (due to its own voltage plus the incident fields
from all other antennas).
V
n E
n
I
n
46
If there are a total of N antennas

r
E
n

r
E
n1
+
r
E
n2
+L+
r
E
nN
Therefore,

Z
n

1
I
n
r
E
nm
m1
N

b
n
d
r
l
n
Define

n
b
n nm nm
d E V l
r r
The impedance becomes
{
nm
N
m
Z
n
nm
N
m
V
N
m
nm
n
n
Z
I
V
V
I
Z
nm n

1 1 1
1
3 2 1
For example, the impedance of dipole n=1 is written explicitly as

Z
1
Z
11
+ Z
12
+L+ Z
1N
When m = n the impedance is the self impedance. This is approximately the impedance
that we have already computed for an isolated dipole using the method of moments.
47
The method of moments can also be used to compute the mutual coupling between
antennas. The mutual impedance is obtained from the definition
Z
mn

V
m
I
n
I
m
0
mutual impedance at port m due to a

current in port n, with port m open ciruited
By reciprocity this is the same as Z
mn
Z
nm

V
n
I
m
I
n
0
. Say that we have two dipoles, one
distant (n) and one near (m). To determine Z
mn
a voltage can be applied to the distant
dipole and the open-circuited current computed on the near dipole using the method of
moments. The ratio of the distant dipoles voltage to the current induced on the near one
gives the mutual impedance between the two dipoles.
Plots of mutual impedance are shown on the following pages:
1. high mutual impedance implies strong coupling between the antennas
2. mutual impedance decreases with increasing separation between antennas
3. mutual impedance for wire antennas placed end to end is not as strong as when
they are placed parallel
48
Mutual Impedance of Parallel Dipoles
0 0.5 1 1.5 2
-40
-20
0
20
40
60
80
Separation, d (wavelengths)
R

o
r

X

(
o
h
m
s
)
d
2 /
12
R
12
X
49
Mutual Impedance of Colinear Dipoles
0 0.5 1 1.5 2
-10
-5
0
5
10
15
20
25
Separation, s (wavelengths)
R

o
r

X

(
o
h
m
s
)
2 /
s
21
R
21
X
50
Mutual-Impedance Example
Example: Assume that a half wave dipole has been tuned so that it is resonant ( X
a
0).
We found that a resonant half wave dipole fed by a 50 ohm transmission line has a VSWR
of 1.46 (a reflection coefficient of 0.1870). If a second half wave dipole is placed parallel
to the first one and 0.65 wavelength away, what is the input impedance of the first dipole?
From the plots, the mutual impedance between two dipoles spaced 0.65 wavelength is
Z
21
R
21
+ jX
21
24.98 j7.7
Noting that
12 21
Z Z the total input impedance is
Z
in
Z
11
+ Z
21
73 + (24.98 j7.7)
48.02 j7.7
The reflection coefficient is

Z
in
Z
o
Z
in
+ Z
o
(48.02 j7.7) 50
(48.02 j7.7) + 50
0.0139 j0.0797 0.081e
j 99.93
o
which corresponds to a VSWR of 1.176. In this case the presence of the second dipole has
improved the match at the input terminals of the first antenna.
51
Broadband Antennas (1)
The required frequency bandwidth increases with the rate of information transfer (i.e., high
data rates require wide frequency bandwidths). Designing an efficient wideband antenna
is difficult. The most efficient antennas are designed to operate at a resonant frequency,
which is inherently narrow band.
Two approaches to operating over wide frequency bands:
1. Split the entire band into sub-bands and use a separate resonant antenna in each band
Advantage: The individual antennas are easy to design (potentially inexpensive)
Disadvantage: Many antennas are required (may take a lot of space, weight, etc.)
F
R
E
Q
U
E
N
C
Y
M
U
L
T
I
P
L
E
X
E
R
BROADBAND
INPUT SIGNAL
1
2
N

M
f
1
f
2
f
N

L
f
1
f
2
f
N
f
BANDWIDTH OF AN
INDIVIDUAL ANTENNA
TOTAL SYSTEM
BANDWIDTH
52
2. Use a single antenna that operates over the entire frequency band
Advantage: Less aperture area required by a single antenna
Disadvantage: A wideband antenna is more difficult to design than a narrowband antenna
Broadbanding of antennas can be accomplished by:
1. interlacing narrowband elements having non-overlapping sub-bands (stepped band
approach)
Example: multi-feed point dipole
d
min
d
max
TANK CIRCUIT USING
LUMPED ELEMENTS
OPEN CIRCUIT AT
HIGH FREQUENCIES
SHORT CIRCUIT AT
LOW FREQUENCIES
53
2. design elements that have smooth geometrical transitions and avoid abrupt
discontinuities
Example: biconical antenna
d
min
d
max
spiral antenna
FEED POINT
WIRES
d
min
d
max
54
Circular Spiral in Low Observable Fixture
55
Another example of a broadband antenna is the log-periodic array. It can be classified as a
single element with a gradual geometric transitions or as discrete elements that are
resonant in sub-bands. All of the elements of the log periodic antenna are fed.
d
min d
max
DIRECTION
OF MAXIMUM
RADIATION
The range of frequencies over which ana antenna operates is determined approximately by
the maximum and minimum antenna dimensions
d
min

