22.8.

Problems

1041

22.2 Using the higher-order terms in the series (22.3.23), show that the input impedance Zin =
R + jX of a small dipole is given as follows to order (kl)4 , where L = ln(2a/l):

R=

1
1
(kl)2 +
(kl)4 ,
12
360

X=

4(1 + L)

kl

1
3

23






2
1
11
L+
(kl)
L
(kh)3
3

180

30

Appendices

22.3 Consider a small dipole with a linear current given by Eq. (22.3.25). Determine the radiation
vector, and the radiated electric and magnetic elds at a far distance r from the dipole. Calculate the radiated power Prad by integrating the radial Poynting vector over a large sphere.
Then identify the radiation resistance R through the denition:

Prad =

1
R|I0 |2
2

and show R is the same as that given by Eq. (22.3.24)

A. Physical Constants
We use SI units throughout this text. Simple ways to convert between SI and other
popular units, such as Gaussian, may be found in Refs. [123126].
The Committee on Data for Science and Technology (CODATA) of NIST maintains
the values of many physical constants [112]. The most current values can be obtained
from the CODATA web site [1433]. Some commonly used constants are listed below:
quantity

symbol

speed of light in vacuum

permittivity of vacuum
permeability of vacuum
characteristic impedance
electron charge
electron mass
Boltzmann constant
Avogadro constant
Planck constant
Gravitational constant
Earth mass
Earth equatorial radius

value

units

c0 , c
0
0
0 , Z0

299 792 458

8.854 187 817 1012
4 107
376.730 313 461

m s1
F m1
H m 1

e
me

1.602 176 462 1019

9.109 381 887 1031

C
kg

k
NA , L
h

1.380 650 324 1023

6.022 141 994 1023
6.626 068 76 1034

J K1
mol1
J/Hz

6.672 59 1011
5.972 1024
6378

m3 kg1 s2
kg
km

G
M
ae

In the table, the constants c, 0 are taken to be exact, whereas 0 , 0 are derived
from the relationships:

1
0

0 =

0 c2

0 =

0

= 0 c

The energy unit of electron volt (eV) is dened to be the work done by an electron
in moving across a voltage of one volt, that is, 1 eV = 1.602 176 462 1019 C 1 V, or
1 eV = 1.602 176 462 1019 J

B. Electromagnetic Frequency Bands

1043

In units of eV/Hz, Plancks constant h is:

h = 4.135 667 27 1015 eV/Hz = 1 eV/241.8 THz

that is, 1 eV corresponds to a frequency of 241.8 THz, or a wavelength of 1.24 m.

B. Electromagnetic Frequency Bands

The ITU divides the radio frequency (RF) spectrum into the following frequency and
wavelength bands in the range from 30 Hz to 3000 GHz:
RF Spectrum
band designations
ELF
VF
VLF
LF
MF
HF
VHF
UHF
SHF
EHF

Extremely Low Frequency

Voice Frequency
Very Low Frequency
Low Frequency
Medium Frequency
High Frequency
Very High Frequency
Ultra High Frequency
Super High Frequency
Extremely High Frequency
Submillimeter

frequency
30300
3003000
330
30300
3003000
330
30300
3003000
330
30300
300-3000

An alternative subdivision of the low-frequency

bands is to designate the bands 330 Hz, 30300 Hz,
and 3003000 Hz as extremely low frequency (ELF),
super low frequency (SLF), and ultra low frequency
(ULF), respectively.
Microwaves span the 300 MHz300 GHz frequency range. Typical microwave and satellite communication systems and radar use the 130 GHz
band. The 30300 GHz EHF band is also referred to
as the millimeter band.
The 1100 GHz range is subdivided further into
the subbands shown on the right.

Telecommunication Union.

23. Appendices

The SHF microwave band is used in radar (trafc control, surveillance, tracking, missile guidance, mapping, weather), satellite communications, direct-broadcast satellite
(DBS), and microwave relay systems. Multipoint multichannel (MMDS) and local multipoint (LMDS) distribution services, fall within UHF and SHF at 2.5 GHz and 30 GHz.
Industrial, scientic, and medical (ISM) bands are within the UHF and low SHF, at 900
MHz, 2.4 GHz, and 5.8 GHz. Radio astronomy occupies several bands, from UHF to LW
microwave bands.
Beyond RF, come the infrared (IR), visible, ultraviolet (UV), X-ray, and -ray bands.
The IR range extends over 3300 THz, or 1100 m. Many IR applications fall in the
120 m band. For example, optical ber communications typically use laser light at
1.55 m or 193 THz because of the low ber losses at that frequency. The UV range lies
beyond the visible band, extending typically over 10400 nm.

wavelength

Hz
Hz
kHz
kHz
kHz
MHz
MHz
MHz
GHz
GHz
GHz

110
1001000
10100
110
1001000
10100
110
10100
110
110
1001000

Mm
km
km
km
m
m
m
cm
cm
mm
m

band
infrared
ultraviolet
X-Ray
-ray

L
S
C
X
Ku
K
Ka
V
W

GHz
GHz
GHz
GHz
GHz
GHz
GHz
GHz
GHz

frequency

energy

1001 m
40010 nm
10 nm100 pm
< 100 pm

3300 THz
750 THz30 PHz
30 PHz3 EHz
> 3 EHz

0.124124 keV
> 124 keV

Visible Spectrum

frequency
12
24
48
812
1218
1827
2740
4075
80100

wavelength

The CIE denes the visible spectrum to be the wavelength range 380780 nm, or
385789 THz. Colors fall within the following typical wavelength/frequency ranges:

Microwave Bands
band

Some typical RF applications are as follows. AM radio is broadcast at 5351700

kHz falling within the MF band. The HF band is used in short-wave radio, navigation,
amateur, and CB bands. FM radio at 88108 MHz, ordinary TV, police, walkie-talkies,
and remote control occupy the VHF band.
Cell phones, personal communication systems (PCS), pagers, cordless phones, global
positioning systems (GPS), RF identication systems (RFID), UHF-TV channels, microwave
ovens, and long-range surveillance radar fall within the UHF band.
International

1044

color

wavelength

red
orange
yellow
green
blue
violet

780620
620600
600580
580490
490450
450380

nm
nm
nm
nm
nm
nm

frequency
385484
484500
500517
517612
612667
667789

THz
THz
THz
THz
THz
THz

X-ray frequencies fall in the PHz (petahertz) range and -ray frequencies in the EHz
(exahertz) range. X-rays and -rays are best described in terms of their energy, which is
related to frequency through Plancks relationship, E = hf . X-rays have typical energies
of the order of keV, and -rays, of the order of MeV and beyond. By comparison, photons
in the visible spectrum have energies of a couple of eV.
The earths atmosphere is mostly opaque to electromagnetic radiation, except for
three signicant windows, the visible, the infrared, and the radio windows. These
three bands span the wavelength ranges of 380-780 nm, 1-12 m, and 5 mm20 m,
respectively.
Within the 1-10 m infrared band there are some narrow transparent windows. For
the rest of the IR range (11000 m), water and carbon dioxide molecules absorb infrared
radiationthis is responsible for the Greenhouse effect. There are also some minor
transparent windows for 1740 and 330370 m.
Commission
1

Internationale de lEclairage (International Commission on Illumination.)

THz = 1012 Hz, 1 PHz = 1015 Hz (petahertz), 1 EHz = 1018 Hz (exahertz).

C. Vector Identities and Integral Theorems

1045

Beyond the visible band, ultraviolet and X-ray radiation are absorbed by ozone and
molecular oxygen (except for the ozone holes.)

C. Vector Identities and Integral Theorems

Algebraic Identities
2

|A| |B| = |A B| + |A B|

(C.1)

(A B)C = (B C)A = (C A)B

(C.2)

A (B C) = B (A C)C (A B)

(BAC-CAB rule)

(A B)(C ) = (A C)(B )(A )(B C)





(A B)(C ) = (A B) C (A B)C

1046

23. Appendices

Differential Identities
) = 0
(

(C.12)

A) = 0
(

(C.13)

A
(A) = A +

(C.14)

A + A
(A) =

(C.15)

B)+B (
A)
(A B) = (A )B + (B )A + A (
A)A (
B)
(A B) = B (
B)B(
A)+(B )A (A )B
(A B) = A(

(C.18)

(C.4)

A) = (
A) A
(

(C.19)

(C.6)

is any unit vector, and A , A are the components of A perpendicular and

where n
. Note also that n
(A n
)= (n
A)n
. A three-dimensional vector can
parallel to n
equally well be represented as a column vector:

ax

a = ay
bz

B)
Ax Bx + Ay By + Az Bz = (A )B + A (

(C.20)

A)
Bx Ax + By Ay + Bz Az = (B )A + B (

(C.21)

A)+(n
A)
)A = n
(
)A n
(
(n

(C.22)

(C.7)





)E E (n
)= (n
)(E)+ n
(E) n
(E)
(n


n
E)(n
E (n
E)
(
E)
+ n

Consequently, the dot and cross products may be represented in matrix form:

bx

a b = [ax , ay , az ] by = ax bx + ay by + az bz
bz

bx
ay bz az by
0
az
ay

0
ax by = az bx ax bz
Ab = az
ay
ax
0
bz
ax by ay bx

ab

ab

(C.8)

A2 = aaT (aTa)I

2 ,
n
T N
I=n

where

x
n

y ,
=n
n
z
n

=
N

0
z
n
y
n

0
x
n


x2 + y2 + z2 , and the unit vector
r = r/r , we have:

r2

r = 3,

r = 0,

r=

(C.24)

The theorems involve a volume V surrounded by a closed surface S. The divergence or

Gauss theorem is:


V

y
n

x ,
n

(C.23)

Integral Theorems for Closed Surfaces

(C.10)

z
n

r 2 = 2r ,

(C.9)

, this identity reads:

where I is the 33 identity matrix. Applied to a unit vector n

+yy
+ z
z, r = |r| =
With r = x x

r =
r,

The cross-product matrix A satises the following identity:

(C.17)

(C.3)

(C.5)

(A n
)+(n
A)n
= A + A
A = n

+ ay y
+ az
z
a = ax x

(C.16)

Tn
= 1 (C.11)
n

This corresponds to the matrix form of the parallel/transverse decomposition (C.6).

