99 views

Uploaded by Dania Alashari

Antenna and Electromagnetic

- 8-Yangui-et-al.pdf
- Solution Manual for Time-Harmonic Electromagnetic Fields - Harrington
- Cyclic Symmetry Topics
- s511p10
- wideband simple dieletric resonator antenna
- A Study on the Vibrational Effects of Adding an Auxiliary Chassis to a 6-Ton Truck.pdf
- Harrington, Roger F_[Time-harmonic electromagnetic fields].pdf
- Lecture 17
- Lec 1b Introduction
- Antenas Inteligentes
- Ecc Derivation
- 3P7_0879
- Stasiun Bumi LABUHA Stabillo
- Nkwenda3cap_5 and 6 Acceptance
- L12_Loop
- Andrew DBXLH-6565C-VTM.pdf
- antenna design
- BPH-202 Waves & Optics
- HFSS Antennas - Arrays.pdf
- Sten_TAP_01

You are on page 1of 378

Yaduvanshi

HarishParthasarathy

Rectangular

Dielectric

Resonator

Antennas

Theory and Design

Rectangular Dielectric Resonator Antennas

Rajveer S. Yaduvanshi Harish Parthasarathy

Rectangular Dielectric

Resonator Antennas

Theory and Design

123

Rajveer S. Yaduvanshi Harish Parthasarathy

Department of ECE Department of ECE

Ambedkar Institute of Advanced Netaji Subhas Institute of Technology

Communication Technologies and New Delhi, Delhi

Research India

New Delhi, Delhi

India

DOI 10.1007/978-81-322-2500-3

Springer India 2016

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part

of the material is concerned, specically the rights of translation, reprinting, reuse of illustrations,

recitation, broadcasting, reproduction on microlms or in any other physical way, and transmission

or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar

methodology now known or hereafter developed.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this

publication does not imply, even in the absence of a specic statement, that such names are exempt from

the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors and the editors are safe to assume that the advice and information in this

book are believed to be true and accurate at the date of publication. Neither the publisher nor the

authors or the editors give a warranty, express or implied, with respect to the material contained herein or

for any errors or omissions that may have been made.

Preface

onstrated by Richtmeyer in 1939. Richtmeyer showed that these dielectric ceramics

can resonate. Theory of DRA was expanded by Okaya and Brash in 1960. More

experimental work on DRAs, done by Long in 1980, proved that DRAs can become

efcient radiators and can be used as antennas. S.A. Long experimentally imple-

mented DRAs of different shapes and sizes as a low-prole antenna.

Analysis and studies on characteristic equation, radiation patterns, and excitation

methodology made DRAs popular by providing a new avenue compared to tradi-

tional patch antennas suffering from low gain and low bandwidth. Aldo Petosa

made DRAs a very successful candidate as functional antennas. Both the limitations

of low gain and low bandwidth in patch antennas can be eliminated by the use of a

rectangular dielectric resonator antenna (RDRA) operating in higher modes and

hybrid modes.

The modes theory of RDRA gives an important analysis on current distribution,

impedance, and radiation patterns of an antenna. Modes form a real, orthogonal

basis function for currents on the antenna. These are dened by boundary value

problems using eigenvalues and eigenvectors. The scope of this book has been

restricted to RDRAs, however, the concept can be extended to other geometries. In

RDRAs, once the excitation is given, the total distributed current on the antenna

structure becomes a weighted sum of eigen currents or a superposition of various

modes at any instant of time.

Resonant modes in RDRAs can be classied as dominant and higher modes.

Dominant modes correspond to lowest resonant frequency. These are called as TE,

TM, and HEM modes. E and H eld formats inside the RDRA at any instant of time

at a known frequency are termed as resonant modes. Modes excitation is directly

related to the surface current densities of the structure due to applied RF current.

This current gets converted into modal elds based on Maxwells equations. These

elds are restricted by RDRA boundary conditions. Reflection and refraction of

electromagnetic waves takes place because of dielectric interface at the boundary.

The generation of higher modes generally depends on RF excitation, device

dimensions, permittivity of dielectric material and coupling techniques used in

v

vi Preface

design of the antenna. The higher-order modes and hybrid modes have much

flexibility and design space in RDRA for different applications, but the excitation

techniques are complex. Rectangular DRA has a high degree of design flexibility

due to two aspect ratios (a/d and b/d), low cost, simplicity, and ease of fabrication.

It can retrot to the existing patch antenna technology for gain improvements.

Researchers have long felt the need for a rigorous theoretical analysis on reso-

nant modes of RDRA, and resonators have become a demanding eld for industry

and academia. This is because knowledge of resonant modes gives physical insight

to the antenna designer, based on which input impedance and radiation character-

istics can be predicted. We hope that this book will help to ll the gap.

The investigations and theory developed are based on applying waveguide

theory models. Propagation of electromagnetic elds has been taken along z-axis,

i.e., exp(c z). Initially, these are exploited via the Maxwells curl equations and

then manipulating them to express the transverse components of the elds in terms

of the partial derivatives of the longitudinal components of the elds w.r.t. x and

y (i.e., the transverse coordinates).

Waveguide models of four different boundary conditions lled with homoge-

neous as well as inhomogeneous dielectric materials with linear and nonlinear

permittivity, permeability, and conductivity have been developed to determine TE

and TM propagating electromagnetic elds. These have resulted in different sine

cosine combinations. TE modes generation required Hz elds as longitudinal elds

and Ex, Ey, Hx, and Hy elds as transverse elds.

If input excitation is applied along x-axis as partial elds, y-axis will have xed

variation and z-axis will have desired variation in propagating elds. For example,

TE d13 . Similar cases can be developed for TM modes so as to propagate Ez elds

as longitudinal and Ex, Ey, Hx, and Hy as transverse elds. Hz eld will get vanished

because of boundary conditions.

An equivalent but computationally simpler way to pass on from waveguide

@

physics to resonator physics is to just replace (c) by ( @z ) in all the waveguide

formulae that express the tangential eld components in terms of the longitudinal

components. This is done after solving the full 3D Helmholtz equations using

separation of variable as x, y, z.

x2 Ez

r 2

2

0

c H z

L, C oscillators with different L, C values. The outcome of all this analysis enables

us to write down the E and H elds inside the resonator, as superposition of four

and three vector-valued basis functions.

Preface vii

X

1 n o

E x; y; z; t Re Cmnpejxmnpt wEmnp x; y; z

mnp1

X

1 n o

E x; y; z

Re Dmnpejxmnpt / mnp

mnp1

and

X

1 n o

H x; y; z; t Re Cmnpejxmnpt wH

mnp

x; y; z

mnp1

X

1 n o

H x; y; z

Re Dmnpejxmnpt / mnp

mnp1

We note that there are only two sets {C(mnp)} and {D(mnp)} of linear com-

bination of coefcients using from the Ez and Hz expansions. The vector-valued

complex functions are as follows:

E ; wH ; /

wEmnp ; / H 2 R3

mnp mnp mnp

fcos; sing functions and hence for m0 ; n0 ; p0 6 m; n; p, each function of the set,

where m, n, p are integers.

n o

; wH ; /

wEmnp ; /

E H

mnp mnp mnp

n o

E ; wH ; /

wEm0 n0 p ; / H

0

mnp0 mnp 0

mnp 0 0

w.r.t. the measure of dx dy dz over surface of RDRA [0, a] [0, b] [0, d], where

a, b, and d are RDRA dimensions. The exact form of the function / E; /

H ; wE ; wH

depends on the nature of RDRA boundaries.

Excitation of RDRA plays very important role for modal analysis. To calculate

the amplitude coefcients {C(mnp)} and {D(mnp)}, we assume that at z = 0, an

e e

excitation Ex x; y; t or Ey x; y; t is applied for some time say t [0, T] and then

removed, as usually is done in L, C oscillators. Then, the Fourier components in this

excitation corresponding to the frequencies {mnp} are excited, and their solutions

are the oscillations for t > T. The other Fourier components decay within the

resonator.

viii Preface

orthonormality:

X

jxmnpt E

Re Cmnpe wmnp X x; y; 0

mnp

E x; y; 0 E e x; y; t

Re Dmnpejxmnpt / mnp X x

and

X

Re C mnpejxmnpt wEmnp Y x; y; 0

mnp

E x; y; 0 Ee x; y; t

Re Dmnpejxmnpt / mnp Y y

By using orthogonality of {wEmnp X x; y; 0, / mnp X

we write p xed and likewise of {wEmnp Y x; y; 0;/ E x; y; 0}; in addition, we

mnp Y

need to use KolmogorovArnoldMoser (KAM) type of time averaging to yield:

E x; y; 0

C mnpwEmnp X x; y; 0 Dmnp/ mnp X

ZT

lim 1

Exe x; y; tejxmnpt dt

T ! 1 2T

T

and likewise

CmnpwEmnp Y x; y; 0 Dmnp/

E

x; y; 0

mnp Y

ZT

lim1 e

EY x; y; tejxmnpt dt

T ! 1 2T

T

In this book, RDRA resonant modes theoretical as wells as practical aspects have

been investigated along with rigorous mathematical analysis for TE, TM, and HEM.

Higher modes generation and control of resonant modes have been experimented.

Shifting of dominant mode toward higher modes and vice versa is desired phenom-

enon for recongurability, merging of neighboring resonant modes have been

exploited with simulation results. Use of higher modes for practical applications in

antennas has been described. Merging of neighboring modes signicantly increased

antenna bandwidth. The device miniaturization using high-permittivity materials has

been described. The devising control on modes has imparted reconguration of

operating frequency, beam pattern, beam width, polarization, gain, and bandwidth.

Higher modes radiation pattern, sensitivity analysis by changing dimensions, and

permittivity analysis by changing permittivity have been mathematically modeled, and

each is supported with simulated and experimental results. Selecting and cancelling a

particular resonant mode has also been described. The concept of modes has been

Preface ix

modes in RDRA can be used for software-dened radios and military applications,

where frequent change of antenna parameters is operational requirement. For auto-

mation on modes control, microcontrollers equipped with lookup table can be used.

The modes have been modeled by R, L, C networks. Antenna far elds patterns

and impedance have been computed and measured. Analysis on hybrid modes in

RDRA has been discussed. Hybrid modes are complex to determine. Their math-

ematical formulations have been described. These modes are diversied.

The excitation of hybrid modes is complex, and their effective control can

revolutionize the antenna technology. Detailed study of mathematical modeling of

hybrid modes has been described. Hybrid modes are more popular for azimuthally

eld variations. The transcendental equation and characteristic equation for RDRA

modes are used for determining propagation constants and then resonant frequency.

The solution of resonant modes can be obtained using the following:

(a) Hz and Ez elds are expressed as umnp(x, y, z), vmnp(x, y, z) and xmnp based on

solving Maxwells equations with given boundary conditions.

(b) At z = 0, surface (x, y) excitation with applied surface current density is

Jsx x; y; t; Jsy x; y; t

Js x; y; d Jsx ; Jsy ^z H Hy ; Hx ;

Hz is terms Dmnp, and Ez terms as Cmnp.

(d) Equate tangential component of Ez at boundary, i.e., Ey jz0 to zero, and

compute the coefcients Dmnp for Hz and Cmnp of Ez.

(e) Excited by mnp and arbitrary feed position in xy plane x0 ; y0 /0 ; h0

" #

X n o

H? ~

Re Dmnp e jxmnpt

r? ~umnp x; y; z

mnp

X n o

Re C~ mnp ejxmnpt r? ~umnp x; y; z

mnp

and similarly E? .

Depending on the boundary conditions, four cases have been developed.

In RDRA, these four walls are assumed as perfect magnetic conductors and top and

bottom walls are taken as perfect electric conductors.

umnp sin sin sin Ez

x Preface

Top and bottom walls are perfect magnetic conductors, and all four sidewalls are

PEC

Transcendental equation is used to solve propagation constants, i.e., kx, ky, and kz.

The propagation constant gives rise to resonant frequency with the help of charac-

teristic equation. These wave numbers kx, ky, and kz are in x, y, and z-directions,

respectively. The free space wave number is k0. The resonant frequency can be

determined from combined solution of transcendental equation and characteristic

equation of rectangular DRA. Time-averaged electric energy = time-averaged

magnetic energy

0

e0 k02 kx2 ky2 kz2

0

kz 6 pp=d

kz

tankz d q ;

k0 er 1 kz2

2

The contents of this book are the outcome of our research work on RDRA

higher-order resonant modes. In this book, analyses have been restricted to rect-

angular resonators higher modes, however, the concept can be extended to other

geometry resonators, such as cylindrical, conical, and hemispherical. With this

book, we hope to ll the gap for rigorous theoretical analysis on RDRA resonant

modes. The work is supported with live projects data and their case studies. This

book should be very useful for antenna designers, both in research and development

and for practical implementations. This book is written in a simple and reader

friendly manner and can be easily understood with an initial knowledge of basic

Preface xi

electromagnetic theory. All the chapters are self-reliant, and no initial specialization

is required to understand the contents. We hope that this book will help open the

design space for a new class of antenna implementations.

This book is organized into 12 chapters including rigorous theoretical analysis of

modes along with case studies and design data annexure. Introduction along with

history of RDRA is given in Chap. 1. Introduction of resonant modes is explained

in Chap. 2. Mathematical derivations for modes and the generation of TE/TM

modes have been discussed in Chap. 3. Chapter 4 presents the derivation of RDRA

transcendental equations. In Chap. 5, mathematical description of amplitude coef-

cients of even and odd modes is presented. Chapter 6 contains radiation param-

eters and mathematical explanations of RDRA. Chapter 7 describes derivations of

higher-order resonant modes and their applications for high-gain antenna designs.

Chapter 8 explains the effect of angular variation on excitation to produce various

types of radiation patterns to meet military requirements. Chapter 9 discusses

sensitivity analysis and mathematical modeling of radiation pattern solutions in

RDRA. Chapter 10 presents the excitation of hybrid modes in RDRA and their

possible applications. Chapter 11 covers inhomogeneous solution along with

measurements. Basic RDRA resonant frequency formulations, materials required,

and their sources are given in the annexures. Complete and detailed solutions of

RDRA have been explained in case studies. Design data are provided in the

annexures. Chapter 12 discusses case studies.

Harish Parthasarathy

Acknowledgments

We would like to express special thanks to reviewers and editors for sparing their

valuable time in reviewing this manuscript. We acknowledge the support and

cooperation extended by all our family members, without whom it would not have

been possible to complete this book.

We in person acknowledge the healthy discussion with our students and col-

leagues. Special thanks to Sujata, R. Gupta, Promod, Vipin, and Chander P. for

imparting their help and support in preparing this manuscript, i.e., typing work and

proofreading. We extend special thanks to Springer for publishing this book.