H
2
and d
max

L
2
56
Yagi-Uda Antenna
A Yagi-Uda (or simply Yagi) is used at high frequencies (HF) to obtain a directional
azimuth pattern. They are frequently employed as TV/FM antennas. A Yagi consists of a
fed element and at least two parasitic (non-excited) elements. The shorter elements in the
front are directors. The longer element in the back is a reflector. The conventional design
has only one reflector, but may have up to 10 directors.
REFLECTOR
DIRECTORS
FED
ELEMENT
MAIN LOBE
MAXIMUM
DIRECTIVITY
BACK LOBE
MAXIMUM
DIRECTIVITY
POLAR PLOT OF
RADIATION PATTERN
The front-to-back ratio is the ratio of the maximum directivity in the forward direction to
that in the back direction: D
max
/ D
back
57
Ground Planes and Images (1)
In some cases the method of images allows construction of an equivalent problem that is
easier to solve than the original problem.
When a source is located over a PEC ground plane, the ground plane can be removed and
the effects of the ground plane on the fields outside of the medium accounted for by an
image located below the surface.
ORIGINAL PROBLEM EQUIVALENT PROBLEM
h h
h
ORIGINAL
SOURCES
IMAGE
SOURCES
REGION 1
REGION 2
GROUND PLANE
REMOVED
I dl
I dl
I dl
I dl
I dl
I dl
PEC
The equivalent problem holds only for computing the fields in region 1. It is exact for an
infinite PEC ground plane, but is often used for finite, imperfectly conducting ground
planes (such as the Earths surface).
58
The equivalent problem satisfies Maxwells equations and the same boundary conditions
as the original problem. The uniqueness theorem of electromagnetics assures us that the
solution to the equivalent problem is the same as that for the original problem.
Boundary conditions at the surface of a PEC: the tangential component of the electric field
is zero.
ORIGINAL PROBLEM EQUIVALENT PROBLEM
PEC
h
h
h
Idz
Idz
Idz
1
E
2
E
E
||
0
1
TANGENTIAL
COMPONENTS
CANCEL
A similar result can be shown if the current element is oriented horizontal to the ground
plane and the image is reversed from the source. (A reversal of the image current direction
implies a negative sign in the images field relative to the source field.)
59
Half of a symmetric conducting structure can be removed if an infinite PEC is placed on
the symmetry plane. This is the basis of a quarter-wave monopole antenna.
z 0
I
V
g
+
2b
I
V
g
+
b
/ 4 / 4
ORIGINAL
DIPOLE
MONOPOLE
PEC
The radiation pattern is the same for the monopole as it is for the half wave dipole
above the plane z 0
The field in the monopole gap is twice the field in the gap of the dipole
Since the voltage is the same but the gap is half of the dipoles gap
56 . 36
2
12 . 73
dipole
2
1
monopole

a a
R R ohms
60
Crossed Dipoles (1)
Crossed dipoles (also known as a
turnstile) consists of two orthogonal
dipoles excited 90 degrees out of phase.
I
x
I
o
I
y
I
o
e
j / 2
jI
o

I
x
x
z
y
I
y
The radiation integral gives two terms
,
_
2 /
2 /
sin sin
2 /
2 /
cos sin
sin cos cos cos

4
l
l
l
l
43 42 1 43 42 1
y d e I j x d e I
jkr o k y j
y
o
k x j
x
o
e
r
jk
E

If 1 << l k then

2 /
2 /
cos sin
l
l
l x d e
k x j
and similarly for the y integral. Therefore,
( ) ( )

sin cos cos sin cos cos

4
j
jkr
o
j
jkr
E
o
r
e
E
r
e I jk
E
o
+
43 42 1
l
61
Crossed Dipoles (2)
A similar result is obtained for E
( ) ( )