N
a)= N
2a .
(n
Ta) and a = (n
a)n
= n
(n
a)= N(
Indeed, we have a  = n
2 )a = a  + a .
n
T N
Therefore, a = Ia = (n


A dV =

dS
An

(Gauss divergence theorem)

(C.25)

is the outward normal to the surface. Greens rst and second identities are:
where n


2 + dV =

dS
n
 





dS
2 2 dV =

n
n
V
S
V

(C.26)

(C.27)

C. Vector Identities and Integral Theorems

where

1047

. Some related theorems are:

is the directional derivative along n
=n
n




dS =
dS
2 dV =
(C.28)
n
V
S
S n


dS
dV =
n
(C.29)
V


V

2 A dV =

)A dS =
(n

)A dS =
(n

(C.30)

(C.31)

A dV =

A dS
n

(C.32)


E

E
dS =
n
n



n
E)(n
E (n
E)
(
E) dS
=
n


S

dS =
n

A B A B) dV =
(

(B A A B) dV =

 

(A B) dS
n

(C.34)

(A B B A) dS
n

(C.35)

(C.40)

(C.41)

(C.42)

A dS =
n

A dl

(Stokes theorem)

(C.36)


dS =
A (n
A)
n


C




dS =
) n
A (n
A)
(

)A dl
(

dS =
n

dl
C

g(z)=

ejkr
4r

(D.2)

ej|z|
2j

(D.3)

3xi xj r 2 ij

g(r)
i, j = 1, 2, 3
(D.4)
r4
where i = /xi and xi stands for any of x, y, z. By summing the i, j indices, Eq. (D.4)
reduces to (D.1). Using this identity, we nd for the Greens function G(r)= ejkr /4r :



1
1  3xi xj r 2 ij
2 xi xj
i j G(r)= ij (3) (r)+ jk +

k
(D.5)
G(r)
3
r
r3
r2




 2

1  3
r(
r p)p
p G(r) = p (3) (r)+ jk +
+ k2
r (p
r) G(r) (D.6)
2
3
r
r

(C.38)

The second term on the right is simply the left-hand side evaluated at points away
from the origin, thus, we may write:


C

G(r)=

(D.1)

(C.37)


S

A d l

1
4r

This reduces to Eq. (D.2) upon summing the indices. For any xed vector p, Eq. (D.5)
is equivalent to the vectorial identity:

where dl is the tangential path length around C. Some related theorems are:

g(r)=

where r = |r|. Eqs. (D.2) and (D.3) are appropriate for describing outgoing waves. We
considered other versions of (D.3) in Sec. 21.3. A more general identity satised by the
Greens function g(r) of Eq. (D.1) is as follows (for a proof, see Refs. [134,135]):
1
3

r dl


2z + 2 g(z)= (z)

Stokes theorem involves an open surface S and its boundary contour C:

dl A

D. Greens Functions

Integral Theorems for Open Surfaces



E

dS +
E d l =
E
n
n
C



n
E)(n
E (n
E)
(
E) dS
=
n

i j g(r)= ij (3) (r)+

 2

+ k2 G(r)= (3) (r)

(C.33)

1
2

2 g(r)= (3) (r)

The vectorial forms of Greens identities are [1243,1240]:

A)+(n
A) dS =
(
)A n
(
n

The Greens functions for the Laplace, Helmholtz, and one-dimensional Helmholtz equations are listed below:

Using Eqs. (C.23) and (C.31), we nd:


)A dS =
(n

A)+(n
A) dS = 0
(
)A n
(
n

Eq. (C.41) is a special case of (C.40). Using Eqs. (C.23) and (C.40) we nd:

A
dS
n


 

23. Appendices


S

1048

(C.39)




 2


p G(r) = p (3) (r) + p G(r)
3

r=0

(D.7)

D. Greens Functions

1049

Then, Eq. (D.7) implies the following integrated identity, where is with respect to r :

 



2
P(r )G(r r )
P(r )G(r r ) dV = P(r)+

3

r =r

dV (D.8)

and r is assumed to lie within V. If r is outside V, then the term 2P(r)/3 is absent.
Technically, the integrals in (D.8) are principal-value integrals, that is, the limits as
0 of the integrals over VV (r), where V (r) is an excluded small sphere of radius
centered about r. The 2P(r)/3 term has a different form if the excluded volume V (r)
has shape other than a sphere or a cube. See Refs. [1282,486,498,624] and [129133]
for the denitions and properties of such principal value integrals.
Another useful result is the so-called Weyl representation or plane-wave-spectrum
representation [22,26,1282,27,541] of the outgoing Helmholtz Greens function G(r):
jkr

G(r)=

4r

 

j(kx x+ky y) jkz |z|

2jkz

dkx dky
(2)2

(D.9)

1050

23. Appendices

kx x + ky y = k cos( ). Setting dx dy = d d = r dr d
, the latter following
from r 2 = 2 + z2 , we obtain from Eq. (D.11) after replacing = r 2 z2 :


  jkr
ejkr j(kx x+ky y)
e
dx dy =
e
ejk cos() r dr d
4r
4r

 2

 

1
d jk cos()
1
=
e
dr ejkr
=
dr ejkr J0 k r 2 z2
2 |z|
2

2
|z|
0

g(kx , ky , z) =

where we used the integral representation (17.9.2) of the Bessel function J0 (x). Looking
up the last integral in the table of integrals [1402], we nd:

g(kx , ky , z)=


k2 k2 ,

kz =
j k2 k2 ,

if

k k ,

(propagating modes)

if

k > k ,

(evanescent modes)

 

G(x, y, z) =

 

G(x, y, z)ej(kx x+ky y) dx dy =

 

j(kx x+ky y)

g(kx , ky , z)e

ejkz |z|
2jkz

(x)(y)=

ej(kx x+ky y)

(D.11)

dkx dky
(2)2

dkx dky
,
(2)2


2z + k2z g(kx , ky , z)= (z)

ejkz |zz |
2jkz

(D.14)


0

ejkz z

ejkz |zz |
2jkz

ejkz z
ejkz z

kz2 k2z 2kz (kz kz ) ,

dz =

ejkz z

,
2kz (kz + kz )

for

z0
(D.15)

for

z<0

The proof is obtained by splitting the integral over the sub-intervals [0, z] and
[z, ). To handle the limits at innity, kz must be assumed to be slightly lossy, that is,
kz = z jz , with z > 0. Eqs. (D.14) and (D.15) can be combined into:

ej k r
ej kr


2
2
k k
2kz (kz kz )
 
ej k r G(r r ) dV =

V+
ej k r

,
2kz (kz + kz )

for z 0
(D.16)
for z < 0

where V+ is the half-space z 0, and k, k , k are wave-vectors with the same kx , ky

components, but different kz s:
+ ky y
+ kz
z
k = kx x

we nd from Eq. (D.2) that g(kx , ky , z) must satisfy the one-dimensional Helmholtz
Greens function equation (D.3), with k2z = k2 k2x k2y = k2 k2 , that is,

(D.13)

One can also show the integral:

Writing (3) (r)= (x)(y)(z) and using the inverse Fourier transform:

 

 
 ejkz |z|
dr ejkr J0 k r 2 z2 =
2jkz
|z|

ej(kx x +ky y ) G(r r )dx dy = ej(kx x+ky y)

(D.10)

The propagating modes are important in radiation problems and conventional imaging systems, such as Fourier optics [1285]. The evanescent modes are important in the
new subject of near-eld optics, in which objects can be probed and imaged at nanometer
scales improving the resolution of optical microscopy by factors of ten. Some near-eld
optics references are [520540].
To prove (D.9), we consider the two-dimensional spatial Fourier transform of G(r)
and its inverse. Indicating explicitly the dependence on the coordinates x, y, z, we have:

g(kx , ky , z) =

where kz must be dened exactly as in Eq. (D.10). A direct consequence of Eq. (D.11)
and the even-ness of G(r) in r and of g(kx , ky , z) in kx , ky , is the following result:

where k2z = k2 k2 , with k = k2x + k2y . In order to correspond to either outgoing

waves or decaying evanescent waves, kz must be dened more precisely as follows:

1
2

+ ky y
kz
k = kx x
z


+ ky y
+
k = kx x

(D.17)

kz
z

(D.12)

whose outgoing/evanescent solution is g(kx , ky , z)= ejkz |z| /2jkz .

A more direct proof of (D.9) is to use cylindrical coordinates, kx = k cos , ky =
k sin , x = cos , y = sin , where k2 = k2x + k2y and 2 = x2 + y2 . It follows that

where we note that k2 k2 = (k2x + k2y + kz2 )(k2x + k2y + k2z )= kz2 k2z .
The Greens function results (D.8)(D.17) are used in the discussion of the EwaldOseen extinction theorem in Sec. 14.6.