Rajveer S. Yaduvanshi

Harish Parthasarathy

xiii

Contents

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2 History of DRA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.3 Working Mechanism of RDRA. . . . . . . . . . . . . . . . . . . . . . . 2

1.4 Antenna Radiation Parameters . . . . . . . . . . . . . . . . . . . . . . . 4

1.5 Advantage of RDRA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.6 Resonant Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.7 Characterization of Resonant Modes . . . . . . . . . . . . . . . . . . . 6

1.8 Magnetic Dipole Moment . . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.9 Spring Resonator of Length L. . . . . . . . . . . . . . . . . . . . . . . . 8

References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.2 Type of Modes (TE, TM, HEM). . . . . . . . . . . . . . . . . . . . . . 14

2.3 Solutions of Helmholtz Equation. . . . . . . . . . . . . . . . . . . . . . 15

2.4 Rectangular Waveguide Analysis . . . . . . . . . . . . . . . . . . . . . 17

2.5 Two-Dimensional Resonator. . . . . . . . . . . . . . . . . . . . . . . . . 20

2.6 Basic Mathematical Representation of Resonant Modes . . . . . . 21

2.7 Voltage Source Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

2.8 Resonant Modes Generation . . . . . . . . . . . . . . . . . . . . . . . . . 27

2.9 MATLAB Simulated Results . . . . . . . . . . . . . . . . . . . . . . . . 29

Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

3.1 Rectangular DRA with Homogeneous Medium . . . . . . . . . . . . 34

3.2 Rectangular DRA Mathematical Modeling . . . . . . . . . . . . . . . 36

3.2.1 Model-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

3.2.2 Model-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

xv

xvi Contents

3.2.3 Model-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

3.2.4 Model-4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

3.2.5 Basic Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

in Rectangular DRA . . . . . . . . . . . . . . . . . . . . . . . . . . . ....... 57

4.1 Case-1: Top and Bottom Walls as PMC and Rest

of the Four Walls are PEC . . . . . . . . . . . . . . . . . . . . . . . . . . 60

4.2 Case-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

4.3 MATLAB Simulation Results. . . . . . . . . . . . . . . . . . . . . . . . 78

4.4 Resonant Frequency of RDRA for Experimentations . . . . . . . . 87

5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

5.2 Amplitude Coefficients Cmnp . . . . . . . . . . . . . . . . . . . . . . . . 104

5.3 RDRA Maxwells Equation-Based Solution . . . . . . . . . . . . . . 106

5.4 RDRA Inhomogeneous Permittivity and Permeability . . . . . . . 112

5.5 RDRA with Probe Current Excitation . . . . . . . . . . . . . . . . . . 118

5.6 RDRA Resonant Modes Coefficients

in Homogeneous Medium . . . . . . . . . . . . . . . . . . . . . . . . . . 121

5.7 RDRA Modes with Different Feed Position . . . . . . . . . . . . . . 123

5.8 R, L, C Circuits and Resonant Modes . . . . . . . . . . . . . . . . . . 125

5.9 Resonant Modes Based on R, L, C Circuits . . . . . . . . . . . . . . 130

6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 135

6.2 Radiation Pattern of RDRA Due to Probe Current i(t)

and Probe Length dl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

6.2.1 Radiation Pattern. . . . . . . . . . . . . . . . . . . . . . . . . . 139

6.3 Poynting Vector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140

6.4 Moat-Shaped RDRA Radiation Pattern . . . . . . . . . . . . . . . . . 141

6.5 Quality Factor of RDRA . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

7.1 Introduction to Higher Modes. . . . . . . . . . . . . . . . . . . . . . . . 148

7.2 Resonant Frequency and RDRA Structure . . . . . . . . . . . . . . . 156

7.2.1 Fields in Rectangular DRA. . . . . . . . . . . . . . . . . . . 158

7.3 Modes (Resonant) Mathematical Solution. . . . . . . . . . . . . . . . 165

7.4 Top-Loading RDRA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166

7.5 Simulated HFSS Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

7.6 Modes at Varying Heights of RDRA . . . . . . . . . . . . . . . . . . . 168

7.7 Distortions Due to Overlap of Dipole Moment . . . . . . . . . . . . 168

7.8 Prototype and Anechoic Chamber Experimentations . . . . . . . . 168

7.9 Adjacent Modes Combination for Broadband Applications . . . . 169

Contents xvii

7.10 Effect of Air Gap Between RDRA and Ground Plane . . . . . . . 169

7.11 Effect of Asymmetrical Wells Inside RDRA. . . . . . . . . . . . . . 170

7.12 Effect of Moat Insertion Inside RDRA. . . . . . . . . . . . . . . . . . 170

7.13 Effect of a/b and d/b Aspect Ratio . . . . . . . . . . . . . . . . . . . . 171

Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

and Resonant Modes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 181

8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 181

8.2 Angular Shift in Excitation. . . . . . . . . . . . . . . . . . . . . ..... 185

8.3 Radiation Pattern Based on Angle 0 ; /0

Variation in xy Plane . . . . . . . . . . . . . . . . . . . . . . . . . ..... 189

8.4 Replacing Probe with Slot of Finite Dimensions

(Ls, Ws) at an Angle h0 ; /0 . . . . . . . . . . . . . . . . . . . ..... 190

8.5 HFSS Computed Radiation Pattern with Shifted hi ; /i

Slot Positions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 192

8.6 Experimentations . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 193

9.1 MATLAB Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206

9.2 HFSS Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

9.2.1 HFSS Result. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208

9.3 Radiation Pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210

10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211

10.2 Mathematical Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214

10.3 Modes in Homogeneous Medium with Source Terms . . . . . . . 218

10.4 Current Density in RDRA . . . . . . . . . . . . . . . . . . . . . . . . . . 219

10.5 E and H Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220

10.6 Mathematical Modeling of Hybrid Modes . . . . . . . . . . . . . . . 221

10.7 General Solution of Hybrid Modes (HEM) . . . . . . . . . . . . . . . 225

10.8 HFSS Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230

10.9 Prototype RDRA Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 232

Solution in Rectangular DRA . . . . . . . . . . . . . . . . . . . . . . . .... 233

11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 233

11.2 Mathematical Model . . . . . . . . . . . . . . . . . . . . . . . . . . .... 234

11.3 Applications: Hybrid Modes Generation Inside RDRA

Can Be Used for Polarization Diversity . . . . . . . . . . . . . .... 250

11.3.1 RF Measurements for Antenna Parameters . . . . .... 250

xviii Contents

12.1 Structure and Hardware Experimentations . . . . . . . . . . . . . . . 251

12.1.1 RDRA Antenna Results . . . . . . . . . . . . . . . . . . . . . 251

12.2 RDRA with ManganeseManganese Material as Dielectric . . . . 256

12.3 Dual-Feed RDRA with Measurements Results . . . . . . . . . . . . 269

12.4 Isolated and Grounded RDRA . . . . . . . . . . . . . . . . . . . . . . . 279

12.4.1 S11 Plot. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279

12.4.2 Gain Plot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

12.4.3 Impedance (Z) Plot . . . . . . . . . . . . . . . . . . . . . . . . 281

12.4.4 Design of RDRA with Ground Plane . . . . . . . . . . . . 282

12.4.5 S11 Plot. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

12.4.6 Gain Plot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

12.4.7 Impedance Plot . . . . . . . . . . . . . . . . . . . . . . . . . . . 283

12.4.8 Comparison of DRA With and Without

Ground Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284

12.4.9 Detailed Design of Aperture-Coupled DRA . . . . . . . 285

12.4.10 Return Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286

12.4.11 Radiation Pattern. . . . . . . . . . . . . . . . . . . . . . . . . . 287

Annexure-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289

Annexure-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295

Annexure-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

Annexure-4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313

Annexure-5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329

Annexure-6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365

About the Authors

Department of ECE at Ambedkar Institute of Advanced Communication

Technologies and Research Delhi for the last 7 years. Before that, he was working

as a senior scientic ofcer in Ministry of Defence, Government of India. He has a

total of 31 years of experience in teaching and research. He is the author of a book

on MHD Antenna, Design and Applications and has published 27 papers in reputed

journals and conferences. He has supervised 151 B.Tech. projects and 11 M.Tech.

projects and is currently supervising seven Ph.D. students. He holds an M.Tech.

degree from NIT Allahabad and a Ph.D. degree from NSIT, Delhi (Delhi

University). His research interests include design of Dielectric resonator antennas,

MHD embedded antennas and analysis of higher modes in DRA.

Dr. Harish Parthasarathy is working as a professor in the Department of ECE at

Netaji Subhas Institute of Technology. He is the author of more than 11 books and

has guided several Ph.D. students. He holds his Ph.D. and B.Tech. degrees from IIT

Delhi. His research involves antenna and signal processing with specialization in

mathematical modeling.

xix

Chapter 1

Rectangular DRA Fundamental

Background

Working mechanism of rectangular DRA (RDRA) has been explained. Survey

work along with citations on related works based on the available literature has

been described. RDRA as a new candidate in the eld of antennas whose com-

parison has been made with existing patch antennas. Their advantages have been

listed. Mathematical solution of one-dimensional resonator has been derived.

Working mechanism Survey

Characteristics Advantages One-dimensional resonator

1.1 Introduction

Antenna is usually visualized as metallic device for radiating and receiving elec-

tromagnetic waves. It is an interface (transducer) between space and communica-

tion device. For wireless communication system or radar system, antenna is used to

couple radio energy from transmitter to space in transdirection, and space to

receiver in receive direction. Antennas are frequency dependent. The design of

antenna corresponds to specic bandwidth and resonant frequency. These are

purely designed as per requirements. The antenna rejects all signals beyond their

bandwidth. An antenna is an integral part of any wireless communication. Hence,

its development must be in synchronization with communication system. There

have been revolutionary developments in communication systems since last dec-

ades. The emergent requirements are being felt in antenna development. The

Gigabytes of data transmission at very high speeds are todays communication

requirements. To match todays advanced communications requirements, rectan-

gular DRA (RDRA) is the most suitable candidate. Rectangular dielectric resonator

antenna is new kind of antenna, which is different from traditional metal or patch

antenna. The patch or metal antennas generally suffer from low bandwidth, high

conducting loss and low gain. RDRA has high gain and wide bandwidth antenna.

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_1

2 1 Rectangular DRA Fundamental Background

material having permittivity greater than 10 F/m. In 1939, R.D. Richtmeyer showed

that non-metalized dielectric material objects can resonate and function as an

antenna, these are called as dielectric resonator antenna [1, 2]. There were no

practical applications of these DRAs until 1960. Dielectric resonator antenna was

rst introduced by S.A. Long in 1980 [3]. Since then, vast research has been carried

out for analysis of DRA material properties, and their effective use as DRA. Various

shapes and excitation methods for DRA have been developed. Many research

papers have been published in reputed journals by researcher such as Kishk [4], Lee

[5], Leung [6], Luk [7], Mongia [8], Shum [9], Junker [4], Antar [10], and Petosa

[11] till date. No rigorous theoretical analysis for RDRA is available in the literature

so for. It is felt that if good literature along with sound mathematical analysis on

RDRA is made available, it can benet the society in large. Only few books are

available on introduction of DRA, but no book is available for sound theoretical

analysis supported with mathematical computations of RDRA.

atoms causes acceleration or deaccelerations which become electromagnetic wave

radiation. The radio waves are introduced into ceramics forming resonator as shown

in Fig. 1.1 from RF transmitter circuits. These RF waves bounce back and forth

between resonator walls, thus forming standing waves, hence stores electrical

energy. Oscillating current introduces oscillating magnetic elds, H elds, and

oscillating electric elds, E elds. The time-varying eld radiates away from

antenna into space due to accelerrating currents. The walls of ceramic formed

partially transparent magnetic walls, and the magnetic energy leaks through these

transparent walls due to fringing effect. Thus, radio power is radiated into space.

Let RDRA having dimensions a, b, d lengths is excited by external electric elds

Eix(x, y) and Eiy(x, y) in x, y plane. The equivalent circuit as shown in Fig. 1.2 is

1.3 Working Mechanism of RDRA 3

RLC circuit

drawn based on the electrical properties of this ceramic RDRA. This results into

longitudinal E z and H z elds. The probe currents are equated with RDRA radiating

currents as per principle of conservation of energy. In other words, time-average

(KAM) electric energies inside the RDRA are equated with time-average magnetic

energies. Figure 1.2 presents RDRA equivalent RLC circuit for computing quality

factor of RDRA.

Z Z

jE j2 dV jH j2 dV; 1:1

V V

h2mn c2 h2 ; 1:2

where c jpp

d .

These elds are computed using Helmholtz equations, taking into account of

source and RDRA boundaries. Mathematical solution of transverse and longitudinal

elds is obtained by half-wave and full-wave Fourier analysis, taking inside

medium and outside medium into consideration.

Figure 1.3 is shown as RDRA placed on innite ground plane. The image theory

can be applied to this RDRA for reducing its height. Varying sinusoidal in time,

energy flow in particular direction can be treated as power radiated per unit solid

angle (energy per unit area per unit time).

plane

4 1 Rectangular DRA Fundamental Background

Antenna radiation pattern

Power

Gain

Polarization

Impedance

Efciency radiation

Most suitable at microwaves and millimeter waves;

Compact in size and portable;

kg

Dimensions of RDRA are of the order of p ; choosing higher er RDRA size

er

can be reduced signicantly.

Ease of fabrication;

No frequency drift due to change in temperature;

High-power handling capability;

High gain and high bandwidth;

Can be integrated with MIMC;

RDRA has advantage of two aspect ratios. Hence, various modes can be gen-

erated by varying any of the aspect ratio;

Simple coupling schemes;

Bandwidth can be variable by choosing dielectric constant; and

High Q factor.

In RDRA, resonant modes represent the radiating phenomena with the help of

E and H eld patterns. These elds inside RDRA are presented azimuthally. With

the knowledge of modes, radiation characteristics of an antenna can be predicted.

The designer can get insight of antenna design and hence can provide correction in

the antenna design. Resonant modes are real current vectors. These modes are

found by orthogonal Fourier basis functions. These are generated based on the

current distribution on the surface of antenna due to eld perturbations. These can

be classied as TE or TM modes. The loss tangent d introduced is due permit-

tivity of the material. The principle conservation of energy is applied, in which,

1.6 Resonant Modes 5

to compute radiated elds. RDRA is excited by input radio frequency currents at

proper impedance match at input port. The transverse components are dened in

terms of longitudinal components and vice versa. The modal eld equations are

developed using Fourier basis functions of cosine or sine terms appearing based on

the RDRA boundary conditions, i.e., six walls of RDRA can be PMC, PEC, or any

combination of these PMC and PEC walls. Hence, resonant modes bring physical

insight into the radiating phenomena taking place inside the RDRA. The resonant

modes form a set of orthogonal functions to compute total current on the surface of

RDRA.

Figure 1.4 shows the resonant modes conguration generated into RDRA. Wave

can only propagate if wave vector k > kc, where kc is cutoff frequency. The lowest

resonance is called dominant mode.

This is solved and the solution consists of a superposition of a source (particular

solution) term and a homogeneous term (i.e., general solution of the homogeneous

yz plane

6 1 Rectangular DRA Fundamental Background

part). Two constants in the homogeneous part are determined by applying the

vanishing boundary conditions on Hz and Ez at top and bottom surfaces, i.e., at

z = 0, d.