cos sin cos sin

4
j
jkr
o
j
jkr
E
o
r
e
E
r
e I jk
E
o
3 2 1
l
Consider the components of the wave propagating toward an observer on the z axis
0 : E
E
o
, E
jE
o
, or
( )
y j x
r
e
E E
jkr
o
+
r
which is a circularly polarized wave. If the observer is not on the z axis, the projected
lengths of the two dipoles are not equal, and therefore the wave is elliptically polarized.
The axial ratio (AR) is a measure of the waves ellipticity at the specified , :
AR
E
max
E
min
, 1 AR
For the crossed dipoles
AR
E
1
cos
2
cos
2
+sin
2
( )
1
cos
62
Crossed Dipoles (3)
The rotating linear pattern is shown. A linear receive antenna rotates like a propeller blade
as it measures the far field at range r. The envelope of the oscillations at any particular
angle gives the axial ratio at that angle. For example, at 50 degrees the AR is about
1/ 0.64 1.56 1.93 dB.
0 50 100 150 200 250 300 350
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
N
O
R
M
A
L
I
Z
E
D

F
I
E
L
D
PATTERN ANGLE, (DEGREES)
E
1
ROTATING LINEAR
POLARIZATION
63
Crossed Dipoles (4)
Examples of rotating linear patterns on crossed dipoles that are not equal in length
0.5
1
30
210
60
240
90
270
120
300
150
330
180 0
-30
-20
-10
0
30
210
60
240
90
270
120
300
150
330
180 0
VOLTAGE PLOT DECIBEL PLOT
64
Polarization Loss (1)
For linear antennas an effective height (

r
h
e
) can be defined

V
oc

r
E
inc

r
h
e
V
oc

r
E
inc

r
h
e
The open circuit voltage is a maximum when the antenna is aligned with the incident
electric field vector. The effective height of an arbitrary antenna can be determined by
casting its far field in the following form of three factors
[ ] [ ] ) , ( ) , , (
e
jkr
o
h
r
e
E r E
r r
1
]
1
The effective height accounts for the incident electric field projected onto the antenna
element. The polarization loss factor (PLF) between the antenna and incident field is

PLF, p
r
E
inc

r
h
e
2
r
E
inc
2
r
h
e
2
65
Example: The Hertzian dipoles far field is
[ ]
43 42 1
l
r
r
) , (
sin
4
) , , (

e
h
jkr
o
r
e kI j
r E
1
]
1
1
]
1
If we have a second dipole that is rotated by an angle in a plane parallel to the plane
containing the first dipole, we can calculate the PLF as follows. First,

V
oc

r
E
inc

r
h
e

r
E
inc
z
r
h
e
z
r
E
inc
r
h
e
z z
r
E
inc
r
h
e
cos
V
oc

r
E
inc

r
h
e
z
x
z
z
y y
TRANSMIT
RECEIVE
66
The PLF is
2
2 2
inc
2
2 2
inc
2 2
inc
2
inc
cos
cos

e
e
e
e
h E
h E
h E
h E
p
r r
r r
r r
r r
When the dipoles are parallel, p=1, and there is no loss due to polarization mismatch.
However, when the dipoles are at right angles, p=0 and there is a complete loss of signal.
A more general case occurs when the incident field has both and components

r
E
inc
E
i

+ E
i

( )
2 2
2

e i i
e i i
h E E
h E E
p
r
r

+
+
Example: The effective height of a RHCP antenna which radiates in the +z direction is
given by the vector ( )

j h h
o e

r
. A LHCP field is incident on this antenna (i.e., the
incident wave propagates in the z direction):
( )
jkz
o
e j E E

inc

r
67
The PLF is
( ) ( )
0
2 2

2 2
2
o o
jkz
o o
h E
e j j h E
p

If a RHCP wave is incident on the same antenna, again propagating along the z axis in the
negative direction, ( )
jkz
o
e j E E

inc
+
r
. Now the PLF is
( ) ( )
1
2 2

2 2
2
o o
jkz
o o
h E
e j j h E
p

Finally, if a linearly polarized plane wave is incident on the antenna,
jkz
o
e E E
inc

r
( )
2 / 1
2

2
2
2
o o
jkz
o o
h E
e j h E
p

If a linearly polarized antenna is used to receive a circularly polarized wave (or the reverse
situation), there is a 3 dB loss in signal.
68
Antenna Polarization Loss
Summary of polarization
losses for polarization
mismatched antennas
TRANSMIT
ANTENNA
RECEIVE
ANTENNA
RHCP
LHCP
V
H
RHCP
V
0 dB
3 dB
3 dB
> 25 dB
0 dB
3 dB
> 25 dB
69
Aircraft Blade Antenna
Blade antennas are used for telemetry and communications. They have nearly
hemispherical coverage, allowing the aircraft to maneuver without a complete loss of
signal. The resulting polarization is often called slant because it contains both horizontal
and vertical components.
Blade antenna
3. 0 (7.62 cm)
1.5 8
(4.01cm)
h