E. Coordinate Systems

1051

A related Weyl-type representation is obtained by differentiating Eq. (D.9) with respect to z. Assuming that z 0 and interchanging differentiation and integration (and
multiplying by 2), we obtain the identity:

2
z

ejkr
4r


=

 

ejkx x ejky y ejkz z

dkx dky
,
(2)2

z0

(D.18)

This just means that the left-hand side is the two-dimensional inverse Fourier transform of ejkz z with kz given by Eq. (D.10). Replacing r by r r , and r by R = |r r |,
and noting that z = z , we also obtain:

2 
z

ejkR
4R


=

 

ejkx (xx ) ejky (yy ) ejkz (zz )

dkx dky
,
(2)2

z z (D.19)

ejkr
4r


=

 

kx jkx x jky y jkz z dkx dky

e
e
e
,
kz
(2)2

23. Appendices

Cylindrical Coordinates

1 +
+
z

z


1
1 2
2

2 =
+

+ 2

2
z2

(E.2a)

(E.2b)

1 A
Az
(E.2c)
+

z






A
A
1 Az
1 (A )
A Az +

A=
(E.2d)
z

z
z

A=

1 (A )

(3) (r r )=

This result establishes the equivalence between the Kirchhoff-Fresnel diffraction formula and the plane-wave spectrum representation as discussed in Sec. 17.17. For the
vector diffraction case, we also need the derivatives of G with respect to the transverse
coordinates x, y. Differentiating (D.9) with respect to x (or with respect to y), we have:

2
x

1052

(  )(  )(z z )

(E.2e)

Spherical Coordinates

1
1

+
+
r
r
r sin





1
1

2
+ 2
r2
2 = 2
sin
+ 2
2
r
r sin

r r
r sin 2
r
=

z0

(D.20)

(E.3a)

(E.3b)

E. Coordinate Systems
The denitions of cylindrical and spherical coordinates were given in Sec. 14.8. The
expressions of the gradient, divergence, curl, Laplacian operators, and delta functions
are given below in cartesian, cylindrical, and spherical coordinates.

A=

1 (r 2 Ar )

r2

A =
r

r


r sin

A
(sin A )
1
(E.3c)
+

r sin



(sin A )
(rA )
1 Ar
A
1
(E.3d)

r sin
r
+

r sin

Cartesian Coordinates

= x

2 =

+y
+
z
x
y
z
(3) (r r )=

2
2
2
+
+
x2
y2
z2

Ay
Az
Ax
+
+
x
y
z






Ay
Ay
Ax
Az
Ax
Az
A=x

+y

z
y
z
z
x
x
y


 x

y
z 








=
 x y z 


 Ax Ay Az 

(rA )
Ar

(r r  )(  )(  )

(E.3e)

Transformations Between Coordinate Systems

A=

(3) (r r )= (x x )(y y )(z z )

r 2 sin

A vector A can be expressed component-wise in the three coordinate systems as:

(E.1)

Ax + y
Ay +
A=x
z Az
A +
A +
=
z Az

(E.4)

A +
A
=
r Ar +
The components in one coordinate system can be expressed in terms of the components of another by using the following relationships between the unit vectors, which

F. Fresnel, Exponential, Sine, and Cosine Integrals

1053

were also given in Eqs. (14.8.1)(14.8.3):

sin
cos
=
x
cos
sin +
=
y

(E.5)

sin

r cos
z =

r sin + cos

(E.6)

=x
cos + y
sin

= x
sin + y
cos

x = cos
y = sin

sin

r=
z cos +
=
cos

z sin +

= r sin
z = r cos

cos sin + y
sin sin +
r=x
z cos
=x
cos cos + y
sin cos

z sin
= x
sin + y
cos

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

1054

23. Appendices

At x = 0, we have F(0)= 0 and F (0)= 1, so that the Taylor series approximation

is F(x) x, for small x. The asymptotic expansions of C(x), S(x), and F(x) are for
large positive x:
1j
j jx2 /2
F(x) =
+
e
2
x

C(x) =
S(x) =

(E.7)

x
C2 (x)=

(E.8)

sin

r cos
z =

A=
(x
x
y

Ax + y
Ay +
) Ax + (
) Ay + (
A =
z Az )= (
z) A z

A = cos cos Ax + sin cos Ay sin Az

A = sin Ax + cos Ay

(E.9)

A +
A )= sin Ar + cos A
A=
(
A =
r Ar +
A +
A )= cos Ar cos A
Az =
zA=
z (
r Ar +

(E.10)

The Fresnel functions C(x) and S(x) are dened by [1401]:

x
S(x)=


sin


t2 dt

x2

2t

x
dt ,

S2 (x)=

sin t

2t

dt

(F.5)

x
F(x)= C(x)jS(x)=

x2

(F.2)

(F.6)


x2


F(x)= F2


x2

(F.7)

if

0x4

if

x>4

(F.8)

% Fresnel integrals F2 (x) = C2 (x)jS2 (x)

The ordinary Fresnel integral F(x) can be computed with the help of Eq. (F.7). The
MATLAB function fcs calculates F(x) for any vector of values x by calling fcs2:
% Fresnel integrals F(x) = C(x)jS(x)

In calculating the radiation patterns of pyramidal horns, it is desired to calculate a

Fresnel diffraction integral of the type:

C(x), S(x), and F(x) are odd functions of x and have the asymptotic values:
1j
F()=
2

S(x)= S2

ejt

dt
2t

 n
 
11

x
x

jx

(an + jbn )
,

e
4 n=0
4

F2 (x)=
 n
11

4 
4
1j
jx

+
e
(c
+
jd
)
,

n
n

2
x n=0
x

F = fcs(x);

ej(/2)t dt

and if x < 0, we set F(x)= F(x)= F2 (x2 /2).

The Fresnel function F2 (x) can be evaluated numerically using Boersmas approximation [1259], which achieves a maximum error of 109 over all x. The algorithm
approximates the function F2 (x) as follows:

F2 = fcs2(x);

(F.1)

They may be combined into the complex function:

1
C()= S()= ,
2

(F.4)

where the coefcients an , bn , cn , dn are given in [1259]. Consistency with the small- and

large-x expansions of F(x) requires that a0 + jb0 = 8/ and c0 + jd0 = j/ 8. We

have implemented Eq. (F.8) with the MATLAB function fcs2:

F. Fresnel, Exponential, Sine, and Cosine Integrals


t2 dt ,

x2

x

C(x)= C2

Similarly, using Eq. (E.6) the cylindrical components A , Az can be expressed in terms
of spherical components as:

F2 (x)= C2 (x)jS2 (x)=

The dot products can be read off Eq. (E.7), resulting in:

cos t

The two types are related by, if x 0:

A=
(x
x
y

Ax + y
Ay +
) Ax + (
) Ay + (
A =
z Az )= (
z) Az

cos

1
1

cos
2
x

They are combined into the complex function:

For example, to express the spherical components A , A in terms of the cartesian

components, we proceed as follows:

x

Associated with C(x) and S(x) are the type-2 Fresnel integrals:

cos cos
sin
=
x
r sin cos +

= r sin sin + cos sin + cos

y

C(x)=

1
1
+
sin
2
x

1
(F.3)

F0 (v, )=

ejv ej(/2)

(F.9)

F. Fresnel, Exponential, Sine, and Cosine Integrals

1055

Making the variable change t = v/ , this integral can be computed in terms of

the Fresnel function F(x)= C(x)jS(x) as follows:

F0 (v, )=

ej(/2)(v

/ 2 )


F




v
v
+ F

(F.10)

where we also used the oddness of F(x). The value of Eq. (F.9) at v = 0 is:
1

F()
F0 (0, )=
F()F() = 2

(F.11)

F1 (v, )=


cos

ejv ej(/2)


1
F0 (v + 0.5, )+F0 (v 0.5, )
2

(F.13)

F1 (0, )= F0 (0.5, )=

j/(8 2 )


=

 



1
1
F
+ F

2
2

1j
2 j/(8 2 )

,
e
2

(F.15)

F1 (v, 0)=


cos


ejv d =


for small


1
F0 (v + 0.5, 0)+F0 (v 0.5, 0)
2


% diffraction integral F(v, , a), Eq. (F.18)

 c2

sin (v + 0.5)
sin (v 0.5)
4 cos(v)
=
+
=
(v + 0.5)
(v 0.5)
1 4v 2


cos

a
2

c1

ejv ej(/2)

(F.18)

For a = 0, we have:

F(v, , 0)=

ej(/2)(v

/ 2 )

 



v
v
F
c1 F
c2

(F.19)

For a = 0, we can express F(v, , a) in terms of F(v, , 0):

F(v, , a)=


1
F(v + 0.5a, , 0)+F(v 0.5a, , 0)
2

(F.20)

For a = 0 and = 0, F(v, , a) reduces to the complex sinc function:



sin (c2 c1 )v/2 j(c2 +c1 )v/2
ejvc2 ejvc1
e
= (c2 c1 )
jv
(c2 c1 )v/2

(F.21)

Stationary Phase Approximation

(F.16)

The Fresnel integrals nd also application in the stationary-phase approximation for

evaluating integrals. The approximation can be stated as follows:

For = 0, the integral F1 (v, ) reduces to the double-sinc function:

1

F = diffint(v,sigma,a,c1,c2);

F(v, 0, 0)=

Using the asymptotic expansion (F.4), we nd the expansion valid for small :

Actually, the most general syntax of diffint is as follows:

(F.14)

It can be veried easily that F0 (0.5, )= F0 (0.5, ), therefore, the value of F1 (v, )
at v = 0 will be given by:
1

% diffraction integral F1 (v, ), Eq. (F.13)

% diffraction integral F0 (v, ), Eq. (F.9)

F(v, , a)=

Writing cos(/2)= (ej/2 + ej/2 )/2, the integral F1 (v, s) can be expressed in
terms of F0 (v, ) as follows:

F1 (v, )=

% diffraction integral F0 (v, ), Eq. (F.9)

The vectors v,sigma can be entered either as rows or columns, but the result will
be a matrix of size length(v) x length(sigma). The integral F0 (v, ) can also be
calculated by the simplied call:

(F.12)

F0 = diffint(v,sigma,0);
F1 = diffint(v,sigma,1);

It evaluates the more general integral:

sin(v)

From either (F.11) or (F.12), we nd F0 (0, 0)= 2. A related integral that is also
required in the theory of horns is the following:

1

23. Appendices

F0 = diffint(v,sigma);

Eq. (F.10) assumes that = 0. If = 0, the integral (F.9) reduces to the sinc function:

F0 (v, 0)= 2

1056

(F.17)

From either Eq. (F.16) or (F.17), we nd F1 (0, 0)= 4/.