The radiation can be identied as magnetic dipoles. Any function can be decom-

posed or separated by projecting that function into basis function, i.e., inner product

with basis function

Zb

hF1 ; F2 i rF2 ~

F1 ~ r dR; 1:3

a

The resonance modes are E and H eld patterns inside the RDRA. Figure 1.5 has

shown that electric elds are always associated with magnetic elds and vice versa.

These can be three types, i.e., TE, TM, and HEM modes. The amplitude coefcients

and phase of RDRA are Cmnp ; wmnp and Dmnp ; /mnp : Ez and Hz elds are based on

the orthonormality. These can be determined by applying principle of orthonor-

mality. The characteristics equations of RDRA are given as follows:

where k0 is free space wave and kx ; ky ; kz are propagation constants in x-, y-, and

z-directions, respectively. Also, k02 x20 l0 e0 ; hence, resonant frequency in free

space can be determined based on the free space wave number. To determine

propagation constants, i.e., kx ; ky and kz , knowledge of transcendental equation is

required. The transcendental equation is developed for RDRA when elds are

propagating in z-direction and given below as

q

d

kz tan kz er 1k02 kz2 ; 1:5

2

1.7 Characterization of Resonant Modes 7

the resonant mode generated, i.e., TE111 ; TM111 ; TE11d ; TE1d1 ; and TE11d . Fields

are expanded into summation of their modal functions, which may be by Cmnp and

Dmnp amplitude coefcients. Applying continuity equation across regional inter-

faces tangential elds, current distribution along surfaces of an antenna can be

computed as Je , i.e., current density. Input impedance Zin and radiation pattern Prad

can be computed based on the current distribution. Eigenvectors or Eigen functions

are formed as characteristic modes. Modes are orthogonal over source region.

E z electric elds produced by J n characteristic currents on the surfaces. These

modes are mainly dependent on the RDRA boundary and excitation.

Electrical walls of RDRA:

E tan n E 0; 1:6

H nor n H 0; 1:7

H tan n H 0; 1:8

E nor n E 0; 1:9

The solution of resonant modes shall vary in terms of sine and cosine as these are

dependent on PEC and PMC walls of RDRA. E z and H z fields can be determined

as linear combinations of these functions sin or cosine in xy plane and z-component

of source, to get these propagation constant. Propagating elds in particular

direction x or y or z is assumed to be continuous inside and outside the RDRA.

While taking into account inside the resonator both, reflected as wells as propa-

gating elds are available, outside the RDRA only outgoing eld components are

taken and reflected component is negated. This solves the transcended equation for

RDRA. The modal characteristics of antenna give rise to elds, i.e., resonant

modes. These are also known as eigenvector and eigenvalues. Eigenvectors are

current amplitudes Cmnp and Dmnp , and eigenvalues are resonant frequencies xmnp .

The resonant frequency can be given as follows:

r

c mp2 np2 pp2

fr m; n; p p ; 1:10

2p el a b d

8 1 Rectangular DRA Fundamental Background

The radiation in RDRA is taking place due to short magnetic dipole formation.

X

d e r; 1:11

where

d dipole moment

e charge

r distance between two charges

dX X

d er e v; 1:12

dt

dX

d :: e v; 1:13

dt

Hence, charges can radiate only if they move with acceleration. There will be no

radiation even if they move with xed or uniform velocity.

As shown in Fig. 1.6, one single string of AB length is applied with external

excitation to produce oscillations. These oscillations will give rise to resonant

frequency of the resonator.

xt C1 e jxt

1.9 Spring Resonator of Length L 9

Hence, C1 x2Cx

2

2

0

jxt

xt xC22x

e

2 , if x0 x; then, x(t) will be 1.

0

Now, x x0 d; when d is small deviation,

C2 e jxt

; 1:15

x0 xx0 x

C2 e jxt

d2x0

@2 1 @2

f x; t 0; at boundaries

@x2 c2 @t2

f 0; t 0 and f L; t 0

@ 2 x2 ^

f x; x 0; 1:16

@x2 c2

xx xx

C1 sin C2 cos 0 or ^f 0; x 0 or ^f L; x 0

c c

xL

sin c 0. Hence, kL np; sine values to be zero.

npc

x kc ; when n 1; 2; 3. . . where k x=c; 1:17

L

Length is 2L/3, dominant frequency 3x1 ; Eqs. (1.1)(1.17) used in this chapter

presented the mathematical concept of topic.

References

2. Richtmyer RD (1939) Dielectric resonator. J Appl Phys 10:391398

3. Wakino K, Tamura H, Sudo T (1987) Dielectric resonator materials and their applications.

Microw J 6:133148

4. Junker GP, Kajfez D, Kishk AA, Lisson AW (1995, May 11) Effect of aperture lling on slot

coupled dielectric resonator antennas operating in HEM11 mode. Electron Lett 31(10):774775

10 1 Rectangular DRA Fundamental Background

5. Lee RQ, Simons RN (1994) Bandwidth enhancement of dielectric resonator antennas. In:

IEEE antennas and propagation society international symposium, Seattle, WA, June 1994,

pp 15001503

6. Luk KM, Leung KW, Chow KY (1997) Bandwidth and gain enhancement of a dielectric

resonator antenna with the use of stacking element. Microw Opt Technol Lett 14(4):215217

7. Leung KW, Chow KY, Luk KM, Yung EKN (1997) Excitation of dielectric resonator antenna

using a soldered through probe. Electron Lett 33(5):349350

8. Mongia RK, lttipiboon A, Cuhaci M, Roscoe D (1994) Radiation Q-factor of rectangular

dielectric resonator antennas theory and experiment. In: IEEE antennas and propagation

society international symposium, Seattle, WA, pp 764767, June 1994

9. Shum SM, Luk KM, Leung WK, Wa K (1994) Mutual impedance of hemisphere dielectric

resonator antennas. IEEE Trans Antennas Propagat 42(12):16521654

10. Antar YMM, Chang D, Sequin G, Henry B, Keller MG (1998, Oct 5) Modied wave guide

model (MWGM) for rectangular dielectric resonator antennas. Microw Opt Technol Lett 19

(2):158160

11. Petosa A, littipiboon A, Cuhaci M, Larose R (1996) Bandwidth improvement for a micro strip

fed series array of dielectric resonator antennas. Electron Lett 32(7):608609

Chapter 2

Rectangular DRA Resonant Modes

and Sources

analysis for generation of different resonant modes have been presented in this

chapter. Realization of resonant modes based on MATLAB has also been worked.

Modes are generated by applying voltage source. Various types of resonant modes

have been described along with all possible applications.

Keywords Cavity resonator Resonant modes Type of modes Wave guide

analysis Mathematical description of resonant modes Simulated work

2.1 Introduction

In the early 1960s, Okaya and Barash [1] reported the rst ever DRA in the form of

a single-crystal TiO2. Since then, no rigorous theoretical analysis has been devel-

oped so far in the literature to evaluate the resonant modes in Rectangular DRA.

Based on Cherenkov principle of radiations, an external electric eld brings the

charges of the molecules of the dielectric into a certain ordered arrangement in

space and creates acceleration phenomenon in dielectric material itself. The

dielectric polarization P is equal to the total dipole moment induced in the volume

of the material by the electric elds. In most cases, the magnitude of polarization is

directly proportional to the intensity of the electric eld at a given point of a

dielectric. The relative permittivity is related to the dielectric susceptibility.

A dielectric resonator is dened as object of dielectric material which functions as

a resonant cavity by means of reflections at the dielectric air interface. The

discontinuity of the relative permittivity at the resonator surface allows a standing

electromagnetic wave to be supported in its interior at a particular resonant fre-

quency, thereby leading to maximum connement of energy within the resonator.

Certain elds distribution or modes will satisfy Maxwells equations and

boundary conditions. Resonant modes are eld structures that can exist inside the

DRA. Modes are the pattern of motion which repeat itself sinusoidally. Innite

number of modes can excited at same time. Any motion is superposition or

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_2

12 2 Rectangular DRA Resonant Modes and Sources

weighted sum of all the modes at any instant of time by combining amplitudes and

phases. As in the case of all resonant cavities, there are many possible resonant

modes that can be excited in dielectric resonators. The boundary conditions are

n H = 0; where H denotes the electric eld intensity and n denotes the normal to

the surface of the resonator.

And, n E = 0, is not necessarily satised at all the surfaces of the RDRA by all

the modes. Different resonant modes have distinct electromagnetic eld distribu-

tions within the DRA, and each mode may provide a different radiation pattern.

Operation of DRA is based on the process that if excitation is applied, then a

high magnetic eld is created inside the dielectric object placed on a ground plane.

Phenomena which occur like a charge particle passing to the eld create the

physical environment like any metal ball passing through liquid. Thus, there will be

change in the eld, contraction, and expansion, which causes fringing effect. This

way dielectric object starts to radiate. Another phenomenon that occurs is that there

might be reflection of the eld from sidewalls of dielectric object due to change in

the refractive index of the medium. The dielectric object acts as an oscillator.

Theory of characteristic modes can be applied in the design of antenna or DRA.

These modes give insight into physical phenomenon taking place inside device in

terms of current vectors as maxima and minima. This helps to locate the feeding

point and desired dimension of RDRA.

In 1968, modes were introduced by Garbacz and later by Harrington. Inagaki

gave simpler theory on modes for radiation mechanism in an antenna. It requires lot

of computation, for loading quality factor, double feeding to improve bandwidth,

and circular polarization. Characteristic modes are current modes or eigenvectors,

which are the solution of characteristic equation. These are orthogonal functions

that can predict total current on surfaces of body of the antenna. Also, desired mode

can be excited for specic radiating pattern. Excitation of mode mainly depends on

feeding arrangement, geometry of the device, and dielectric material used. In time

domain, varying electric eld can produce magnetic elds and vice versa. By

applying RF, excitation currents in RDRA get converted into surface current

density distributed over the surfaces, i.e., RF excitation with proper impedance

match can generates J. This probe current produced magnetic vector potential A.

The radiated magnetic elds are presented in the form E-electric eld intensity

using Lorentz gauge transformation. An antenna can propagate electromagnetic

elds, if wave vector k > kc . The cutoff wave vector kc determines the cutoff

frequency. There can be dominant resonant frequency or higher-order resonant

frequency. The propagation takes place along x-axis if propagation constant k [ np b.

There will not be any propagation if k np b , as it will lead to formation of standing

waves. Similar conditions persist for propagation along y-axis and z-axis.

Maxwells equations dene the behavior of electromagnetic wave propagation,

while the solution of Maxwell equation is dened by Helmholtz equation. The

radiated power is given by Parsevals power theorem. Half-wave Fourier analysis is

used to determine the time domain behavior of antenna radiations. The magnitude

and phase of the radiated eld is given by Poynting vector (S = E H). The image

2.1 Introduction 13

extending ground plane to an isolated RDRA. Resonance in RDRA is created due

to formation of standing waves inside the device. Frequency xmnp is the spectral

solution of an antenna, and this can determine the base half-wave Fourier analysis.

Principle of orthonormality is used to determine radiation parameters by equating

electric time average energy equal to magnetic time average energy by KAM

(KalmogorovArnoldMoser).

At any instant of time, n number of modes exist. The particular mode can be

excited by increasing weighted amplitude of desired mode. More than one mode

can also be excited into RDRA. Blocking modes can take place if Ez and Hz elds

of same frequency are available in RDRA at any instant of time. Hence, mode

spectrum will result into corresponding resonant frequency generation. Wave

propagation can be dened by Helmholtz equation. The Maxwells equation

describes the behavior of electromagnetic elds and forms the basis of all EM

classical phenomena. The size of antenna can be reduced to half by image theorem,

converting isolated cavity into innite ground plane. Dielectric resonator antenna is

formed with high permittivity substrate. The abrupt change in permittivity due to

change in medium forms standing waves. These waves establish resonance as they

bounce back and forth in-between two walls due to elds perturbation. Modes are

spectral resolution of electromagnetic elds of waves radiated by the RDRA. Modal

excitation mainly depends on:

(a) Position of probe;

(b) Magnitude of probe current; and

(c) Phase of input current.

To compute resonant modes, vector principle of orthonormality on half-wave

Fourier analysis has been applied, i.e., radiated magnetic energies time averaged are

compared with applied electrical energies time averaged in the case of resonator

antennas. More number of modes along z-axis in RDRA can be generated either by

increasing electrical height d of RDRA or by increasing resonant frequency of

DRA. Figure 2.1 depicts the prototype RDRA with moat under fabrication.

14 2 Rectangular DRA Resonant Modes and Sources

modes are formed. The propagation constant c itself is taking discrete values. This

forces the natural frequencies of the eld oscillations to take discrete values (mnp)

indexed by three positive integers, namely m, n, and p. The solutions of the

waveguide problem yield discrete values of c, i.e., cm; n; x for a given frequency

x by applying boundary conditions to the electromagnetic elds on the sidewalls.

The corresponding eld amplitudes are solutions to the 2-D Helmholtz equations

corresponding to the transverse Laplacian r2? : These amplitudes are called the

waveguide modes and are of the form given below in Sect. 2.2.

Transverse electric and magnetic (TEM) mode

Transverse electric (TE) mode

Transverse magnetic (TM) mode

Hybrid electric and magnetic (HEM) or HE odd and EH even mode

Modes propagation depends mainly on following conguration:

1. Excitation

2. Dimensions

3. Coupling

4. Medium

5. Point of excitation

6. Input impedance

Cross-polarization solution can be the outcome of modes. They can be merged,

separated and mixed depending upon the requirements. Half Fourier analysis can be

used to describe modes of propagation and excitation. Even and odd modes can be

studied. They can be analyzed with magnetic dipole moments. They help to predict

far eld radiation patterns. Modulated bandwidth and gain control can be achieved.

High gain at higher modes can be used for hardware implementations. Device

dimensions can be minimized by proper selection of modes for resonant frequen-

cies. In case of milli metric (mm) wave, device size can be enlarged for easy

hardware development or hardware implementation. The solution is based on

waveguide method when boundaries have been all six electrical walls. The solution

is based on solution of Maxwells equations and then restricted to given boundary

conditions for conned modes of EM waves.

2.3 Solutions of Helmholtz Equation 15

@D

rH J 2:1

@t

D E

B lH r A

1

H r A

l

E Em ejxt 2:2a

H Hm ejxt 2:2b

Now,

@B

rE 2:3

@t

or

r E jxlH jxr A

r E jxA 0 2:4

r re 0

E jxA re

E jxA re

r r A r r A r 2 A 2:5

16 2 Rectangular DRA Resonant Modes and Sources

Now,

r lH rr A r2 A

or

lr H rr A r2 A

r H J jxE

or

B

r J jxE

l

or

r B lJ jxlE

or

r r A lJ jxlE

or

rr A r2 A lJ jxlE

rr A r2 A lJ jxljxA re

or

rr A r2 A lJ x2 lA jxlre

or

r2 A k2 A lJ jxlre rr A

or

r2 A k2 A lJ rjxle r A

where k2 x2 l.