45
o
/ 4 THIN WIRE
Method of moments model:
Location of source
excitati on
Comparison of measured and calculated
patterns for a blade installed on a cylinder
70
Blade Antennas Installed on an Aircraft
The top antenna (A) provides coverage in the upper hemisphere, while the bottom antenna
(B) covers the lower hemisphere. The two antennas can be duplexed (switched) or their
signals combined using a coupler.
Method of moments patch model: Elevation pattern: (signals
combined with 3 dB coupler)
Roll
Pit ch
Yaw
A
B
x
y
z
71
Small Loop Antenna (1)
By symmetry, we expect that the field of a small wire loop located in the z = 0 plane will
depend only on the angle off of the wire axis, . Because of the azimuthal symmetry
cylindrical coordinates are required to solve this problem. If the wire is very thin (a
filament) and has a constant current
o
I flowing in it, the radiation integral is
( )

2
0
4
) , , (
43 42 1
r
r r
l d
r r jk
d a I e
r
jk
r E e
o
jkr
z
x
y
a r
r
r
r
R
r
Using the transformation tables
( )
( )

+
+
+ +
sin sin cos cos sin
sin cos
cos sin sin cos sin
a r r
y x a a r
z y x r
r
r
We also need
in terms of the cartesian

unit vectors (
is not a constant that can be

moved outside of the integral)
+ cos sin
y x
72
The radiation integral becomes
( )
( )

2
0
sin sin cos cos sin
cos sin
4
) , , ( d e y x e
r
a I jk
r E
jka
jkr o
r
For a small loop 1 << ka and the exponential can be represented by the first two terms of a
Taylors series to get
( )
( )

+ +
+
sin sin cos cos sin 1
cos sin sin cos sin
jka e
jka
Inserting the approximation in the integral
( )
( )
[ ]

2
0
sin sin cos cos sin 1 cos sin d jka y x
Since 0 cos sin
2
0
2
0

d d the 1 in the square brackets can be dropped. The
remaining terms in the integrand involve the following factors:

of function odd an is cos sin because zero to integrates cos sin sin
cos cos
of function odd an is cos sin because zero to integrates cos cos sin
sin sin
2
2
73
The only two terms that do not integrate to zero are of the form

2
0
2
0
2 2
cos sin d d
Therefore,
( )
( )
jkr o jkr o
jkr
o
e
r
a I k
y x e
r
a I k
d y d x e
r
jka I jka
r E

,
_

sin
4
cos sin sin

4
cos cos sin sin
4
sin
) , , (
2 2 2 2
2 2
2
0
2
0
4 4 4 4 3 4 4 4 4 2 1
r
The radiation pattern of the small loop is the same as that of a short dipole aligned with the
loop axis. The radiated power is
12
4 2 4
rad
a I k
P
o

and the radiation resistance

4
4
4 4 4 4
2
rad
320
6
) 120 ( ) / 2 (
6
2
,
_
a a a k
I
P
R
o
a
74
Helix Antenna (1)
A helix is described by the following parameters:
D = diameter
A = axial length
C = circumference ( D )
S = center-to-center spacing
between turns
L = length of one turn
N = number of turns

) / ( tan
1
D S pitch angle
d
D
S
D C
S
L
GROUND PLANE
z
Conventional helices are constructed with air or low-dielectric cores. A helix is capable of
operating in several different radiation modes and polarizations, depending on the
combination of parameter values and the frequency.
75
Helix Antenna (2)
Radiation modes of a helix:
> C
SMALL (LOOKS
LIKE A LOOP)
LARGE
NORMAL MODE AXIAL MODE
In both cases the radiation is circularly
polarized:
Normal mode:
2 2
2
AR
D
S
Axial mode:
n
n
2
1 2
AR
+
In the axial mode, the beamwidth decreases

with increasing helix length, NS
/ ) / (
115
BWFN
NS C
o
for
o o
13 12 < < .

Antenas N Wave Propagation

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Antenas N Wave Propagation

Transféré par

Droits d'auteur :

Formats disponibles

Naval Postgraduate School Distance Learning

Antennas & Propagation

The radiation integrals apply to an unbounded medium. For antenna problems

For a thin dipole (radius, << a ) the

For a thin short dipole ( << l ) we assume

. Note that the

Sample Matlab code for the rectangular rule

The loss due to reflection at the antenna terminals is

The basis functions or expansion functions { }

mutual impedance at port m due to a

sin cos cos cos

sin cos cos sin cos cos

cos sin cos sin

in terms of the cartesian

is not a constant that can be

cos sin sin

and the radiation resistance

In the axial mode, the beamwidth decreases

Vous aimerez peut-être aussi