The MATLAB function diffint can be used to evaluate both Eq. (F.9) and (F.13) for
any vector of values v and any vector of positive numbers , including = 0. It calls
fcs to evaluate the diffraction integral (F.9) according to Eq. (F.10). Its usage is:


j(x)

f (x)e

dx 

2j
f (x0 )ej(x0 )
 (x0 )

(F.22)

where x0 is a stationary point of the phase (x), that is, the solution of  (x0 )= 0,
where for simplicity we assume that there is only one such point (otherwise, one has a
sum of terms like (F.22), one for each solution of  (x)= 0). Eq. (F.22) is obtained by
expanding (x) in Taylor series about the stationary point x = x0 and keeping only up
to the quadratic term:
1
2

1
2

(x) (x0 )+ (x0 )(x x0 )+  (x0 )(x x0 )2 = (x0 )+  (x0 )(x x0 )2

F. Fresnel, Exponential, Sine, and Cosine Integrals

1057

Making this approximation in the integral and assuming that f (x) is slowly varying
in the neighborhood of x0 , we may replace f (x) by its value at x0 :

j(x)

dx 

f (x)e

j (x0 )+ (x0 )(xx0 )2 /2

= f (x0 )e

j (x0 )(xx0 )2 /2

while for z 0, we have Si (z)= Si (z) and Ci (z)= Ci (z)+j. Conversely, we have
for z > 0:





E1 (jz)= Ci (z)+j Si (z)
= ln(z)+Cin (z)+j Si (z)
2

dx


 (x

ej

2
0 )(xx0 ) /2

dx =

Using F()F()

 (x0 )

eju

/2

du =

dx




F()F()
 (x0 )

y = Si(z);
y = Ci(z);
y = Cin(z);

% sine integral, Eq. (F.24)

% sine integral, Eq. (F.24)
% sine integral, Eq. (F.25)

A related integral that appears in calculating mutual and self impedances is what
may be called a Greens function integral:

h
Gi(d, z0 , h, s)=

= 2F ()= 1 + j = 2j, we obtain



2j
j (x0 )(xx0 )2 /2
e
dx =
 (x0 )

h

Several antenna calculations, such as mutual impedances and directivities, can be reduced to the exponential integral, which is dened as follows [1401]:

E1 (z)=

(exponential integral)

z
Si (z)=

sin u

(sine integral)

Ci (z)= + ln z +

cos u 1

z

1 cos u

v0 = ju0 ,
v1 = ju1 ,




E1 (jz)E1 (jz)
+
Si (z)=
= Im E1 (jz) +
2j
2
2
E1 (jz)+E1 (jz)
2


= Re E1 (jz)

dv
dz
=
v
R

 v1
v0

eu
du ,
u

or,

(F.29)



d2 + z20 sz0


u1 = k
d2 + (h z0 )2 + s(h z0 )
u0 = k

The function Gi evaluates Eq. (F.29), where z0 , s, and the resulting integral J, can be
vectors of the same dimension. Its usage is:


0

% Greens function integral, Eq. (F.29)

cos( cos ) cos

d = Si (2)sin Cin (2)cos
sin

(F.30)

Its proof is straightforward by rst changing variables to z = cos , then using

partial fraction expansion, and nally changing variables to u = (1 + z), and using
the denitions (F.24) and (F.25):
(F.25)


0

For z 0, the sine and cosine integrals are related to E1 (z) by [1401]:

Ci (z)=

where

(cosine integral)

du = + ln z Ci (z)

Another integral that appears commonly in antenna work is:

where is the Euler constant = 0.5772156649... . A related cosine integral is:

(F.24)

du

(F.28)



ejkR jksz
e
dz = sejksz0 E1 (ju0 )E1 (ju1 )
R

J = Gi(d,z0,h,s);

du
z

Cin (z)=

Gi(d, z0 , h, s)=

(F.23)

where z is a complex number with phase restricted such that |arg z| < . This range
allows pure imaginary zs. The built-in MATLAB function expint evaluates E1 (z) at an
array of zs. Related to E1 (z) are the sine and cosine integrals:

ejkR jksz
e
dz = sejksz0
R
h

et
dt
z+t

s = 1

which gives

Exponential, Sine, and Cosine Integrals


R = d2 + (z z0 )2 ,



v = jk R + s(z z0 )

eu
du = ez
u

ejkR jksz
e
dz ,
R

This integral can be reduced to the exponential integral by the change of variables:

Normally, the phase depends on a positive parameter in the form (x)= (x),
and the stationary-phase approximation is justied in the limit .

(F.27)

The MATLAB functions Si, Ci, Cin evaluate the sine and cosine integrals at any
vector of zs by using the relations (F.26) and the built-in function expint:

The last integral can be reduced to the complex Fresnel integral by the change of

variables (x x0 )= / (x0 ) u:

23. Appendices

f (x0 )e

j(x0 )

1058

=
(F.26)

cos( cos ) cos

d =
sin

1
2

1
1

 2
=

1

cos(z) cos
1
dz +
1+z
2

cos(u ) cos

cos(z) cos
dz
1 z2

1

cos(z) cos
dz =
1z

du = sin

 2
0

sin u

du cos

1
1

 2
0

cos(z) cos
dz
1+z
1 cos u

du

G. Gauss-Legendre Quadrature

1059

G. Gauss-Legendre Quadrature

f (x) dx 

N


wi f (xi )

i=1

f (x1 )

f (x2 )

= wT f (x)
f (x) dx 
wi f (xi )= [w1 , w2 , . . . , wN ]
.

..

i=1
f (xN )

wix =

ba

J=

f (x) dx = e2 e1 + ln 2 = 5.36392145

This integral can be evaluated numerically by the MATLAB code:

N = 5;
[w,x] = quadr(1,2,N);
f = exp(x) + 1./x;
J = w*f

% number of weights and nodes

% calculate weights and nodes for the interval [1, 2]
% evaluate f (x) at the node vector
% approximate integral

This produces the exact value with a 4.23107 percentage error. If the integration
interval is split in two, say, [1, 1.5] and [1.5, 2], then the second line above can be
replaced by
J. Stoer and R. Burlisch, Introduction to Numerical Analysis, Springer, NY, (1980); and, G. H. Golub and
J. H. Welsch, Calculation of Gauss Quadrature Rules, Math. Comput., 23, 221 (1969).

(G.3)


zi +

b+a

(G.4)

wi

P(z) dz =

N


wi P(zi )

(G.5)

i=1

1
(f , g)=

2
,

provided that the zi are the N roots of the Legendre polynomial PN (z).
The Legendre polynomials Pn (z) are obtained via the process of Gram-Schmidt orthogonalization of the non-orthogonal monomial basis {1, z, z2 , . . . , zn . . . }. Orthogonality is dened with respect to the following inner product over the interval [1, 1]:

As an example, consider the following function and its exact integral:

f (x)= ex +

ba

1

Gauss-Legendre quadrature over subintervals

single interval
two subintervals, [a, c] and [c, b]
three subintervals, [a, c], [c, d], and [d, b]
subintervals, [a, a+c, a+2c, . . . , a+Mc], with a + Mc = b

b+a

where the scaling of the weights follows from the scaling of the differentials dx =
dz(b a)/2, so the value of the integral (G.1) is preserved by the transformation.
Gauss-Legendre quadrature is nicely tied with the theory of orthogonal polynomials
over the interval [1, 1], which are the Legendre polynomials. For N-point quadrature,
the nodes zi , i = 1, 2, . . . , N are the N roots of the Legendre polynomial PN (z), which
all lie in the interval [1, 1]. The method is justied by the following theorem:
For any polynomial P(z) of degree at most 2N 1, the quadrature formula (G.1) is
satised exactly, that is,

where ab is an array of endpoints that dene the subintervals, for example,

ab = [a, b] ,
ab = [a, c, b] ,
ab = [a, c, d, b] ,
ab = a : c : b ,

z+

If wi and zi are the weights and nodes with respect to the interval [1, 1], then those
with respect to [a, b] can be constructed simply as follows, for i = 1, 2, . . . , N:

(G.2)

The function quadrs allows the splitting of the interval [a, b] into subintervals,
computes N weights and nodes in each subinterval, and concatenates them to form the
overall weight and node vectors w, x:

Gauss-Legendre quadrature

[w,x] = quadrs(ab,N);

ba

xi =

The function quadr returns the column vectors of weights w and nodes x, with usage:
[w,x] = quadr(a,b,N);

x=

N


% or by, [w,x] = quadrs(1:0.5:2, N);

which has a percentage error of 1.28109 . Next, we discuss the theoretical basis of
the method.
The interval [a, b] can be replaced by the standardized interval [1, 1] with the
transformation from a x b to 1 z 1:

(G.1)

where wi , xi are appropriate weights and evaluation points (nodes). This can be written
in the vectorial form:

b

23. Appendices
[w,x] = quadrs([1,1.5,2],N);

In many parts of this book it is necessary to perform numerical integration. GaussLegendre quadrature is one of the best integration methods, and we have implemented
it with the MATLAB functions quadr and quadrs. Below, we give a brief description of
the method. The integral over an interval [a, b] is approximated by a sum of the form:

b

1060

f (z)g(z)dz

(G.6)

The standard denition of the Legendre polynomials is:

Pn (z)=

1
2n n!


dn  2
(z 1)n ,
dzn

n = 0, 1, 2, . . .