2.3 Solutions of Helmholtz Equation 17

r A jxele

or

1

e r A

jxle

Hence, r2 A k2 A lJ:

Propagation in waveguide has been taken along z-axis, and all the four sidewalls of

waveguide are PEC; the elds computed are as follows:

Hx Ex

Hy Ey

Hz Ez

X

1 mpx npy

Ez x; y; z C m; n sin sin expcmn z; 2:7a

m;n1

a b

@Ez

Ez;y ;

dy

@Ez

Ez;x

dx

X

1 mpx npy

Hz x; y; z Dm; n cos cos expcmn z 2:7b

m;n1

a b

sides rst

18 2 Rectangular DRA Resonant Modes and Sources

i j k

@ @Ez @Ey @Ez @Ex @Ey @Ex

r E @x @

@y

@

@z i j k

Ex Ey Ez @y @z @x @z @x @y

2:9a

i j k

@ @Hz @Hy @Hz @Hx @Hy @Hx

r H @x @

@y

@

@z i j k

Hx Hy Hz @y @z @x @z @x @y

2:9b

Comparing with RHS in both equations and getting value of Hx, Hy, Hz from (2.9a)

and Ex, Ey, Ez from (2.9b) we get

1 @Ez @Ey 1 @Hz @Hy

Hx ; Ex ; 2:10a

jxl @y @z jx @y @z

1 @Ez @Ex 1 @Hz @Hx

Hy ; Ey ; 2:10b

jxl @x @z jx @x @z

1 @Ey @Ex 1 @Hy @Hx

Hz ; Ez ; 2:10c

jxl @x @y jx @x @y

@Hz

Hz;y

dy

@Hz

Hz;x

dx

Similarly,

jx c Ex Hz;y

c jxl Hy Ez;x

2.4 Rectangular Waveguide Analysis 19

and

jx x Ey Hz;x

c jxl Hx Ez;x

On manipulating them

Ey jxl c Hz;x

Hx c jx Ez;y

Hence, on simplication

jxl c

Ey Hz;x 2 Ez;y

h2m;n hm;n

mpx

npy

c z

X Dm; n np

b cos a sin b e

mn

Ex jxl 2

m;n

hm;n

mp

mpx

npy

c z

X C m; n

a cos a sin b e

mn

c 2

; 2:11

hm;n

mp

X Dm; n np

b cC m; n a mpx npy cmn z

jxl cos sin e

m;n

h2m;n a b

mp2 np2

h2m;n

a b

cmn ! propagation constant

Za Zb mpx npy

2

Eix m; n Eix x; y cos sin dxdy

ab a b

00 2:12a

np

mp

a

;

h2m;n

Za Zb mpx npy

2

Eiy m; n Eiy x; y sin cos dxdy

ab a b

0 0

! 2:12b

jxl mp c Cm; n np

m;n

2 Dm; n b

;

hm;n a h2m;n

20 2 Rectangular DRA Resonant Modes and Sources

#

Eix m; n a h2m;n b h2m;n Cm; n

Eiy m; n c np

hm;n

2 b hjxl

2

mp Dm; n

m;n m;n a

boundary conditions are given as:

8 9

< x 0; a =

y 0; b

: ;

z 0; d

np

mp

h mpx npyi

X jxl Dm; n cm;n Cm; n

Eix x; y b a

cos sin ;

m;n

h2m;n a b

2:13a

mp

np

h mpx npyi

X jxl Dm; n cm;n Cm; n

Eiy x; y a b

sin cos :

m;n

h2m;n a b

2:13b

2 0 2:14

@x2 @y2 c @t2

w0; y; t wa; y; t 0

wx; 0; t wx; b; t 0

@2w @ @w @ @w

rdxdy T dy dx T dx dy 2:15

@t2 @x @x @y @y

Y 00 X 00

ky2 ; kx2 ;

Y X

2.5 Two Dimensional Resonator 21

@2w

c2 r 2 w 0 2:16

@t2

wx; y; t X xY yT t 2:17

00

T 00 t 2 X x Y 00 y

x

2

c

T t X x Y y

let

X x sinkx x

Y y sin ky y

x2

kx2 ky2

c2

mp np

kx ; ky 2:18

a b

q

Frequency can be written as: xmn cp ma2 nb2 :

Modes

r2 Ax k2 Ax 0; 2:19

kr 1 near eld pattern

where Az is the magnetic vector potential and k is the wave vector or wave number

along z-axis.

C4 sinky yC5 coskz z C6 sinkz z;

22 2 Rectangular DRA Resonant Modes and Sources

Z

l e jkR 3 0

Az J z0 d z; 2:20

4p R

2p q

kc ; k0 kx2 ky2 kz2 x2 l; where k is the wave number:

k

The wave number can be dened as rate of change of phase w.r.t. distance in the

direction of propagation. Resonant frequency x xmnp in RDRA and its

mathematical expression is given below:

r

mp2 np2 pp2

c

fr m; n; p p ; 2:21

2p

er a b d

H rA

dA

E r ; scalar and magnetic vector potential from Lorentz Gauge

dt

conditions:

S E H ; S is Poynting vector energy flow or flux:

Prad

Z 2 input Impedance:

jIj

Ex ; E y ; E z ; H x ; H y ; H z are electric and magnetic fields

I n npx npxo I n

mpy mpyo

L cos ; sin L cos ; sin ; 2:22

a a b b

where L denotes linear components. It turns out that depending on the nature of

wall surfaces (PEC or PMC), four possible linear combinations can appear

(cos sin; sin cos; sin sin; and cos cos).

In rectangular DRA, weve got to applying in additional boundary conditions on

top and bottom surfaces to be the linear combinations as compared to waveguide.

d ; when p 1; 2; 3. . . and have two possible

ppz

ppz

Thus, the possible frequencies x obtained by solving cm; n; x pp d ; then

comes out to be:

2 1=2

m n 2 p2

xm; n; p p 2 2 2 : 2:23

a b d

2.6 Basic Mathematical Representation of Resonant Modes 23

@

physics to resonator physics is to just replace c by @z in all the waveguide

formulae that express the tangential eld components in terms of the longitudinal

components. This is done after solving the full 3-D Helmholtz equations using

separation of variable in x, y, z.

x2 Ez

r2 2 0 2:24

c Hz

C oscillators with different L, C values. The outcome of all this analysis enables us

to write down the E and H elds inside the resonator, as superposition of four, three

vector-valued basis functions.

X

1 n o

Ex; y; z; t Re Cmnp e jxmnpt wEmnp x; y; z

mnp1

n o 2:25

X

1

E x; y; z ;

Re Dmnp e jxmnpt / mnp

mnp1

and

X

1 n o

H x; y; z; t Re Cmnpe jxmnpt wH

mnp

x; y; z

mnp1

n o 2:26

X

1

Re H x; y; z

Dmnpe jxmnpt / ;

mnp

mnp1

We note that there are only two sets fCmnp g and fDmnp g of linear combination of

coefcients from the Ez and Hz expansions. The vector-valued complex functions

E wH /

are wEmnp , / H R3 (where R is autocorrelation) and contains compo-

mnp mnp mnp

nents fcos; sing fcos; sing fcos; sing; functions and hence for m0 n0 p0 6

mnp; each function of the set:

n o

E ; wH ; /

wEmnp ; / H ;

mnp mnp mnp

n E H

o

wEm0 n0 p ; /m0 n0 p ; wH

mnp

; / m0 n 0 p

;

24 2 Rectangular DRA Resonant Modes and Sources

The exact form of the function / E; / H ; wE ; wH depends on the nature of the

boundaries. The next problem addressed can be on excitations of RDRA. To cal-

culate the amplitudes coefcients {Cmnp} and {Dmnp}, we assume that at z = 0,

e e

excitations Ex x; y; t or Ey x; y; t are applied for some time say t 2 [0, T] and

then removed. Then, the Fourier components in this excitation corresponding to the

frequencies xfmnpg are excited and their solutions are the oscillations, while the

waveguide for t > T. The other Fourier components decay within the resonator.

{Cmnp, Dmnp} are the components of the form:

X

Exe x; y; t Re Cmnpe jxmnpt wEmnp x x; y; 0

mnp

2:27

Re E x; y; 0

Dmnpe jxmnpt / mnp x

and

X

Eye x; y; t Re Cmnpe jxmnpt wEmnp y x; y; 0

mnp

2:28

Re E x; y; 0

Dmnpe jxmnpt / ;

mnp y

By using, orthogonality of fwEmnp x x; y; 0; / mnp x

E x; y; 0g:

we write p to be xed and likewise of fwEmnp y x; y; 0; / mnp y

In addition, we need to use KAM type of time averaging to yield eld

components:

C mnpwEmnp x x; y; 0 Dmnp/ E x; y; 0

mnp x

ZT

lim 1

Exe x; y; tejxmnpt dt: 2:29

T ! 1 2T

T

and likewise

C mnpwEmnp y x; y; 0 Dmnp/ E x; y; 0

mnp y

ZT

lim 1

Eye x; y; te jxmnpt dt: 2:30

T ! 1 2T

T

2.7 Voltage Source Model 25

source to an LC circuit for sometimes and then switching it off. After a sufciently

long time, all frequencies in the LC circuit decay away except the frequency p1

LC

.

We can more generally compare a resonator with the material medium having

non-zero conductivity. Thus, the medium is characterized by the triplet (; b; r

which corresponds to an array of (C, L, R) = RLC circuits.

Such a resonator is analyzed in the same way replacing by 0 jr x , i.e.,

complex permittivity depending on frequency. The resonant frequencies xmnp

now have a non-zero imaginary part corresponding to decay of the eld with time.

Their frequencies and elds may also be determined by applying separation of

variables with boundary conditions to the Helmholtz equations.

E

r2 jxlr jx z 0; 2:31

Hz

To have sustained oscillations in such a resonator, we must never switch off the

excitation. We may for example apply a surface current source at z d0 , where

0\d0

d: Letting Jsx x; y; x and Jsy x; y; x be this surface current excitations in

the Fourier domain, the current density corresponds to this is given as:

^ Jsy x; y; xY^ dz d0 ;

J e x; y; z; x Jsx x; y; xX 2:32

Curl E jxl H;

homogeneous equations, i.e., with J e = 0 and a particular solutions for J e 6 0. The

general solutions to the homogeneous problem are the same as earlier explained,

i.e., containing only the frequencies fxmnpg. Particular solution to the

J e 6 0 inhomogenous problem is obtained by taking the curl of the second

equation and substituting the elds into third equation to obtain

26 2 Rectangular DRA Resonant Modes and Sources

X

1

L1 m; n; x expcmnxzumn x; y; x

m;n1 2:34a

b1 m; n; x expcmnxzvmn x; y; x for d z [ d; 2:14

X

1

H p x; y; z; x L2 m; n; x exp cmnxzumn x; y; x

m;n1 2:34b

b2 m; n; x expcmnxzvmn x; y; x for 0 z\d; 2:15

are multiples (x-dependent) of

n mpx npxo

mpy mpy

L cos ; sin L cos ; sin

a a p p

To meet the boundary conditions on the sidewalls, if z = 0, d; if the walls are PEC,

Hpz 0; when z = 0, d. That gives use

X

Hpz x; y; z; x Lm; n; x sin hfcm; n; xz d gumnz x; y; x; d\z d;

m;n

2:35

and

X

Hpz x; y; z; x bm; n; x sin hfcm; n; xzgumnz x; y; x; 0 z\d;

m;n

2:36

The elds Hp? x; y; z; x are easily determined from these equations in the

region z > d and z < d by differentiating them w.r.t. x, y, z; wherever c comes in the

@

multiple w.r.t. expcz, we replace it by @z etc.

In this way, we get

X

1

Hpx x; y; z; x L1 m; n; xwmnx x; y; z; x; for z [ d; 2:37a

m;n1

and

X

1

Hpy x; y; z; x L2 m; n; xwmny x; y; z; x; for z\d; 2:37b

m;n1

2.7 Voltage Source Model 27

X

1

Hpx x; y; z; x

bm; n; x/

E

x; y; z; x; 2:38a

mnp x

m;n1

and

X

1

Hpy x; y; z; x

bm; n; x/

E

x; y; z; x; 2:38b

mnp y

m;n1

The coefcients Lm; n; x and bm; n; x are obtained from the boundary

conditions

^ Jsy x; y; xY:

J Jsx x; y; xX ^ 2:39

The Fig. 2.2 presents how the generated modes look like. This will be able to tell us

the number of resonant modes in particular direction. The transverse components of

EM waves are expressed as Ex ; Ey ; Hx ; Hy : If propagation of wave is along

z-direction, Ez; Hz elds are the longitudinal components. These elds are modal

solutions, solved based on Helmholtz equations using standard boundary condi-

tions. The RDRA is basically a boundary value problem. The linear combinations

of sine and cosine terms give rise to TE and TM modes. The generation of various

kinds of modes in an antenna and propagation is very critical issue; it need through

study. Now, rewriting Helmholtz equation for source-free medium (Fig. 2.3)

28 2 Rectangular DRA Resonant Modes and Sources

generated in RDRA by HFSS.

x x x

a TEd12 , b TEd14 , c TEd16

modes

r2 W k2 W 0;

1 @ 2

W kx2 2:40a

Wx @x

1 @ 2

W ky2 2:40b

Wy @y

1 @ 2

W kz2 2:40c

Wz @z

W or Hz or Ez A sin kx x B cos kx x C sin ky y D cos ky y ejkz z

Xn mpx npyo

Hz C mn cos cos e jkz z ; C mn Fourier Coefficients; 2:41a

m;n

a b

Xn mpx npyo

Ez Dmn sin sin e jkz z ; Dmn Fourier Coefficients; 2:41b

m;n

a b

mp2 np2

c2 x2 l kx2 ky2

a b

mp 2 np 2

kz2 x2 l

a b

2.8 Resonant Modes Generation 29

mp 2 np 2

x2 l [0

a b

s

1 mp2 np2

xc p

l a b

It means, waveguide will support all waves having greater than xc to propagate.

Now, rewriting Hz and Ez

Xn mpx npyo

Hz C mn cos cos e jkz z 2:42

m;n

a b

Xn mpx npyo

Ez Dmn sin sin e jkz z 2:43

m;n

a b

Here, Cmn and Dmn are the coefcients of Fourier cosine and sine series.

q

cm;n h2m;n x2 le

Hence, C mn and Dmn gives us relative amplitudes and phase. Hence, we get solution

of possible amplitudes and phase of wave propagating through rectangular wave-

guide called as modes of propagation.