(G.7)

The rst few of them are listed below:

P0 (z) = 1
P1 (z) = z



P2 (z) = (3/2) z2 (1/3)


P3 (z) = (5/2) z3 (3/5)z


P4 (z) = (35/8) z4 (6/7)z2 + (3/35)

(G.8)

G. Gauss-Legendre Quadrature

1061

They are normalized such that Pn (1)= 1 and are mutually orthogonal with respect
to (G.6), but do not have unit norm:

1
(Pn , Pm )=

Pn (z)Pm (z)dz =

2
2n + 1

nm

23. Appendices

By introducing the same scaling factors into each term of the recurrence (G.10), we
nd that the renormalized Pn (z) satisfy:

1, z, z2 , . . . , zn

n


qk zk =

ck Pk (z)

k=0

P(z)= PN (z)Q(z)+R(z)

(f1 , P0 )
P0 (z)= z
(P0 , P0 )

(G.14)

where Q(z) and R(z) are the quotient and remainder of the division by the Legendre
polynomial PN (z), and both will have order N 1. Then, the integral of P(z) can be
written in inner-product notation as follows:

zdz = 0. Then, construct P2 by:

1

(f2 , P0 )
(f2 , P1 )
P0 (z)
P1 (z)
P2 (z)= f2 (z)
(P0 , P0 )
(P1 , P1 )

where now we have (f2 , P1 )= (z , z)=

1
1

z2 dz =

z dz = 0, and

2
,
3

P(z)dz = (P, 1)= (PN Q + R, 1)= (PN Q, 1)+(R, 1)= (Q, PN )+(R, 1)

But (Q, PN )= 0 because Q(z) has order N 1 and PN (z) is orthogonal to all such
polynomials. Thus, the integral of P(z) can be expressed only in terms of the integral
of the remainder polynomial R(z), which has order N 1:

1
2

(f2 , P0 )= (z2 , 1)=

n


with the expansion coefcients calculated from ck = (Q, Pk ). This also implies that if
Q(z) has order n 1 then, it will be orthogonal to Pn (z).
Next, we turn to the proof of the basic Gauss-Legendre result (G.5). Given a polynomial P(z) of order 2N 1, we can expand it uniquely in the form:

(fn , Pk )
Pk (z)
(Pk , Pk )

1
1

P0 (z), P1 (z), P2 (z), . . . , Pn (z)

k=0

A few steps of the construction will clarify it:

where (f1 , P0 )= (z, 1)=

Q(z)=

for n = 1, 2, 3, . . . , do

P1 (z)= f1 (z)

Thus, any polynomial Q(z) of degree n can be expanded uniquely in either basis:

initialize P0 (z)= f0 (z)= 1

k=0

(G.13)

monomial basis by the orthonormalized Legendre basis:

(G.10)

The Gram-Schmidt orthogonalization process of the monomial basis fn (z)= zn is

the following order-recursive construction:

n
1

4n2 1

This relationship can be assumed to be valid also at n = 0, provided we dene

P1 (z)= 0. For each order n, the Gram-Schmidt procedure replaces the non-orthogonal




n
n+1
Pn1 (z)+
Pn+1 (z)
2n + 1
2n + 1

Pn (z)= fn (z)

n =

zPn (z)= n Pn1 (z)+n+1 Pn+1 (z) ,

(G.9)

Moreover, they satisfy the three-term recurrence relation:

zPn (z)=

1062

1
(P0 , P0 )= (1, 1)=

1

1

dz = 2

P(z)dz = (P, 1)= (R, 1)=

R(z)dz

(G.15)

The right-hand side of the integration rule (G.5) can also be expressed in terms of R(z):

Therefore,
1
2/3
= z2
P2 (z)= z
2
3
2

N


Then, normalize it such that P2 (1)= 1, and so on. For our discussion, we are going
to renormalize the Legendre polynomials to unitnorm. Because of (G.9), this amounts
to multiplying the standard Pn (z) by the factor (2n + 1)/2. Thus, we re-dene:


Pn (z)=


2n + 1 1 dn  2
(z 1)n ,
2
2n n! dzn

1
2

i=1

n = 0, 1, 2, . . .

(G.11)

N


wi PN (zi )Q(zi )+

i=1

N


wi R(zi )

(G.16)

i=1

and, because we assumed that PN (zi )= 0,

N


wi P(zi )=

i=1

N


wi R(zi )

(G.17)

i=1

Thus, combining (G.15) and (G.17), we obtain the following condition, which is equivalent to Eq. (G.5),

Thus, (G.9) becomes (Pn , Pm )= nm . In particular, we note that now

P0 (z)=

wi P(zi )=

1

(G.12)

R(z)dz =

N

i=1

wi R(zi )

(G.18)

G. Gauss-Legendre Quadrature

1063

Because R(z) is an arbitrary polynomial of degree N 1, and has only N coefcients,

this condition can be satised with a common set of N weights wi for all such R(z). If
we had not assumed initially that the zi were the zeros of PN (z), and took them to be
an arbitrary set of N distinct points in [1, 1], then (G.18) would read as

1
1

R(z)dz =

N


wi PN (zi )Q(zi )+

i=1

N


wi R(zi )

i=1

In order for this to be satised for all R(z) and all Q(z), then (G.18) must still be
satised by setting Q(z)= 0, which xes the weights wi . Therefore, the rst term in
the right-hand side must be zero for all polynomials Q(z) of degree N 1, and one can
show that this implies that PN (zi )= 0, that is, the zi must be the zeros of PN (z).
Condition (G.18) can be used to determine the weights by expanding R(z) into either
the monomial basis or the Legendre basis, that is, because R(z) has degree N 1:

R(z)=

N
1

rk zk =

k=0

N
1

ck Pk (z)

1064

23. Appendices

Because the vector c is arbitrary, we must have the condition:

Pw =

1
zki wi =

i=1

zk dz =

1 + (1)
,
k+1
k

k = 0, 1, . . . , N 1


R(z)dz = (R, 1)=

2 (R, P0 )=

w = F1 u

N
1

(G.21)

ck (PK , P0 )=

k=0

N
1

ck k0 =

2 c0

k=0

The right-hand side of (G.18) may be written as follows. Dening the NN matrix Pki = Pk (zi ), i = 1, 2 . . . , N, and k = 0, 1, . . . , N 1, and the row vector cT =
[c0 , c1 , . . . , cN1 ] of expansion coefcients, we have,
N

i=1

wi R(zi )=

N
N
1 

ck Pk (zi )wi = cT P w

k=0 i=1

Thus, (G.18) now reads, where u0 = [1, 0, 0, . . . , 0]T :

cT P w =

2 c0 =

2 P1 u0

(G.22)

zP1 (z) = 1 P0 (z)+2 P2 (z)

zP2 (z) = 2 P1 (z)+3 P3 (z)
zP3 (z) = 3 P2 (z)+4 P4 (z)

0
P0 (z)
P (z)
1
1
=
z
P2 (z) 0
0
P3 (z)

(G.20)

Alternatively, we may use the Legendre basis, which is more elegant. The left hand
side of (G.18) will receive contribution only from the k = 0 term because P0 is orthogonal
to all the succeeding Pk . Indeed, using the denition (G.12), we have:

1

which can be written in matrix form:

Dening the matrix Fki = zki and the vector uk = 1 + (1)k /(k + 1), we may write
(G.20) in the compact matrix form:

Fw = u

w=

zP0 (z) = 1 P1 (z)

Inserting, for example, the monomial basis into (G.18) and matching the coefcients
of rk on either side, we obtain the system of N equations for the weights:
N


2 u0

The matrix P has some rather interesting properties. First, it has mutually orthogonal
columns. Second, these columns are the eigenvectors of a Hermitian tridiagonal matrix
whose eigenvalues are the zeros zi . Thus, the problem of nding both zi and wi is
reduced to an eigenvalue problem.
These eigenvalue properties follow from the recursion (G.13) of the normalized Legendre polynomials. For n = 0, 1, 2, 3, the recursion reads explicitly:

(G.19)

k=0

2 cT u0

0
P0 (z)

0
P1 (z)

0
3 P2 (z)
0
4 P4 (z)
P3 (z)

and more generally,

P0 (z)
0
P (z)
1

P2 (z) 0

= .
z
..
..

PN2 (z) 0
0
PN1 (z)

0
..
.

..

..

0
0

N2

N1

..

0
0
.

0
P0 (z)

P (z)
0
1

P2 (z)

..
..

.
.

0
N1
PN2 (z)
0
N PN (z)
PN1 (z)
0
0
0
..
.

Now, if z is replaced by the ith zero zi of PN (z), the last column will vanish and we
obtain the eigenvalue equation:

.
..

0
0

0
..
.

..

..

0
0

N2

N1

..

0
0
.

0
0
0
..
.

P0 (zi )
P1 (zi )
P2 (zi )

P0 (zi )
P1 (zi )
P2 (zi )

= zi

.
.

..
..