Results of resonant frequency obtained on various sizes RDRAs using HFSS have

been placed in Table 2.1. The MATLAB programs are being developed for modes

graphical view. Resonant modes and resonant frequencies are being obtained based

on formulations. The programs and simulated results are given below:

30 2 Rectangular DRA Resonant Modes and Sources

S. No. Permittivity Dimension (a b h) mm Resonant frequency

1 10.0 14.3 25.4 26.1 3.5

2 10.0 14 8 8 5.5

3 10.0 15.24 3.1 7.62 6.21

4 20.0 10.2 10.2 7.89 4.635

5 20.0 10.16 10.2 7.11 4.71

6 35.0 18 18 6 2.532

7 35.0 18 18 9 2.45

8 100.0 10 10 1 7.97

The graph shown in Fig. 2.4 represents inverse relationship between height and

resonant frequency as k-wavelength is inversely proportional to resonant frequency

fr. MATLAB simulation shown in Fig. 2.5 represents number of modes generated

in x, y, z directions. The mathematical expression on the topic is expressed in

Eqs. (2.1)(2.31).

2.9 MATLAB Simulated Results 31

1

0.8

0.6

0.4

0.2

0

-0.2 z

-0.4

-0.6

-0.8

-1

3

2

1 x 4 5

y 3

0 1

2

-1 0

-2 -1

-2 -3

-3 -5 -4

TE7,10,6

32 2 Rectangular DRA Resonant Modes and Sources

Reference

1. Okaya A, Barash LF (1962) The dielectric microwave resonator. Proc IRE 50:20812092

Chapter 3

Mathematical Analysis of Rectangular

DRA

gular dielectric resonator antenna (RDRA) has been introduced. The investigations

are based on the rst applying waveguide theory, then converting it to resonator by

replacing c to d/dz. Initially, these elds are exploited using the Maxwell curl

equations, then manipulating them to express the transverse components of the

elds in terms of partial derivatives of the longitudinal components of the elds

with respect to x and y axis (i.e., the transverse coordinates). Waveguide models of

four rectangular DRAs with specied boundary conditions with linear permittivity

have been realized.

Waveguide Homogeneous medium Boundary conditions Surface interface

dielectric resonator antenna (RDRA) has been introduced. The investigations are

based on rst applying waveguide theory, then converting it to resonator by

replacing c to d/dz. Through out this book, electromagnetic eld propagation has

been taken along z-axis, i.e., exp(z). Initially, these elds are exploited using the

Maxwell curl equations, then manipulating them to express the transverse com-

ponents of the elds in terms of partial derivatives of the longitudinal components

of the elds with respect to x and y axis (i.e., the transverse coordinates).

Waveguide models of four different rectangular DRAs with specied boundary

conditions with homogeneous material having linear permittivity have been

mathematically modeled. The elds are realized to determine TE and TM modes of

propagating elds. These have resulted into different sinecosine combinations.

Propagation of these elds have been split as inside the RDRA and outside RDRA.

The interfacing surface is having two different dielectrics. The solution is developed

as transcendental equation, which purely characterized rectangular DRA frequency

and propagating elds in terms of propagation constants and dominant resonant

frequency. TE modes generation required Hz as longitudinal elds and Ex, Ey, Hx,

and Hy as transverse elds. Excitation is applied along x-axis as partial elds, y-axis

will have xed variation, and z-axis will have desired variation in propagating

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_3

34 3 Mathematical Analysis of Rectangular DRA

elds, for example, TE d13 and TE d43 . Similar cases can be developed for other

modes, so as to propagate Ez elds as longitudinal and Ex, Ey, Hx, and Hy as

transverse elds. In this case, Hz shall get vanished because of boundary conditions.

Resonant modes, i.e., amplitude coefcient of these elds Cmnp and Dmnp inside the

DRA can be determined by comparing magnetic energies equal to electrical ener-

gies based on principle orthonormality or law of conservation. The derivation for

the quality factor and radiation pattern have been developed for deeper antenna

analysis.

In Rectangular DRA as shown in Fig. 3.1, top and bottom walls of RDRA are PMC

and rest of the other walls are PEC. On magnetic walls (PMC), n E = 0, where

E denotes the electric eld intensity and n denotes the normal to the surface of

the resonator. Similarly, n H = 0 is not necessarily satised at all the surfaces of the

DRA by all the modes. Different resonant modes shall have different electromagnetic

Fig. 3.1 a Rectangular DRA with aperture-coupled feed. b RDRA with input excitation

3.1 Rectangular DRA with Homogeneous Medium 35

eld distribution inside the RDRA, and each mode may provide a different resonant

frequency and radiation pattern, i.e., eigen vector and eigen frequency.

Excitation-based resonant modes can generate desired radiation pattern for different

coverage requirements. By making use of this mechanism, internal as well as

associated external elds distribution can be obtained.

Rectangular DRA is better choice due to flexible aspect ratio, i.e., b/a or d/

a options can generate different modes. The existence of two independent aspect

ratios in a rectangular DRA offers better design flexibility. Assuming the ground

plane to be innitely large, image theory is applied to replace the isolated RDRA by

a grounded resonator of half-size. In this RDRA, two of the six surfaces of the

resonator are assumed to be perfect magnetic walls, while the remaining four are

assumed to be perfect electric walls. Electromagnetic theory is then applied to study

its theoretical analysis, and later three more cases have been developed based on the

different boundary conditions. For example, the elds undergo one half-wave

variation along the dimension a and remains constant along dimension b. They

undergo less than a half-wave variation along z-axis, i.e., variation along DRA

height d. The resonant mode is therefore identied as TEz10 d. The propagation

direction has been assumed in z-direction. TEz310 resonant elds undergo three

half-wave variations along length of DRA a and one half-wave variation along

breadth b, and no variation along height d. To adapt these formulae to an DRA,

we note that the propagation constants along z can be c with the linear combi-

nations of coefcients chosen, so as to meet the boundary conditions at z = 0 and

z = d, i.e., the top and bottom surfaces of the RDRA, which have been taken as PEC

(permanent electrical conducting) walls. On a PEC, the tangential components

(n E = 0) of the electric eld and the normal component (n H = 0) of the

magnetic elds get vanished. While on a PMC wall, by directly, the normal

component of the electric eld (n E = 0) and the tangential components

(n H = 0) of the magnetic eld get vanished.

To compute resonant modes, vector principle of orthonormality on half-wave

Fourier analysis has been applied, i.e., radiated magnetic energies are compared

with applied electrical energies in RDRA. More number of modes along z-axis in

RDRA can be generated either by increasing electrical height d of RDRA or by

increasing excitation resonant frequency. Given below are the two rectangular

DRAs with different congurations shown in Fig. 3.1.

In Fig. 3.1, PMC and PEC walls conguration is labeled. The mathematical

solution is developed based on this conguration. The boundary conditions of

interface walls shall form linear combinations of sinecosine terms. Accordingly,

they will decide whether transverse electric elds or magnetic elds will vanish.

Propagation of longitudinal elds shall depend on the direction of excitation.

Excitation of resonant modes in rectangular boundaries are easier as compared to

cylindrical. Transcendental equation and characteristics equations have been

developed for rectangular DRA. This has provided complete solution of resonant

frequency and propagation constants.

36 3 Mathematical Analysis of Rectangular DRA

In this chapter, four different solutions are presented, each RDRA is associated with

different boundaries. The resultant eld formed the resonant modes of different

kinds.

Figure 3.2 described E and H elds pattern forming resonant modes, i.e.,

dominant or higher-order excited modes inside the RDRA.

3.2.1 Model-1

(a) Here, top and bottom walls are assumed as PMC and rest of the other four walls

are PEC as per Fig. 3.1.

Given top and bottom surfaces of RDRA as PMC at z 0; d;

)nH 0

n E 0;

Hy Hx 0;

Ez 0;

n E 0;

n H 0;

x 0; a;

Ey Ez 0;

Hx 0;

3.2 Rectangular DRA Mathematical Modeling 37

At,

y 0; b;

Ex Ez 0;

Hy 0;

1 @Hz 1 @ 2 Ez

Ex 3:1

2

jx 1 kc2 @y jxl @z@x

1 1 @ 2 Ez @Hz

Ey 3:2

2

jx 1 kc2 jxl @z@y @x

1 @Ez 1 @ 2 Hz

Hx 3:3

2

jxl 1 kc2 @y jx @z@x

1 1 @ 2 Hz @Ez

Hy 3:4

jxl 1 kc2 jx @z@y @x

2

wz X xY yZz

where

TE mode Ez 0 and H z 6 0

wHz X xY yZz

0@Hz

Ey C ;

@x

38 3 Mathematical Analysis of Rectangular DRA

or,

Ey C0 X 0 xY yZ z;

Now

X 0 x A1 cos kx x A2 sin kx x;

But at

x 0; a; Ey 0;

) 0 A1 cos kx 0 A2 sin kx 0;

or,

mp

A1 0 and kx ;

a

@Hz0

Ex C ;

@y

or,

Ex C0 X xY 0 yZz:

Now

Y 0 y A3 cos ky y A4 sin ky y;

At,

y 0; b; Ex 0;

) 0 A3 cos ky 0 A4 sin ky 0;

or,

np

A3 0 and ky ;

b

3.2 Rectangular DRA Mathematical Modeling 39

1 @ 2 Hz

0

Hx C ;

jx @z@x

or,

Hx C0 X 0 xY yZ 0 z;

Now

Z 0 z A5 cos kz z A6 sin kz z;

At,

z 0; d Hx 0;

) A5 cos kz 0 A6 sin kz 0 0;

pp

A5 0 and kz ;

d

Hence,

mp np pp

Hz A2 A4 A6 cos x cos y cos z 3:8

a b d

mppp mp np pp

Hx C 00 A2 A4 A6 sin x cos y sin z ;

a d a b d

np pp mp np pp

Hy C 00 A2 A4 A6 cos x sin y sin z ;

b d

mp a b d

mp np pp

Ey C 00 A2 A4 A6 sin x cos y cos z ;

a a

np mp np pp b d

Ex C 00 A2 A4 A6 cos x sin y cos z ;

b a b d

wEz X xY yZ z;

40 3 Mathematical Analysis of Rectangular DRA

At,

x 0; a;

Ez 0;

A1 sin kx 0 A2 cos kx 0 0;

mp

) A2 0 and kx ;

a

Also, at

y 0; b;

Ez 0;

A3 sin ky 0 A4 cos ky 0 0;

np

A4 0; and ky

b

At,

z 0; d;

Ez 0;

) A5 sin kz 0 A6 cos kz 0 0;

pp

A6 0 and kz ;

d

Hence,

mp np pp

Ez A1 A3 A5 sin x sin y sin z 3:9

a b d

mppp mp np pp

Ex C 00 A1 A3 A5 cos x sin y cos z ;

a d a b d

np pp mp np pp

Ey C 00 A1 A3 A5 sin x cos y cos z ;

b d

mp a b d

mp np pp

Hy C 00 A1 A3 A5 cos x sin y sin z ;

a a

np mp np pp b d

Hx C 00 A1 A3 A5 sin x cos y sin z ;

b a b d

3.2 Rectangular DRA Mathematical Modeling 41

3.2.2 Model-2

(b) Top and bottom walls are PEC and rest of the other walls are PMC:

Assuming the top and bottom surface plane be at z 0; d;

n E 0;

n H 0;

Ey Ex 0;

Hz 0;

n H 0;

n E 0;

At,

x 0; a;

Hy Hz 0;

Ex 0;

At,

y 0; b;

Hx Hz 0;

Ey 0;

1 @Hz 1 @ 2 Ez

Ex ; 3:10

2

jx 1 kc2 @y jxl @z@x

1 1 @ 2 Ez @Hz

Ey ; 3:11

2

jx 1 kc2 jxl @z@y @x

1 @Ez 1 @ 2 Hz

Hx ; 3:12

2

jxl 1 kc2 @y jx @z@x

42 3 Mathematical Analysis of Rectangular DRA

1 1 @ 2 Hz @Ez

Hy ; 3:13

jxl 1 kc2 jx @z@y @x

2

wz X xY yZz;

where

TE mode Ez 0 and H z 6 0

wH z X xY yZ z;

At,

x 0; a;

Hz 0;

A1 sin kx 0 A2 cos kx 0 0;

A2 0;

and

mp

kx ;

a

Also, at

y 0; b;

Hz 0;

A3 sin ky 0 A4 cos ky 0 0;

A4 0;

and

np

ky ;

b

3.2 Rectangular DRA Mathematical Modeling 43

At,

z 0; d;

Hz 0;

A5 sin kz 0 A6 cos kz 0 0;

A6 0;

and

pp

kz ;

d

Hence,

mp np pp

Hz A1 A3 A5 sin x sin y sin z ; 3:17

a b d

mppp mp np pp

Hx C 00 A1 A3 A5 cos x sin y cos z ;

a d a b d

np pp mp np pp

Hy C 00 A1 A3 A5 sin x cos y cos z ;

b d

mp a b d

mp np pp

Ey C 00 A1 A3 A5 cos x sin y sin z ;

a a

np mp np pp b d

Ex C 00 A1 A3 A5 sin x cos y sin z ;

b a b d

TM mode H z 0 and Ez 6 0

wEz X xY yZ z;

1 @ 2 Ez

0

Ey C ;

jxl @z@y

Ey C0 X xY 0 yZ 0 z;

Now

Y 0 y A3 cos ky y A4 sin ky y;

44 3 Mathematical Analysis of Rectangular DRA

At,

y 0; b; Ey 0;

0 A3 cos ky 0 A4 sin ky 0;

np

A3 0 and ky ;

b

01 @ 2 Ez

Ex C ;

jxl @z@x

Ex C 0 X 0 xY yZ 0 z;

X 0 x A1 cos kx x A2 sin kx x;

x 0; a;

Ex 0;

0 A1 cos kx 0 A2 sin kx 0;

A1 0;

and

mp

kx ;

a

Z 0 z A5 cos kz z A6 sin kz z;

At,

z 0; d; Ex 0;

) 0 A5 cos kz 0 A6 sin kz 0;

or,

pp

A5 0 and kz ;

d

Hence,

mp np pp

Ez A2 A4 A6 cos x cos y cos z ; 3:18

a b d

3.2 Rectangular DRA Mathematical Modeling 45

mppp mp np pp

Ex A2 A4 A6 sin x cos y sin z ;

a d a b d

np pp mp np pp

Ey A2 A4 A6 cos x sin y sin z ;

np mp np pp d

b d a b

Hx A2 A4 A6 cos x sin y cos z ;

b a b d

mp mp np pp

Hy A2 A4 A6 sin x cos y cos z ;

a a b d

3.2.3 Model-3

(c) Solution of RDRA, when all six walls are PEC (perfect electrical walls):

Using Maxwell equations:

r E jxB jxlH;

r H jxE;

r E jxlH;

^x ^y ^z

@ @ @

@x @y

@z jxlH;

Ex Ey Ez

@Ez @Ey @Ex @Ez @Ey @Ex

^x ^y ^z jxlH;

@y @z @z @x @x @y

@Ez @Ey

jxlHx ; 3:19

@y @z

@Ex @Ez

jxlHy ; 3:20

@z @x

@Ey @Ex

jxlHz ; 3:21

@x @y

@Hz @Hy

jxEx ; 3:22

@y @z

46 3 Mathematical Analysis of Rectangular DRA

@Hx @Hz

jxEy ; 3:23

@z @x

@Hy @Hx

jxEz ; 3:24

@x @y

1 @Hz 1 @ 2 Ex @ 2 Ez

Ex ; 3:25

jx @y jxl @z2 @x@z

1 @ 2 Ex 1 @Hz 1 @ 2 Ez

Ex 2 2 ; 3:26

k @z jx @y jxl @x@z

c2 1 @Hz c @Ez

Ex 1 2 ;

k jx @y jxl @x

c2 1 c @Ez @Hz

Ey 1 2 ;

k jx jxl @y @x

c2 1 @Ez c @Hz

Hx 1 2 ;

k jxl @y jx @x

c2 1 c @Hz @Ez

Hy 1 2

k jxl jxl @y @x

w X xY yZ z;

TM mode of propagation, H z 0;

Boundary conditions

Electrical walls ! Etan 0 n E;

! Hn 0 n H;

At, x = 0;

y 0; Y y Ay cos0; A4 must be zero:

3.2 Rectangular DRA Mathematical Modeling 47

Therefore,

@

Z z 0;

@z

A5 coskz z A6 sinkz z 0;

Therefore, at

z 0; d;

A5 must be zero;

At,

x a;

X x A1 sin kx a 0;

mp

kx ;

a

At,

Y b; Y y A2 sin ky d 0;

np

ky ;

b

At,

z 0 zz A4 sin kz d 0;

pp

kz ;

d

48 3 Mathematical Analysis of Rectangular DRA

1 @Hz 1 @ 2 Ez

Ex ;

jx 1 ky2

2

@y jxl @x@z

Ez 0;

1

y2

1 k2

k2

1 @Hz 1 @ 2 Ez

Ey ;

2

jx 1 ky2 @x jxl @y@z

A1 A3 A5 ky

y2

k 1 k2

2

1 @Ez 1 @ 2 Hz

Hx ;

2

jxl 1 ky2 @y jx @x@z

ky A1 A3 A5

2

sin kx x cos ky y cos kz z ;

x2 l 1 k2

y

1 1 @ 2 Hz @Ez

Hy ;

2

jxl 1 ky2 jx @y@z @x

kz kx A1 A3 A5

2

cos kx x sin ky y cos kz z ;

x2 l 1 ky2

kx A1 A3 A5

y2

jxl 1 k2

For, TE mode

For PEC walls, electric eld components are assumed to be varying with Hz in

direction of (x, y, z)

3.2 Rectangular DRA Mathematical Modeling 49

@

Ex C 0 Hz

@y

C0 X xY 0 yZ z;

y 0; b;

Y 0 y A3 cos ky y A4 sin ky y 0;

A3 0;

np

ky ;

b

@

Similarly, Ey C00 Hz ;

@x

A1 0;

mp

kx ;

a

Z z A5 sin kz z A6 cos kz z;

At,

z 0; d;

A6 0;

pp

kz ;

d

Hz A2 A4 A5 cos kx x cos ky y sin kz z;

Therefore,

1

Ex 2

A2 A4 A5 cos kx x sin ky sin kz;

jx 1 ky2

A2 A4 A5 kx

Ey 2

sin kx x cos ky y sin kz z ;

jx 1 ky2

kx kz A1 A3 A5

Hx 2

sin kx x cos ky y cos kz z ;

k2 1 k2 y

kz ky A1 A3 A5

Hy 2

cos kx x sin ky y cos kz z ;

x2 l 1 ky2

50 3 Mathematical Analysis of Rectangular DRA

3.2.4 Model-4

(d) When all the six walls of RDRA are assumed to be PMC (permanent magnetic

walls),

wz X xY yZ z where wz is wave function in x, y, and z direction as space.

3:27

where A1A6 are constants and (A1 sin kx x + A2 cos kx x is solution of second-order

differential equation in x direction, i.e., X(x).

When all six walls are PMC, the rectangular DRA solution is

Htan n H 0;

Hnor n E 0;

Applying boundaries,

At,

x 0; a ) Hy and Hz 0; Ex 0;

At,

y 0; b ) Hx and Hz 0; Ey 0;

At,

z 0; d ) Hx and Hy 0; Ez 0;

TE mode of propagation Ez 0; H z 6 0

Using boundary conditions

At,

mp

x 0; a; Hz 0 ) A2 0 and kx ;

a

At,

np

y 0; b; Hz 0 ) A4 0 and ky ;

b

3.2 Rectangular DRA Mathematical Modeling 51

Now,

@ 2 Hz

Hx C 00 C 00 X 0 xY yz0 z

@x@z

z0 z A5 cos kz z A6 sin kz z

At,

z 0; d ) d ) Hx 0;

pp

) Hx 0 ) A5 0; kz ;

d

Hence,

mpx npy ppz

Hz A1 A3 A6 sin sin cos 3:28

a b d

TM mode of propagation Ez 6 0; H z 0

We again look for the conditions, when Hz = 0, i.e., to get the value of Ez

C 0 @Ez

Hz

@y

C X xY y0 Z z;

0

Hx 0 at Y 0; b;

np

) A3 0 at ky ;

b

Similarly,

@Ez

Hy C 00 ;

@x

C 00 X 0 xY y0 Z z;

X 0 x A1 cos kx x A2 sin kx x;

) Hy 0 at x 0; a;

) A1 0;

mp

kx ;

a

52 3 Mathematical Analysis of Rectangular DRA

At,

z 0; d ) Ez 0;

pp

) A5 0 and kz ;

d

At,

z 0; d ) Ez 0;

pp

) A5 0 and kz ;

d

Hence,

mpx npy ppz

Ez A2 A4 A5 cos cos sin : 3:30

a b d

modes are formed. The propagation constant c itself is taking discrete values. This

forces the natural frequencies of the eld oscillations to take discrete values mnp,

indexed by three positive integers m, n, and p. The solutions of the waveguide

problem yield discrete values of c, i.e., cm; n; x for a given frequency x by

applying boundary conditions to the electromagnetic elds on the side walls. The

corresponding eld amplitudes are solutions to the 2-D Helmholtz equations

corresponding to the transverse Laplacian r2? : These amplitudes are called the

waveguide modes and are of the form

I n npx nnpxoo I n

mpy nmpyoo

L cos ; sin L cos ; sin

a a b b

where L denotes linear components. It turns out that, depending on the nature of

wall surfaces (PEC or PMC), four possible linear combinations can appear

cos sin; sin cos; sin sin; and cos cos.

In rectangular DRA, we have got to applying in additional boundary conditions

on top and bottom surfaces to be the linear combinations as compared to the

waveguide.

d ; when p 1; 2; 3. . . and have two possible

ppz

d and cos d .

3.2 Rectangular DRA Mathematical Modeling 53

d and then

comes out to be

2 1=2

m n2 p2

xm; n; p p 2 2 2

a b d

@

physics to resonator physics is to just replace c by @z in all the waveguide

formulae that express the tangential eld components in terms of the longitudinal

components. This is done after solving the full 3-D Helmholtz equations using

separation of variable in x, y, and z.

x2 Ez

r 2

2

0

c Hz

L, C oscillators with different L, C values. The outcome of all this analysis enables

us to write down the E and H elds inside the resonator, as superposition of four or

three vector-valued basis functions.

X

1 n o

Ex; y; z; t Re Cmnpe jxmnpt wEmnp x; y; z

m;n;p1

n o 3:31

X

1

E x; y; z

Re Dmnpe jxmnpt / mnp

m;n;p1

and

X

1 n o

H x; y; z; t Re Cmnpe jxmnpt wH

mnp

x; y; z

m;n;p1

n o 3:32

X

1

H x; y; z

Re Dmnpe jxmnpt / mnp

m;n;p1

We note that there are only two sets of amplitude coefcients {C(mnp)} and

{D(mnp)} of linear combination of coefcients using from the Ez and Hz expansions.

The vector-valued complex functions are wEmnp ; / E ; wH ; / H R3 (where R is

mnp mnp mnp

autocorrelation) and contains components fcos; sing fcos; sing fcos; sing;

functions and hence for m0 n0 p0 6 mnp; each function of the set

n o

E ; wH ; /

wEmnp ; / H ;

mnp mnp mnp

54 3 Mathematical Analysis of Rectangular DRA

n o

E ; wH ; /

wEm0 n0 p ; / H ;

0

mnp0 mnp 0

mnp0 0

The exact form of the function / E; /

H ; wE ; wH depends on the nature of the

boundaries. The next problem addressed can be on excitations of RDRA. To cal-

culate the amplitude coefcients {C(mnp)} and {D(mnp)}, we assume that at z = 0,

e e

an excitation Ex x; y; t or Ey x; y; t is applied for some time say t 2 [0, T] and

then removed. Then, the Fourier components in this excitation corresponding to the

frequencies fxmnpg are excited and their solutions are the oscillations, while the

waveguide for t > T. The other Fourier components decay within the resonator.

{C(mnp), D(mnp)} are components of the form,

X

Exe x; y; t ReCmnpe jxmnpt wEmnp x x; y; 0

m;n;p

n o 3:33

E x; y; 0

Re Dmnpe jxmnpt / mnp x

and

X

Eye x; y; t Re Cmnpe jxmnpt wEmnp y x; y; 0

m;n;p

n o 3:34

E x; y; 0

Re Dmnpe jxmnpt / mnp y

n o

E x; y; 0 , for different (m, n), we

By using orthogonality of wEmnp x x; y; 0; /

n mnp x o

E x; y; 0 ;

can write p to be xed and likewise of wEmnp y x; y; 0; / mnp y

In addition, we need to use KAM (KolmogorovArnoldMoser) type of time

averaging to yield

E x; y; 0

CmnpwEmnp x x; y; 0 Dmnp/ mnp x

ZT

lim 1

Exe x; y; tejxmnpt dt

T!1 2T

T

3.2 Rectangular DRA Mathematical Modeling 55

and likewise

CmnpwEmnp y x; y; 0 Dmnp/

E

x; y; 0

mnp y

ZT

lim 1

Eye x; y; te jxmnpt dt

T ! 1 2T

T

Chapter 4

Mathematical Analysis of Transcendental

Equation in Rectangular DRA

has been derived. Transcendental equation of rectangular DRA provides complete

solution of propagation constants, i.e., kx, ky, and kz. The propagation constant gives

rise to resonant frequency with the help of characteristic equation. The wave

numbers kx, ky, and kz are in x, y, and z direction, respectively. The free space wave

number is k0. The exact solution of RDRA resonant frequency can be determined

from combined solution of transcendental equation and characteristic equation.

These equations have unique solution. RDRA depends upon boundary conditions.

MATLAB-based simulation has been worked for RDRAs. They have been depicted

with examples. This chapter has given a complete design solution of rectangular

DRAs.

Keywords Mathematical analysis Transcendental equation Rectangular DRA

Propagation constant Eigen vectors Effective electrical length Characteristic

equation

agation constants, i.e., kx, ky, and kz. The propagation constant gives rise to resonant

frequency with the help of characteristic equation. The wave numbers kx, ky, and kz

are in x, y, and z direction, respectively. The free space wave number is k0. The

exact solution of RDRA resonant frequency can be determined from combined

solution of transcendental equation and characteristic equation. These equations

have unique solution if RDRA boundary conditions are xed. For example, top and

bottom walls are PMC and rest of the four walls is PEC and vice versa, only two

different transcendental equations will be developed.

To get this solution, Hz elds and derivative of Hz elds need to be solved. They

are solved for continuous propagating elds conditions. The elds are assumed

continuous at interface of RDRA. The RDRA along with eigen vectors is shown in

Fig. 4.1a, b.

CASE#1 RDRA solution:

See Fig. 4.2.

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_4

58 4 Mathematical Analysis of Transcendental

Fig. 4.1 a Rectangular DRA. b Eigen currents (current vectors) versus wavelength

4 Mathematical Analysis of Transcendental 59

To derive transcendental equation, the elds inside the resonator and outside the

resonator are required.

kz

tankz d q transcendental equation 4:1

r 1k0 2 kz 2

kx mp=a 4:3a

ky np=b 4:3b

kz pp=d 4:3c

TE11, TE11, and TE11 are dominant modes.

The solution of resonant frequency can be had if solution of ky propagation

constant is obtained from characteristic equation, r k0 2 kx 2 ky 2 kz 2 , and then

substituted in transcendental equation to compute resonant frequency f0.

Boundary condition

Propagation constant, c2mn k0 2 hmn 2

p

k 2p=k x l x=c;

Z Z

E dV

2

H 2 dV

r k0 2 kx 2 ky 2 kz 2 4:4

0 k0 2 kx 2 ky 2 kz 0

2

4:5

kz pp=d

k z 0 kz 2 0 k0 0 r k0 2

2 2

kz 02 kz 2 0 l0 x2 r l0 x2

kz 02 kz 2 x2 l1 r 4:6

60 4 Mathematical Analysis of Transcendental

of the Four Walls are PEC

Assuming that the top and bottom surface plane be at z 0; d to be PMC

) nH 0

And

nE 0

or,

Hy Hx 0

Ez 0

) nE 0

And

nH 0

At

x 0; a Ey Ez 0

Hx 0

4.1 Case-1: Top and Bottom Walls as PMC and Rest of the Four Walls are PEC 61

At

y 0; b Ex Ez 0

Hy 0

We also know

1 @Hz 1 @ 2 Ez

Ex 4:7a

jx 1 kc2

2

@y jxl @z@x

1 1 @ 2 Ez @Hz

Ey 4:7b

2

jx 1 kc2 jxl @z@y @x

1 @Ez 1 @ 2 Hz

Hx 4:7c

2

jxl 1 kc2 @y jx @z@x

1 1 @ 2 Hz @Ez

Hy 4:7d

jxl 1 kc2 jx @z@y @x

2

wz X xY yZz

where

wHz X xY yZ z

@Hz

0

Ey C

@x

62 4 Mathematical Analysis of Transcendental

or,

Ey C 0 X 0 xY yZ z

Now

X 0 x A1 cos kx x A2 sin kx x

But at

x 0; a Ey 0

) 0 A1 cos kx 0 A2 sin kx 0

or,

mp

A1 0 and kx

a

Similarly,

@Hz

Ex C 0

@y

or,

Ex C0 X xY yZz

Now

Y 0 y A3 cos ky y A4 sin ky y

But at

y 0; b Ex 0

) 0 A3 cos ky 0 A4 sin ky 0

or,

np

A3 0 and ky

b

from

01 @ 2 Hz

Hx C

jx @z@x

4.1 Case-1: Top and Bottom Walls as PMC and Rest of the Four Walls are PEC 63

or,

Hx C 0 X 0 xY yZ 0 z

Now

Z 0 z A5 cos kz z A6 sin kz z

At

z 0; d Hx 0

) A5 cos kz 0 A6 sin kz 0 0

pp

A5 0 and kz

d

Hence,

mp np pp

Hz A2 A4 A6 cos x cos y cos z 4:9

a b d

mppp mp np pp

Hx C00 A2 A4 A6 x cos

sin y sin z 4:10a

a d a b d

nppp mp np pp

Hy C 00 A2 A4 A6 cos x sin y sin z 4:10b

b d a b d

mp mp np pp

Ey C00 A2 A4 A6 sin x cos y cos z 4:10c

a a b d

np mp np pp

Ex C00 A2 A4 A6 cos x sin y cos z 4:10d

b a b d

As we know that at the PMC wall, the tangential component of magnetic eld

and normal component of electric eld are equal to zero at the interface z = 0, d.