PN2 (zi )
N1 PN2 (zi )
0
PN1 (zi )
PN1 (zi )

(G.23)

Denoting the above tridiagonal matrix by A and the column of Pk (zi )s by pi , we

may write compactly:
A pi = zi pi , i = 1, 2, . . . , N
(G.24)

H. Lorentz Transformations

1065

Thus, the eigenvalues of A are the zeros zi and the corresponding eigenvectors are
the columns pi of the matrix P that we introduced in (G.22). Because the zeros zi are
distinct and A is a Hermitian matrix, its eigenvectors will be mutually orthogonal:
2
pT
i pj = di ij

V = PD1

(G.26)

Replacing P in (G.22) by P = VD and using the orthogonality VT V = I of the eigenvector matrix, or V1 = VT , we obtain the solution:
w=

2D

V u0

wi =

1
T
2 d
i (vi u0 )

(G.27)

The matrix D can itself be expressed in terms of V by noting that the top entry of pi

is P0 (zi )= 1/ 2, and therefore, it follows from vi = pi /di that the top entry of vi will

T
1
T
be vi u0 = 1/( 2di ), or, d
i = 2(vi u0 ). It nally follows from Eq. (G.27) that

wi =

2
d
i

23. Appendices

where c is the speed of light in vacuum. Dening the scaled quantities = ct and
= v/c, the above transformation and its inverse, obtained by replacing by , may
be written as follows:

 = ( z)
z = (z )
x = x
y = y

(G.25)

where di = pi  are the norms of the vectors pi . It follows that the orthonormalized
eigenvectors of A will be vi = pi /di , and the orthogonal matrix of eigenvectors having
the vi as columns will be V = [v1 , v2 , . . . , vN ], or, expressed in terms of the matrix P
and the diagonal matrix D = diag{d1 , d2 , . . . , dN }:

1066

2
2(vT
i u0 )

= V(1, i), that is, the rst row of V. Because the eigenIn MATLAB language,
vectors of the Hermitian matrix A are real-valued and unique up to a sign, Eq. (G.28)
allows the unique determination of the weights from the eigenvector matrix V.
The above discussion leads to two possible implementations of the MATLAB function
quadr. In the rst, we obtain the coefcients of the Legendre polynomial PN (z), nd its
zeros using the built-in function root, and then solve the linear equation (G.21) for the
weights. The second approach, implemented by the function quadr2 and the related
function quadrs2, determines zi , wi from the eigenvalue problem of the matrix A.

x = Lx ,

x

where x = ,
y
z

v 
t = t 2 z
c
z = (z vt)


x =x


y =y

1
where =
1 v2 /c2

(H.1)


x

x =  ,
y
z

L=

0
0

0
1
0
0

0
0
1
0

0
0

(H.2)

Such transformations leave the quadratic form (c2 t2 x2 y2 z2 ) invariant, that is,

c2 t2 x2 y2 z2 = c2 t2 x2 y2 z2

(H.3)

Introducing the diagonal metric matrix G = diag(1, 1, 1, 1), we may write the
quadratic form as follows, where xT denotes the transposed vector, that is, the row
vector xT = [, x, y, z]:
xT Gx = 2 x2 y2 z2 = c2 t2 x2 y2 z2

(H.4)

More generally, a Lorentz transformation is dened as any linear transformation x =

Lx that leaves the quadratic form xT Gx invariant. The invariance condition requires
that: xT Gx = xT LT GLx = xT Gx, or

LT GL = G

H. Lorentz Transformations
According to Einsteins special theory of relativity [461], Lorentz transformations describe the transformation between the space-time coordinates of two coordinate systems moving relative to each other at constant velocity. Maxwells equations remain
invariant under Lorentz transformations. This is demonstrated below.
Let the two coordinate frames be S and S . By convention, we may think of S as
the xed laboratory frame with respect to which the frame S is moving at a constant
velocity v. For example, if v is in the z-direction, the space-time coordinates {t, x, y, z}
of S are related to the coordinates {t , x , y , z } of S by the Lorentz transformation:

These transformations are also referred to as Lorentz boosts to indicate the fact that
one frame is boosted to move relative to the other. Interchanging the roles of z and x, or
z and y, one obtains the Lorentz transformations for motion along the x or y directions,
respectively. Eqs. (H.1) may be expressed more compactly in matrix form:

(G.28)

vT
i u0

= ( + z )
z = (z +  )
x = x
y = y

(H.5)

In addition to the Lorentz boosts of Eq. (H.1), the more general transformations
satisfying (H.5) include rotations of the three spatial coordinates, as well as time or
space reections. For example, a rotation has the form:

L=
0

0
where R is a 33 orthogonal rotation matrix, that is, RT R = I, where I is the 33
identity matrix. The most general Lorentz boost corresponding to arbitrary velocity
v = [vx , vy , vz ]T is given by:

L=

I+

2
T
+1

where =

= 

1 T

(H.6)

H. Lorentz Transformations

1067

| =
When v = [0, 0, v]T , or = [0, 0, ]T , Eq. (H.6) reduces to (H.1). Dening = |

2 2
2

T
and the unit vector = /, and using the relationship = 1, it can be
veried that the spatial part of the matrix L can be written in the form:
I+

1 + 2
1 + 1 2

v=

23. Appendices

Ta and a = [ax , ay , az ]T is the spatial part of a. Then,

where a =
T

a =
(
a)
a =

(H.7)

a0

a

0
0

0
1
0
0

0
0
1
0

0
0

1
0
0

1 1

0
1
0
0

1 1

0
0
1
0

0
0

2
0
0

2 2

(H.8)

a = La ,

0
1
0
0

0
0
1
0

a0
a

a = x
ay
az

a
a

= (a a0 )
= a

= 

| ,
= |

2 2
0
0

a0 b0 a b = a0 b0 a b ,

where a =

a0

b=

b0

time and space

ct

frequency and wavenumber

/c

kx

ky

kz

energy and momentum

E/c

px

py

pz

charge and current densities

Jx

Jy

Jz

scalar and vector potentials

cAx

cAy

cAz

(H.12)

(H.13)

For example, under the z-directed boost of Eq. (H.1), the frequency-wavenumber
transformation will be as follows:

 = ( ckz )


kz = kz
c
kx = kx

= ( + ckz )


kz = kz + 
c
kx = kx

c = v ,

= 2
c
c

(H.14)

ky = ky

where we rewrote the rst equations in terms of instead of /c. The change in
frequency due to motion is the basis of the Doppler effect. The invariance property
(H.12) applied to the space-time and frequency-wavenumber four-vectors reads:

 t k r = t k r

(H.15)

This implies that a uniform plane wave remains a uniform plane wave in all reference
frames moving at a constant velocity relative to each other. Similarly, the charge and
current densities transform as follows:

c = (c Jz )
Jz = (Jz c)

(H.11)

(H.10)

Some examples of four-vectors are given in the following table:

ky = ky

(a0 a )
a +
(a a0 )
a

az

Four-vectors transforming according to Eq. (H.9) are referred to as contravariant.

Under the general Lorentz boost of Eq. (H.6), the spatial components of a that are transverse to the direction of the velocity vector v remain unchanged, whereas the parallel
component transforms as in Eq. (H.10), that is, the most general Lorentz boost transformation for a four-vector takes the form:

a0 = (a0 a )

ay

(H.9)

a0 = (a0 + az )
az = (az + a0 )
ax = ax
ay = ay

a0

ax

For example, under the z-directed boost of Eq. (H.1), the four-vector a will transform as:

a0 = (a0 az )
az = (az a0 )
ax = ax
ay = ay

T
I + ( 1)

a0

a0
a
x
,
where a =
ay
az

four-vector

A four-vector is a four-dimensional vector that transforms like the vector x under

Lorentz transformations, that is, its components with respect to the two moving frames
S and S are related by:

from which Eq. (H.11) follows.

For any two four-vectors a, b, the quadratic form aTGb remains invariant under
Lorentz transformations, that is, aTGb = aTGb, or,

the following condition, where 1 = 1/ 1 21 and 2 = 1/ 1 22 :

v1 + v2
1 + v1 v2 /c2

with = v/c. Eq. (H.8) is Einsteins relativistic velocity addition theorem. The same
group property implies also that L1 ()=
 L(). The proof
 of Eq. (H.8) follows from

a
and a = a a  = a

and using Eq. (H.7), the Lorentz transformation (H.6) gives:

Setting =

T
T = I + ( 1)
+1

The set of matrices L satisfying Eq. (H.5) forms a group called the Lorentz group. In
particular, the z-directed boosts of Eq. (H.2) form a commutative subgroup. Denoting
these boosts by L(), the application of two successive boosts by velocity factors 1 =
v1 /c and 2 = v2 /c leads to the combined boost L()= L(1 )L(2 ), where:

1068

Jx

= Jx

Jy = Jy

c = (c + Jz )


Jz = (Jz + c )
Jx = Jx
Jy = Jy

(H.16)

H. Lorentz Transformations

1069

Because Eq. (H.5) implies that LT = GLG, we are led to dene four-vectors that
transform according to LT . Such four-vectors are referred to as being covariant. Given
= Ga. This operation
any contravariant 4-vector a, we dene its covariant version by a
simply reverses the sign of the spatial part of a:


= Ga =
a

1
0



a0


=

a0
a

1070

23. Appendices

A rank-2 tensor is represented by a 44 matrix, say F. Its Lorentz transformation

properties are the same as the transformation of the product of a column and a row
four-vector, that is, F transforms like the quantity abT , where a, b are column fourvectors. This product transforms like a bT = L(abT )LT . Thus, a general second-rank
tensor transforms as follows:

F = LFLT

(H.17)

An antisymmetric rank-2 tensor F denes, and is completely dened by, two threedimensional vectors, say a = [ax , ay , az ]T and b = [bx , by , bz ]T . Its matrix form is:

transforms as follows:
The vector a
 = Ga = GLa = (GLG)(Ga)= LT a

(H.18)




x
=
x =
y

z
Because x = Lx, it follows that

x

=L

xj




=
Lji 
 =
xi

xj
i
j
j
j

(H.19)

x = LT x

 = ( + z )

= ( z )
z = (z  )

x = x
y = y

x = x

(H.20)

The four-dimensional divergence of a four-vector is a Lorentz scalar. For example,

denoting the current density four-vector by J = [c, Jx , Jy , Jz ]T , the charge conservation law involves the four-dimensional divergence:

bx

(H.24)

bx

b
x
=

F
by
bz

bx
0

az
ay

bz
ay

ax

ax

by
az

(H.25)

transforms like F itself. For the z-directed boost of Eq. (H.1), it

Thus, the dual F
follows from (H.23) that the two vectors a, b transform as follows:

(H.21)

(H.22)

Although many quantities in electromagnetism transform like four-vectors, such as

the space-time or the frequency-wavenumber vectors, the actual electromagnetic elds
do not. Rather, they transform like six-vectors or rank-2 antisymmetric tensors.