Hence,

Hx ; Hy 0

and

Ez 0

Thus, we take Ex and Hy :

Now, the propagating wave is continuous at the interface, i.e., Ex Ex0 :

64 4 Mathematical Analysis of Transcendental

Therefore,

mp np mp np 0

A cos x sin y C1 e jkz z C2 ejkz z A cos x sin y C20 ejkz z

a b a b

4:11

or,

0

C1 e jkz z C2 ejkz z C20 ejkz z 4:12

Therefore,

C1 e jkz z C2 ejkz z 0

C1 C2 0

4:13

or; C1 C2

@Hz @Hz0

Hz Hz0 and

@x @x

mp np

Hz B cos x cos y coskz z 4:14a

a b

and

mp np

Hz0 B cos x cos y cos kz0 z 4:14b

a b

mp np mp np 0

B cos x cos y C1 e jkz z C2 ejkz z B cos x cos y C20 ejkz z

a b a b

or,

0

C1 e jkz z C2 ejkz z C20 ejkz z 4:15

4.1 Case-1: Top and Bottom Walls as PMC and Rest of the Four Walls are PEC 65

0

jkz C1 e jkz z C2 ejkz z jkz0 C20 ejkz z 4:17

or,

0

2kz C 1 coskz d kz0 C 02 ejkz z :

j tan kz d 1

0 4:18a

kz kz

Squaring both sides and substituting the value of kz2 from Eq. (4.3c),

kz 02 kz2 x2 lr 1

kz

tankz d q 4:18b

k02 r 1 kz2

4.2 Case-2

For transcendental equation, we need to compute the elds inside the resonator and

outside it.

kz

tankz d q 4:19

r 1k02 kz2

mp

kx 4:20a

a

66 4 Mathematical Analysis of Transcendental

np

ky 4:20b

b

pp

kz 4:20c

d

TE11, TE11, and TE11 are dominant modes.

Boundary condition

p x

Propagation constant, c2mn k02 hmn2 where k 2p

k x l c :

Z Z

E 2 dV H 2 dV:

When top and bottom walls are PMC, rest of the other walls is PEC

Assuming that the top and bottom surface plane be at z 0; d

) nH 0

And

nE 0

or,

Hy Hx 0

Ez 0

) nE 0

And

nH 0

At

x 0; a Ey Ez 0

Hx 0

4.2 Case-2 67

At

y 0; b Ex Ez 0

Hy 0

We also know

1 @Hz 1 @ 2 Ez

Ex 4:21a

2

jx 1 kc2 @y jxl @z@x

1 1 @ 2 Ez @Hz

Ey 4:21b

2

jx 1 kc2 jxl @z@y @x

1 @Ez 1 @ 2 Hz

Hx 4:21c

2

jxl 1 kc2 @y jx @z@x

1 1 @ 2 Hz @Ez

Hy 4:21d

jxl 1 kc2 jx @z@y @x

2

wz X xY yZz 4:22

where

X x A1 sin kx x A2 cos kx x

Y y A3 sin ky y A4 cos ky y

Z z A5 sin kz z A6 cos kz z

wHz X xY yZz

Ez 0

we get

@Hz

0

Ey C

@x

68 4 Mathematical Analysis of Transcendental

or,

Ey C 0 X 0 xY yZ z

Now

X 0 x A1 cos kx x A2 sin kx x

But at

x 0; a Ey 0

) 0 A1 cos kx 0 A2 sin kx 0

or,

mp

A1 0 and kx

a

Similarly,

@Hz

Ex C 0

@y

or,

Ex C 0 X xY 0 yZ z

Now

Y 0 y A3 cos ky y A4 sin ky y

But at

y 0; b Ex 0

) 0 A3 cos ky 0 A4 sin ky 0

or,

np

A3 0 and ky

b

0 1 @ 2 Hz

Hx C

jx @z@x

4.2 Case-2 69

or,

Hx C 0 X 0 xY yZ 0 z

Now

Z 0 z A5 cos kz z A6 sin kz z

At

z 0; d Hx 0

) A5 cos kz 0 A6 sin kz 0 0

pp

A5 0 and kz

d

Hence,

mp np pp

Hz A2 A4 A6 cos x cos y cos z 4:23

a b d

mppp mp np pp

Hx C00 A2 A4 A6 sin x cos y sin z 4:24a

a d a b d

nppp mp np pp

Hy C 00 A2 A4 A6 cos x sin y sin z 4:24b

b d a b d

mp mp np pp

Ey C 00 A2 A4 A6 sin x cos y cos z 4:24c

a a b d

np mp np pp

Ex C00 A2 A4 A6 cos x sin y cos z 4:24d

b a b d

kx kz

Hx sinkx x cosky y sinkz z

jxl0

ky kz

Hy coskx x sinky y sinkz z

jxl0

Ey kx sinkx x cosky y coskz z

Ex ky coskx x sinky y coskz z

kx2 ky2

Hz coskx x cosky y coskz z

jxl0

70 4 Mathematical Analysis of Transcendental

kx2 ky2

Hz0 coskx x cosky y cosk 0z z

jxl0

ky kz

Hy coskx x sinky yC1 e jkz d C2 ejkz d

jxl0

But

Hy 0 at d 0

C1 C2 0

or,

C1 C2

dHy

A0 jkz coskx x sinky yC1 e jkz d C2 ejkz d

dz

or,

dHy

C1 jkz coskx x sinky ye jkz d ejkz d

dz

or,

dHy

C1 2jkz coskx x sinky y coskz d

dz

0

Hy0 C10 coskx x sinky yejkz d outside the cavity

For Hz to be continuous,

dHy dHy0

dz dz

or,

0

C1 2jkz coskx x sinky y coskz d jkz0 C10 coskx x sinky yejkz d

4.2 Case-2 71

or,

0

2C1 kz coskz d kz0 C10 ejkz d 4:25

Ex ky coskx x sin ky y C1 e jkz d C2 ejkz d

At

d 0; Ex 0;

so,

C1 C2 0

or,

C1 C2

) Ex ky coskx x sin ky y C1 e jkz d ejkz d

or,

Ex 2jC1 ky coskx x sin ky y sinkz d

Also

Ex0 ky coskx x sin ky y cos kz0 z

or,

Ex0 C10 ky coskx x sin ky y ejkz d

For Hz to be continuous,

Ex Ex0

or,

or,

72 4 Mathematical Analysis of Transcendental

0 10 jk0 d

2C1 kz coskz d kz C1 e z

or,

j tan kz d 1

0

kz kz

or,

kz

j tan kz d

kz0

kz2

tan2 kz d

kz2 x2 l0 1 r

or,

kz2

tan2 kz d

x2 l0 r 1 kz2

kz

tan kz d q 4:27

r 1k02 kz2

With the help of transcendental equation, we can nd the propagation factor. Also

with the help of this equation, we can obtain resonant frequency.

CASE#3

For transcendental equation, we need to compute the elds inside the resonator and

outside it.

q

kz tankz d r 1k02 kz2 ;

r k02 kx2 ky2 kz2 ;

and

kx mp=a 4:28a

ky np=b 4:28b

4.2 Case-2 73

kz pp=d 4:28c

Boundary conditions

Propagation constant, c2mn k02 h2mn

p

k 2p= k x 2 l x=c;

Z Z

E dV

2

H 2 dV

kz0 6 pp=d

kz02 kz2 0 l0 x2 0 r l0 x2

When top and bottom walls are PEC, rest of the other walls is PMC.

Now,

Assuming that the top and bottom surface plane be at z 0; d

) nE 0

and

nH 0

or,

74 4 Mathematical Analysis of Transcendental

Ey Ex 0

Hz 0

) nH 0

And

nE 0

At

x 0; a Hy Hz 0

Ex 0

At

y 0; b Hx Hz 0

Ey 0

We also know

1 @Hz 1 @ 2 Ez

Ex 4:31a

2

jx 1 kc2 @y jxl @z@x

1 1 @ 2 Ez @Hz

Ey 4:31b

2

jx 1 kc2 jxl @z@y @x

1 @Ez 1 @ 2 Hz

Hx 4:31c

2

jxl 1 kc2 @y jx @z@x

1 1 @ 2 Hz @Ez

Hy 4:31d

jxl 1 kc2 jx @z@y @x

2

wz X xY yZ z

4.2 Case-2 75

where

X x A1 sin kx x A2 cos kx x

Y y A3 sin ky y A4 cos ky y

Z z A5 sin kz z A6 cos kz z

wHz X xY yZ z

At

x 0; a Hz 0;

or,

A1 sin kx 0 A2 cos kx 0 0

mp

) A2 0 and kx

a

Also at

y 0; b Hz 0

or,

A3 sin ky 0 A4 cos ky 0 0

np

) A4 0 and ky

b

At

z 0; d Hz 0

) A5 sin kz 0 A6 cos kz 0 0

pp

A6 0 and kz

d

Hence,

mp np pp

H z A1 A3 A5 sin x sin y sin z 4:32

a b d

76 4 Mathematical Analysis of Transcendental

mppp mp np pp

Hx C 00 A1 A3 A5 cos x sin y cos z 4:33a

a d a b d

nppp mp np pp

Hy C00 A1 A3 A5 sin x cos y cos z 4:33b

b d a b d

mp mp np pp

Ey C00 A1 A3 A5 cos x sin y sin z 4:33c

a a b d

np mp np pp

Ex C 00 A1 A3 A5 sin x cos y sin z 4:33d

b a b d

As we know that at the PEC wall, the tangential component of electric eld and

normal component of magnetic eld is equal to zero at the interface z 0; d:

Hence,

Ex ; Ey 0

and

Hz 0

Thus we take Ey and Hx :

Now, the propagating wave is continuous at the interface, i.e., Hx Hx0 :

Therefore,

mp np mp np 0

A cos x sin y C1 e jkz z C2 ejkz z A cos x sin y C20 ejkz z

a b a b

or,

0

C1 e jkz z C2 ejkz z C20 ejkz z 4:34

Therefore,

C1 e jkz z C2 ejkz z 0

C1 C2 0

4.2 Case-2 77

or,

C1 C2 4:35

Therefore,

@Hz @Hz0

Hz Hz0 and

@x @x

mp np

H z B sin x sin y sinkz z 4:36a

a b

and

mp np

H 0 z B sin x sin y sink0 z z 4:36b

a b

mp np mp np 0

B sin x sin y C1 e jkz z C2 ejkz z B sin x sin y C20 ejkz z

a b a b

or,

0

C1 e jkz z C2 ejkz z C20 ejkz z

0

jkz C1 e jkz z C2 ejkz z jkz0 C20 ejkz z

or,

0

2jkz C1 sinkz d kz0 C 02 ejkz z 4:38

78 4 Mathematical Analysis of Transcendental

Squaring both sides and substituting the value of kz02 from Eq. (4.3c), we get

q

kz tankz d r 1k02 kz2 4:39a

q

kz tankz d=2 r 1k02 kz2 4:39b

clear

clear all

er=9.8;

c=3*10^8;

d=10*10^-3;

for p=1:1:10

f=c*p*(sqrt(1+tan(p*pi/2).^2))/2*d*(sqrt(er-1));

end

plot(p,f);

title('pvsf')

xlabel('p-------->>>');

ylabel('f-------->>');

grid on;

resonant frequency is shown in Fig. 4.4.

The resonant frequency is increasing as the delta length is increasing as shown in

Fig. 4.4. Also, radiation lobe is increasing as the number of resonant mode is

increasing as shown in Fig. 4.5.

4.3 MATLAB Simulation Results 79

frequency vs pdelta

3.00E+07

frequency vs pdelta

frequency-------->>>

2.50E+07

2.00E+07

1.50E+07

1.00E+07

5.00E+06

0.00E+00

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

pdelta---------->>>

using MATLAB.

80 4 Mathematical Analysis of Transcendental

m=5;

n=4;

p=3;

a=10;

b=5;

c=2;

x=linspace(-5,5,51);

y=linspace(-2.5,2.5,51);

z=linspace(-1,1,51);

[xi,yi,zi] = meshgrid(x,y,z);

Ez= (cos(m*pi*xi/a).*cos(n*pi*yi/b)).*sin(p*pi*zi/c);

Ez= Ez.^2;

Ez= sqrt(Ez);

slice(xi,yi,zi,Ez,xslice,yslice,zslice)

colormap hsv

4.3 MATLAB Simulation Results 81

RDRA:

d=9;

w=6;

h=7.6;

c=3e8;

cons=9.8;

syms y real

kx=pi/d;

kz=pi/2/h;

ko=sqrt((kx^2+y^2+kz^2)/cons);

f=real(y*tan(y*w/2)-sqrt((cons-1)*ko^2-y^2));

ky=fzero(inline(f),[0,(pi/w)-0.01]);

fresonance = c/2/pi*sqrt((kx^2+ky^2+kz^2)/cons)/1e7;

The MATLAB-simulated resonant modes in Figs. 4.6, 4.7, 4.8, 4.9, 4.10, 4.11

and 4.12 have been drawn, and resonant frequency using transcendental equation is

placed in table form.