ax = (ax by )

bx = (bx + ay )

ay = (ay + bx )

by = (by ax )

az

bz = bz

= az

(H.27)

These are obtained by equating the expressions:

Under a Lorentz transformation, this remains invariant, and therefore, if it is zero

T
in one frame it will remain zero in all frames. Using T
x = x L, we have:
T




t + J = T
x J = x LJ = x J = t + J

az
by

,
bx

ay
bz

bz
by

y = y

c
J
x
= T
t + J = [ , x , y , z ]
xJ
Jy
Jz

bz
by

az
by

bx

ay
bz

corresponds to the pair (a, b), and F

to (b, a). Their Lorentz transforThus, F
mation properties are:
 = LT FL
1 , F
 = LFL
T
F
(H.26)

For the z-directed boost of Eq. (H.1), we have LT = L1 , which gives:

z = (z + )

ax

a
x
=

F
ay
az

x = LT x

ax

= GFG, and its

Given the tensor F, one may dene its covariant version through F
and obtained by the replacements a b and b a, that is,
dual, denoted by F

x . Indeed, we have component-wise:

a
x
F=
ay
az

where we used the property that G2 = I4 , the 44 identity matrix. The most important
covariant vector is the four-dimensional gradient:

(H.23)

ax

a
0
x

ay
bz
az by

0
1

=
0
0
0

ay
bz
0

bx
0
0
1
0

az
by

=
bx
0

0
ax

0 ay

az

ax
0

bz
by

ay
bz
0

bx

az

by
0

bx 0
0

0
1
0
0

0
0
1
0

0
0

H. Lorentz Transformations

1071

More generally, under the boost transformation (H.6), it can be veried that the
components of a, b parallel and perpendicular to v transform as follows:
a = (a + b )
b = (b a )

= 

1 2

a  = a 

| ,
= |

(H.28)

1072

23. Appendices

Associated with a six-vector (a, b), there are two scalar invariants: the quantities
(a b) and (a a b b). Their invariance follows from Eq. (H.28). Thus, the scalars
(E B), (E E c2 B B), ( H), (c2 H H) remain invariant under Lorentz
transformations. In addition, it follows from (H.30) that the quantity (E B H) is
invariant.
Given a six-vector (a, b) and its dual (b, a), we may dene the following fourdimensional current vectors that are dual to each other:

b  = b 

J=

Thus, in contrast to Eq. (H.11) for a four-vector, the parallel components remain unchanged while the transverse components change. A pair of three-dimensional vectors
(a, b) transforming like Eq. (H.28) is referred to as a six-vector.
It is evident also that Eqs. (H.28) remain invariant under the duality transformation
a b and b a, which justies Eq. (H.26). Some examples of (a, b) six-vector pairs
dening an antisymmetric rank-2 tensor are as follows:
a

cB

c
cP

H
M

(H.29)

B = (B

E )

)
H = (H c
 = ( +

E  = E 

H  = H 

B  = B 

 = 

H )

Hx = (Hx + cDy )

Ey = (Ey + cBx )

Hy = (Hy cDx )

By = (By

c
c

Ey )

Dx = (Dx

Ex )

Dy = (Dy +

Ez = Ez

Hz = Hz

Bz = Bz

Dz = Dz

b
a b


(H.32)




J =

 a 

b   a 


J =

 b 

a   b 


(H.33)

The calculation is straightforward but tedious. For example, for the z-directed boost
(H.1), we may use Eqs. (H.20) and (H.27) and the identity 2 (1 2 )= 1 to show:



Jx =  b   a  x = y bz z by  ax
= y bz 2 (z + )(by ax )2 ( + z )(ax by )


= y bz z by ax = b a x = Jx

J0 =  a  = x ax + y ay + z az

(H.30)

= x (ax by )+y (ay + bx )+(z + )az



= (x ax + y ay + z az )(x by y bx az ) = (J0 Jz )

Ex = (Ex cBy )

J =

Similarly, we have:

= v and /c = v/c2 . Note that the two groups of equations

where we may replace c
transform into each other under the usual duality transformations: E H, H E,
B, B . For the z-directed boost of Eq. (H.1), we have from Eq. (H.30):

Bx = (Bx +

It can be shown that both J and J transform as four-vectors under Lorentz transformations, that is, J = LJ and J = LJ, where J , J are dened with respect to the
coordinates of the S frame:

where P, M are the polarization and magnetization densities dened through the relationships = 0 E + P and B = 0 (H + M). Thus, the (E, B) and (, H) elds have the
following Lorentz transformation properties:

B )
E = (E + c

a
b a

c
c

Hy )
(H.31)

Hx )

In this fashion, one can show that J and J satisfy the Lorentz transformation equations (H.10) for a four-vector. To see the signicance of this result, we rewrite Maxwells
equations, with magnetic charge and current densities m , Jm included, in the fourdimensional forms:

c
H c


=


,

cB
E cB


=

cm


(H.34)

Jm

Thus, applying the above result to the six-vector (c, H) and to the dual of (E, cB)
and assuming that the electric and magnetic current densities transform like fourvectors, it follows that Maxwells equations remain invariant under Lorentz transformations, that is, they retain their form in the moving system:

 c 

H   c 


=

c
J


,

 cB 

E   cB 


=

cm

Jm


(H.35)

I. MATLAB Functions

1073

The Lorentz transformation properties of the electromagnetic elds allow one to

solve problems involving moving media, such as the Doppler effect, reection and transmission from moving boundaries, and so on. The main technique for solving such problems is to transform to the frame (here, S ) in which the boundary is at rest, solve the
reection problem in that frame, and transform the results back to the laboratory frame
by using the inverse of Eq. (H.30).
This procedure was discussed by Einstein in his 1905 paper on special relativity in
connection to the Doppler effect from a moving mirror. To quote [461]: All problems
in the optics of moving bodies can be solved by the method here employed. What is
essential is that the electric and magnetic force of the light which is inuenced by a
moving body, be transformed into a system of co-ordinates at rest relatively to the
body. By this means all problems in the optics of moving bodies will be reduced to a
series of problems in the optics of stationary bodies.

I. MATLAB Functions

1074

23. Appendices

Plasmonic Waveguides
drude
dmda
dmds
dmdcut
pwg
pwga
pwgpower

Drude-Lorentz model for Silver, Gold, Copper, Aluminum

asymmetric DMD plasmonic waveguide - iterative solution
symmetric DMD plasmonic waveguide - iterative solution
cutoff width for asymmetric DMD guides
plasmonic waveguide solution for symmetric guides
plasmonic waveguide solution for asymmetric guides
transmitted power in plasmonic waveguide

Transmission Lines
g2z
z2g
lmin

- reflection coefficient to impedance transformation

- impedance to reflection coefficient transformation
- find locations of voltage minima and maxima

mstripa
mstripr
mstrips

- microstrip analysis (calculates Z,eff from w/h)

- microstrip synthesis with refinement (calculates w/h from Z)
- microstrip synthesis (calculates w/h from Z)

multiline - reflection response of multi-segment transmission line

The MATLAB functions are grouped by category. They are available from the web page:
www.ece.rutgers.edu/~orfanidi/ewa.
Multilayer Dielectric Structures
brewster
fresnel

- calculates Brewster and critical angles

- Fresnel reflection coefficients for isotropic or birefringent media

n2r
r2n

- refractive indices to reflection coefficients of M-layer structure

- reflection coefficients to refractive indices of M-layer structure

multidiel
multidiel1
multidiel2
omniband
omniband2

snel

- calculates refraction angles from Snels law for birefringent media

reflection response of isotropic or birefringent multilayer structures

simplified version of multidiel for isotropic layers
reflection response of lossy isotropic multilayer dielectric structures
bandwidth of omnidirectional mirrors and Brewster polarizers
bandwidth of birefringent multilayer mirrors

swr
tsection

- standing wave ratio

- T-section equivalent of a length-l transmission line segment

gprop
vprop
zprop

- reflection coefficient propagation

- wave impedance propagation
- wave impedance propagation

Impedance Matching
qwt1
qwt2
qwt3

- quarter wavelength transformer with series segment

- quarter wavelength transformer with 1/8-wavelength shunt stub
- quarter wavelength transformer with shunt stub of adjustable length

dualband - two-section dual-band Chebyshev impedance transformer

dualbw
- two-section dual-band transformer bandwidths
stub1
stub2
stub3

- single-stub matching
- double-stub matching
- triple-stub matching

onesect
twosect

- one-section impedance transformer

- two-section impedance transformer

pi2t
t2pi
lmatch
pmatch

Pi to T transformation
Pi to T transformation
L-section reactive conjugate matching network
Pi-section reactive conjugate matching network

input reflection coefficient in terms of S-parameters

output reflection coefficient in terms of S-parameters
constant noise figure circle
noise figure of two-port
transducer, available, and operating power gains of two-port

Quarter-Wavelength Transformers
bkwrec
frwrec

- order-decreasing backward layer recursion - from a,b to r

- order-increasing forward layer recursion - from r to A,B

chebtr - Chebyshev broadband reflectionless quarter-wave transformer

chebtr2 - Chebyshev broadband reflectionless quarter-wave transformer
chebtr3 - Chebyshev broadband reflectionless quarter-wave transformer

Dielectric Waveguides
dguide - TE modes in dielectric slab waveguide
dslab
- solves for the TE-mode cutoff wavenumbers in a dielectric slab
dguide3 - TE and TM modes in asymmetric 3-slab dielectric waveguide

S-Parameters
gin
gout
nfcirc
nfig
sgain

I. MATLAB Functions
sgcirc
smat
smatch
smith
smithcir
sparam
circint
circtan

1075

stability and gain circles

S-parameters to S-matrix
simultaneous conjugate match of a two-port
draw basic Smith chart
add stability and constant gain circles on Smith chart
stability parameters of two-port
circle intersection on Gamma-plane
point of tangency between the two circles

Linear Antenna Functions

dipdir
dmax
dipole
traveling
vee
rhombic

dipole directivity
computes directivity and beam solid angle of g(th) gain
gain of center-fed linear dipole of length L
gain of traveling-wave antenna of length L
gain of traveling-wave vee antenna
gain of traveling-wave rhombic antenna

king
kingeval
kingfit
kingprime

Kings 3-term sinusoidal approximation

evaluate Kings 3-term sinusoidal current approximation
fits a sampled current to Kings 2-term sinusoidal approximation
converts Kings 3-term coefficients from unprimed to primed form

hbasis
hdelta
hfield
hmat
hwrap
kernel
pfield
pmat

basis functions for Hallen equation

solve Hallens equation with delta-gap input
solve Hallens equation with arbitrary incident E-field
Hallen impedance matrix with method of moments and point-matching
wraps a Toeplitz impedance matrix to half its size
thin-wire kernel computation for Hallen equation
solve Pocklingtons equation with arbitrary incident E-field
Pocklington impedance matrix with method of moments and point-matching

hcoupled - solve Hallens equation for 2D array of non-identical parallel dipoles

hcoupled2 - solve Hallens equation for 2D array of identical parallel dipoles
gain2d
gain2s
imped
imped2
impedmat
resonant
yagi

normalized gain of 2D array of parallel dipoles with Hallen currents

normalized gain of 2D array of parallel dipoles with sinusoidal currents
mutual impedance between two parallel standing-wave dipoles
mutual impedance between two parallel standing-wave dipoles
mutual impedance matrix of array of parallel dipole antennas
calculates the length of a resonant dipole antenna
simplified Yagi-Uda array design

1076

23. Appendices

Antenna Array Functions

gain1d
bwidth
binomial
dolph
dolph2
dolph3
multibeam
prol
prolmat
scan
sector
steer
taylornb
taylor1p
taylorbw
uniform
woodward
ville

normalized gain computation for 1D equally-spaced isotropic array

beamwidth mapping from psi-space to phi-space
binomial array weights
Dolph-Chebyshev array weights
Riblet-Pritchard version of Dolph-Chebyshev
DuHamel version of endfire Dolph-Chebyshev
multibeam array design
prolate array design
prolate matrix
scan array with given scanning phase
sector beam array design
steer array towards given angle
Taylor n-bar line source array design
Taylor 1-parameter array design
Taylor B-parameter and beamwidth
uniform array weights
Woodward-Lawson-Butler beams
Villeneuve array design

chebarray - Breslers Chebyshev array design method (written by P. Simon)

Gain Plotting Functions

abp
abz
ab2p
abz2

polar gain plot in absolute units

azimuthal gain plot in absolute units
polar gain plot in absolute units - 2*pi angle range
azimuthal gain plot in absolute units - 2pi angle range

dbp
dbz
dbp2
dbz2

polar gain plot in dB

azimuthal gain plot in dB
polar gain plot in dB - 2*pi angle range
azimuthal gain plot in dB - 2pi angle range

abadd
abadd2
dbadd
dbadd2
addbwp
addbwz
addcirc
addline
addray

add
add
add
add
add
add
add
add
add

gain in absolute units

gain in absolute units - 2pi angle range
gain in dB
gain in dB - 2pi angle range
3-dB angle beamwidth in polar plots
3-dB angle beamwidth in azimuthal plots
grid circle in polar or azimuthal plots
grid ray line in azimuthal or polar plots
ray in azimuthal or polar plots

Aperture Antenna Functions

diffint - generalized Fresnel diffraction integral
diffr
- knife-edge diffraction coefficient
dsinc
- the double-sinc function cos(pi*x)/(1-4*x^2)
fcs
fcs2

- Fresnel integrals C(x) and S(x)

- type-2 Fresnel integrals C2(x) and S2(x)

hband
heff
hgain
hopt
hsigma

horn antenna 3-dB width

aperture efficiency of horn antenna
horn antenna H-plane and E-plane gains
optimum horn antenna design
optimum sigma parametes for horn antenna

Miscellaneous Utility Functions

ab
db

- dB to absolute power units

- absolute power to dB units

c2p
p2c

- complex number to phasor form

- phasor form to complex number

d2r
r2d

- degrees to radians
- radians to degrees

dtft

- DTFT of a signal x at a frequency vector w

I. MATLAB Functions

1077

I0
ellipse
etac
wavenum
poly2

modified Bessel function of 1st kind and 0th order

polarization ellipse parameters
eta and c
calculate wavenumber and characteristic impedance
specialized version of poly with increased accuracy

quadr
quadrs
quadr2
quadrs2

Gauss-Legendre quadrature weights

quadrature weights and evaluation
Gauss-Legendre quadrature weights
quadrature weights and evaluation

Ci
Cin
Si
Gi

cosine integral Ci(z)

cosine integral Cin(z)
sine integral Si(z)
Greens function integral

sinhc
asinhc
sqrte

- hyperbolic sinc function

- inverse hyperbolic sinc function
- evanescent SQRT for waves problems

flip
blockmat
upulse
ustep

flip a column, a row, or both

manipulate block matrices
generates trapezoidal, rectangular, triangular pulses, or a unit-step
unit-step or rising unit-step function

dnv
snv
ellipK
ellipE
landenv

dn elliptic function at a vector of moduli

sn elliptic function at a vector of moduli
complete elliptic integral of first kind at a vector of moduli
complete elliptic integral of second kind at a vector of moduli
Landen transformations of a vector of elliptic moduli

and evaluation points

points on subintervals
and evaluation points
points on subintervals

References

Books
[1] S. A. Schelkunoff, Electromagnetic Waves, Van Nostrand, New York, 1943.

[4] R. W. P. King, R. B. Mack, and S. S. Sandler, Arrays of Cylindrical Dipoles, Cambridge Univ. Press,
Cambridge, 1968.
[5] J. D. Kraus, Antennas, 2nd ed., McGraw-Hill, New York, 1988.
[6] C. A. Balanis, Antenna Theory, Analysis and Design, 2nd ed., Wiley, New York, 1996.
[7] W. L. Stutzman and G. A. Thiele, Antenna Theory and Design, 2nd ed., Wiley, New York, 1998.
[8] R. S. Elliott, Antenna Theory and Design, Prentice Hall, Upper Saddle River, NJ, 1981.
[9] R. E. Collin and F. J. Zucker, eds., Antenna Theory, parts 1 and 2, McGraw-Hill, New York, 1969.
[10] A. W. Rudge, K. Milne, A. D. Olver, and P. Knight, eds., The Handbook of Antenna Design, vols. 1 and
2, 2nd ed., Peter Peregrinus Ltd., London, 1986.
[11] R. C. Johnson, ed., Antenna Engineering Handbook, 3d ed., McGraw-Hill, New York, 1993.
[12] T. S. M. Maclean, Principles of Antennas: Wire and Aperture, Cambridge Univ. Press, Cambridge,
1986.

MATLAB Movies
grvmovie1
grvmovie2
pulsemovie
pulse2movie
RLCmovie
TDRmovie
xtalkmovie
dipmovie

[2] S. A. Schelkunoff and H. T. Friis, Antennas, Theory and Practice, Wiley, New York, 1952.
[3] R. W. P. King, The Theory of Linear Antennas, Harvard Univ. Press, Cambridge, MA, 1956.

[13] J. R. Wait, Introduction to Antennas and Propagation, Peter Peregrinus, Ltd, London, 1986.
-

pulse propagation with slow and negative group velocity (vg < 0)
pulse propagation with slow and fast group velocity (vg > c)
step and pulse propagation on terminated transmission lines
step propagation on two cascaded lines
step getting reflected off a reactive termination
fault location by time-domain reflectometry
crosstalk signals on coupled transmission lines
electric field pattern of radiating Hertzian dipole

[14] T. A. Milligan, Modern Antenna Design, McGraw-Hill, New York, 1985.

[15] R. C. Hansen, Microwave Scanning Antennas, Academic Press, New York, vol. I, 1964, vols II and III,
1966.
[16] R. C. Hansen, Phased Array Antennas, Wiley, New York, 1998.
[17] E. Brookner, ed., Practical Phased Array Antenna Systems, Artech House, Boston, 1991.
[18] R. J. Mailloux, Phased Array Antenna Handbook, Artech House, Norwood, MA, 1994.
[19] R. E. Collin, Antennas and Radiowave Propagation, McGraw-Hill, New York, 1985.
[20] H. Mott, Antennas for Radar and Communications: A Polarimetric Approach, Wiley, New York, 1992.
[21] S. Silver, ed., Microwave Antenna Theory and Design, Peter Peregrinus, Ltd, London, 1984.
[22] R. F. Harrington, Time-Harmonic Electromagnetic Fields, McGraw-Hill, New York, 1961.
[23] G. S. Smith, An Introduction to Classical Electromagnetic Radiation, Cambridge Univ. Press, Cambridge, 1997.
[24] C. G. Someda, Electromagnetic Waves, Chapman and Hall, London, 1998.
[25] E. J. Rothwell and M. J. Cloud, Electromagnetics, CRC Press, Boca Raton, FL, 2001.
[26] L. B. Felsen and N, Marcuvitz, Radiation and Scattering of Waves, IEEE Press, New York, 1994.
[27] L. Tsang, J. A. Kong, and K-H. Ding, Scattering of Electromagnetic Waves, Wiley, New York, 2000.
[28] S. Drabowitch, A. Papiernik, H. Grifths, and J. Encinas, Modern Antennas, Chapman & Hall, London,
1998.
[29] A. D. Olver, Microwave and Optical Transmission, Wiley, Chichester, England, 1992.
[30] J. A. Kong, Electromagnetic Wave Theory, 2nd ed., Wiley, New York, 1990.
[31] H. C. Chen, Theory of Electromagnetic Waves, McGraw Hill, New York, 1983.

1078