1

0.9

0.8

0.7

0.6

z-axis

0.5

0.4

0.3

0.2

5

0.1

0 y-axis

2.5 0

2 1.5 1

0.5

x- axis 0

-0.5

-1 -1.5 -2 -5

-2.5

TE12z

82 4 Mathematical Analysis of Transcendental

1

0.9

0.8

0.7

0.6

0.5 z

0.4

0.3

0.2

0.1

0

3

2

1

x y 3 4 5

0

1 2

-1 -1 0

-2 -3 -2

-3 -5 -4

TE 22z

1

0.8

0.6

0.4

0.2

z

0

2.5 5

2 4

1.5 3

1 2

0.5 1

0

-0.5 -1

0

y

x

-1

-1.5

-2 -4

-3

-2

-2.5 -5

TE33z

Example 1 Calculate the dimension of d in RDRA:

For TE111 mode when

r 100

a 10 mm

b 10 mm

fr 7:97 GHz

4.3 MATLAB Simulation Results 83

1

0.9

0.8

0.7

0.6

0.5

0.4

z

0.3

0.2

5

0.1

0

2.5 2 1.5

0 y

1 0.5 0 -0.5

x -1

-1.5 -2 -2.5 -5

TE12z

1

0.9

0.8

0.7

0.6

0.5

0.4

0.3 z

0.2

0.1

0

3

2

1 4 5

0 2 3

1

0

-1

x -2

-4 -3 -2 -1

y

-3 -5

TE22z

TE222

0.5

-0.5

z

-1

2.5

2

1.5 5

1 4

0.5 3

2

0

-0.5 0 y 1

-1 -1

x -1.5

-3

-2

-2 -4

-2.5 -5

84 4 Mathematical Analysis of Transcendental

4.3 MATLAB Simulation Results 85

r

c m2 n2 p2

fr p

2 r a b d

s

3 108 12

7:97 109 p 1002 1002

2 100 d

s

12

531:33 20000

d

1

512:167

d

d 1:95 mm

r 35

a 18 mm

b 18 mm

fr 2:45 GHz

Solution

r

c m2 n2 p2

fr p

2 r a b d

s

3 108 10002 10002 12

2:45 109 p

2 35 18 18 d

s

2

2

1000 1

9337:222 2

18 d

1

56:252

d

d 17:77 mm

TE111 mode using the given data of RDRA:

r 10

a 14 mm

b 8 mm

d 8 mm

86 4 Mathematical Analysis of Transcendental

Solution

r

c m 2 n2 p 2

fr p

2 r a b d

s

3 108 10002 10002 10002

fr p

2 10 14 8 8

fr 9:04 GHz

Example 4

r 10

a 14 mm

b 8 mm

d 16 mm

Solution

r

c m 2 n2 p 2

fr p

2 r a b d

s

3 108 10002 10002 10002

fr p

2 10 14 8 16

fr 7:44 GHz

Example 5 Calculate the resonant frequency for the TE11d mode using the given

data:

r 10

a 14 mm

b 8 mm

d 8 mm

Solution

s

c m 2 n2 d 2

fr p

2 r a b d

s

3 108 10002 10002

fr p 0

2 10 14 8

fr 6:82 GHz

4.3 MATLAB Simulation Results 87

Example 6

r 10

a 14 mm

b 8 mm

d 16 mm

Solution

r

c m 2 n2 p 2

fr p

2 r a b d

s

3 108 10002 10002

fr p 0

2 10 14 8

fr 6:82 GHz

The RDRAs can be prototyped with various materials and sizes as per the

requirements.

Table 4.1 consists of list of RDRA materials, permittivity, dimensions, and

computed resonant frequency.

Example 7 Compute resonant frequency when RDRA dimensions are 10 10

10 mm3 and dielectric constant of material used is 10.

r

mp2 np2 pp2

c

fr m; n; p p

2p l a b d

Resonant frequencies in isolated case are 49.7 and 25.8 GHz with ground plane

(Table 4.2).

88 4 Mathematical Analysis of Transcendental

Table 4.1 RDRA materials, permittivity, dimensions, and computed resonant frequency

S. Material Permittivity RDRA dimension Resonant Resonant

no. (a b h) mm frequency frequency

simulated by calculated

HFSS

Countis Laboratories

1. MgOSiO2TiO2 9.8 9 6 7.6 7.43 7.6757

(CD-9)

2. MgOSiO2TiO2 9.8 14.3 25.4 26.1 3.5 3.7430

(CD-9)

3. MgOCaOTiO2 20.0 10.16 10.2 7.11 4.71 4.6215

(CD-20)

4. MgOCaOTiO2 20.0 10.16 7.11 10.2 4.55 4.5941

(CD-20)

5. MgOCaOTiO2 20.0 10.2 10.2 7.89 4.635 4.4833

(CD-20)

6. MgOCaOTiO2 100.0 10 10 2 4.57 4.2158

(CD-100)

7. MgOCaOTiO2 100.0 10 10 1 7.97 7.7587

(CD-100)

8. MgOCaOTiO2 100.0 12.7 12.7 1 7.72 7.6628

(CD-100)

9. MgOCaOTiO2 100.0 5 10 1 8.85 8.1828

(CD-100)

10. MgOCaOTiO2 100.0 10 5 1 8.5 8.0147

(CD-100)

Emerson & Cuming Microwave Products N.V.

11. Magnesium 10.0 14 8 8 5.5 5.6117

titanate

(ECCOSTOCK@)

12. Magnesium 10.0 14.3 25.4 26.1 3.92 3.7055

titanate

(ECCOSTOCK@)

13. Zirconia 20.0 10.16 10.2 7.11 4.71 4.6215

(ECCOSTOCK@)

14. Zirconia 20.0 10.16 7.11 10.2 4.55 4.5941

(ECCOSTOCK@)

15. Zirconia 20.0 10.2 10.2 7.89 4.635 4.4833

(ECCOSTOCK@)

16. Strontium titanate 100.0 10 10 2 4.57 4.2158

(ECCOSTOCK@)

17. Strontium titanate 100.0 10 10 1 7.97 7.7587

(ECCOSTOCK@)

18. Strontium titanate 100.0 12.7 12.7 1 7.72 7.6628

(ECCOSTOCK@)

19. Strontium titanate 100.0 5 10 1 8.85 8.1828

(ECCOSTOCK@)

(continued)

4.4 Resonant Frequency of RDRA for Experimentations 89

S. Material Permittivity RDRA dimension Resonant Resonant

no. (a b h) mm frequency frequency

simulated by calculated

HFSS

20. Strontium titanate 100.0 10 5 1 8.5 8.0147

(ECCOSTOCK@)

Morgan Advanced Materials

21. CaMgTi (Mg, Ca 20.0 10.16 10.2 7.11 4.71 4.6215

titanate)

(D20)

22. CaMgTi (Mg, Ca 20.0 10.16 7.11 10.2 4.55 4.5941

titanate)

(D20)

23. CaMgTi (Mg, Ca 20.0 10.2 10.2 7.89 4.635 4.4833

titanate)

(D20)

24. ZrTiSn (Zr, Sn 37.0 18 18 9 2.45 2.1617

titanate)

(D36)

Temex Components & Temex Telecom, USA

25. Zr Sn Ti Oxide 37.0 18 18 9 2.45 2.1617

(E2000)

Trans-Tech Skyworks Solutions, Inc.

26. BaZnCoNb 35.036.5 18 18 6 2.532 2.7081

(D-83)

27. BaZnCoNb 35.036.5 18 6 18 2.835 2.3947

(D-83)

T-CERAM, RF & Microwave

28. E-11 10.8 15.2 7 2.6 11.6 10.379

29. E-11 10.8 15 3 7.5 6.88 7.0937

30. E-11 10.8 15.24 3.1 7.62 6.21 6.9440

31. E-20 20.0 10.16 10.2 7.11 4.71 4.6215

32. E-20 20.0 10.16 7.11 10.2 4.55 4.5941

33. E-20 20.0 10.2 10.2 7.89 4.635 4.4833

34. E-37 37.0 18 18 9 2.45 2.1617

TCI Ceramics, Inc.

35. DR-36 36.0 18 18 6 2.532 2.7081

36. DR-36 36.0 18 6 18 2.835 2.3947

90

Table 4.2 Fringing effect along b dimensions increased effective along y-direction of RDRA

S. Permittivity Dimension (a (length) b (width) d (depth)) Resonant Effective width Multiple % change in

no. mm frequency (b) factor width

37. 10.0 14.3 25.4 26.1 3.5 34.22 1.3474 34.7381

38. 10.0 14 8 8 5.5 14.13 1.7665 76.6535

39. 10.0 15.24 3.1 7.62 6.21 8.33 2.8872 168.7230

40. 20.0 10.2 10.2 7.89 4.635 15.31 1.5014 50.1419

41. 20.0 10.16 10.2 7.11 4.71 15.15 1.4858 48.5797

42. 35.0 18 18 6 2.532 24.12 1.34 33.9973

43. 35.0 18 18 9 2.45 25.64 1.4244 42.4423

44. 100.0 10 10 1 7.97 11.24 1.1242 12.4237

4 Mathematical Analysis of Transcendental

4.4 Resonant Frequency of RDRA for Experimentations 91

%%Dimensions of RDRA

%%length

d=[14.3,14.0,15.24,10.2,10.16,18,18,10];

%%width

w=[25.4,8,3.1,10.2,10.2,18,18,10];

%%height

h=[26.1,8,7.62,7.89,7.11,6,9,1];

%%Mode

m=1;

n=1;

p=1;

c=3e8;

cons=[10.0,10.0,10,20,20,35,35,100];

syms y real

for i=drange(1:8)

kx(i)=pi/d(i);

kz(i)=pi/2/h(i);

ko=sqrt((kx(i).^2+y.^2+kz(i).^2)/cons(i));

f=real(y.*tan(y*w(i)/2)-sqrt((cons(i)-1)*ko.^2-y.^2));

ky(i)=fzero(inline(f),[0,(pi/w(i))-0.01]);

%%Resonant frequency

fre(i)=c/2/pi*sqrt((kx(i).^2+ky(i).^2+kz(i).^2)/cons(i))*1e3;

Effwidth(i)=pi/ky(i);

factor(i)=Effwidth(i)./w(i);

perchangwidth(i)=((Effwidth(i)-w(i))/w(i))*100;

end

Program 1

%%Dimensions of DRA

%%length

d=[14.3,14.0,15.24,10.2,10.16,18,18,10];

%%width

w=[25.4,8,3.1,10.2,10.2,18,18,10];

%%height

h=[26.1,8,7.62,7.89,7.11,6,9,1];

%%Mode

m=1;

n=1;

92 4 Mathematical Analysis of Transcendental

p=1;

c=3e8;

cons=[10.0,10.0,10,20,20,35,35,100];

syms y real

for i=drange(1:8)

kx(i)=pi/d(i);

kz(i)=pi/2/h(i);

ko=sqrt((kx(i).^2+y.^2+kz(i).^2)/cons(i));

f=real(y.*tan(y*w(i)/2)-sqrt((cons(i)-1)*ko.^2-y.^2));

ky(i)=fzero(inline(f),[0,(pi/w(i))-0.01]);

%%Resonant frequency

fre(i)=c/2/pi*sqrt((kx(i).^2+ky(i).^2+kz(i).^2)/cons(i))*1e3;

Effwidth(i)=pi/ky(i);

factor(i)=Effwidth(i)./w(i);

perchangwidth(i)=((Effwidth(i)-w(i))/w(i))*100;

end

Results:

4.4 Resonant Frequency of RDRA for Experimentations 93

Program 2

m=1;

n=1;

p=1;

E_r=10;

a=15.24e-03;

b=3.1e-03;

d=7.62e-03;

c=3e+08;

kx=m*pi/a;

ky=n*pi/b;

kz=p*(pi/d)/2;

ko=sqrt(kx^2+ky^2+kz^2)/sqrt(E_r);

fo=(c*ko/pi)/2;

foghz=fo/(1e+09);

Results:

94 4 Mathematical Analysis of Transcendental

Program 3

MATLAB programs taking parameters a, b, d same and comparing

frequency using:

Program 1: Characteristic equation

m=1

n=1

p=1

E_r=10

a=14.3e-03

b=25.4e-03

d=26.1e-03

c=3e+08

k_x=m*pi/a

k_y=n*pi/b

k_z=p*(pi/d)/2

k_o=sqrt(k_x^2+k_y^2+k_z^2)/sqrt(E_r)

f_o=(c*k_o/pi)/2

f_oGHz=f_o/1e+09

Output:

4.4 Resonant Frequency of RDRA for Experimentations 95

Program 4

Transcendental equation for same dimensions:

m=1;

n=1;

p=1;

E_r=10;

a=14.3e-03;

b=25.4e-03;

d=26.1e-03;

c=3e+08;

syms y real

kx=pi/a;

kz=pi/d/2;

ko=sqrt(kx^2+y^2+kz^2)/sqrt(E_r);

f=real(y*tan(y*b/2)-sqrt((E_r-1)*ko^2-y^2));

ky=fzero(inline(f),[0,(pi/b)-0.01]);

fre=c/2/pi*sqrt((kx^2+ky^2+kz^2)/E_r)*1e3;

effwidth=pi/ky;

factor=effwidth/b;

perchangwidth=((effwidth-b)/b)*100;

Output:

96 4 Mathematical Analysis of Transcendental

Program 5

MATLAB programs taking parameters a, b, d same and comparing

frequency using:

Characteristic equation

Where a=17mm

b=25mm

c=10mm

m=1;

n=1;

p=1;

E_r=10;

a=17e-03;

b=25e-03;

d=10e-03;

c=3e+08;

k_x=m*pi/a;

k_y=n*pi/b;

k_z=p*(pi/d)/2;

k_o=sqrt(k_x^2+k_y^2+k_z^2)/sqrt(E_r);

f_o=(c*k_o/pi)/2;

f_oGHz=f_o/1e+09;

Output:

4.4 Resonant Frequency of RDRA for Experimentations 97

Program 6

Transcendental equation

m=1;

n=1;

p=1;

E_r=10;

a=17e-03;

b=25e-03;

d=10e-03;

c=3e+08;

syms y real

kx=pi/a;

kz=pi/d/2;

ko=sqrt(kx^2+y^2+kz^2)/sqrt(E_r);

f=real(y*tan(y*b/2)-sqrt((E_r-1)*ko^2-y^2));

ky=fzero(inline(f),[0,(pi/b)-0.01]);

fre=c/2/pi*sqrt((kx^2+ky^2+kz^2)/E_r)*1e3;

effwidth=pi/ky;

factor=effwidth/b;

perchangwidth=((effwidth-b)/b)*100;

98 4 Mathematical Analysis of Transcendental

Program 7

frequency using:

Characteristic equation

m=1

n=1

p=1

E_r=10

a=14.3e-03

b=25.4e-03

d=26.1e-03

c=3e+08

k_x=m*pi/a

k_y=n*pi/b

k_z=p*(pi/d)/2

k_o=sqrt(k_x^2+k_y^2+k_z^2)/sqrt(E_r)

f_o=(c*k_o/pi)/2

f_oGHz=f_o/1e+09

Output:

4.4 Resonant Frequency of RDRA for Experimentations 99

Program 8

Transcendental equation

m=1;

n=1;

p=1;

E_r=10;

a=14.3e-03;

b=25.4e-03;

d=26.1e-03;

c=3e+08;

syms y real

kx=pi/a;

kz=pi/d/2;

ko=sqrt(kx^2+y^2+kz^2)/sqrt(E_r);

f=real(y*tan(y*b/2)-sqrt((E_r-1)*ko^2-y^2));

ky=fzero(inline(f),[0,(pi/b)-0.01]);

fre=c/2/pi*sqrt((kx^2+ky^2+kz^2)/E_r)*1e3;

effwidth=pi/ky;

factor=effwidth/b;

perchangwidth=((effwidth-b)/b)*100;

Output:

100 4 Mathematical Analysis of Transcendental

Program 9

4.4 Resonant Frequency of RDRA for Experimentations 101

Program 10

102 4 Mathematical Analysis of Transcendental

Q:No: 2 Compute propagation constants in x-, y-, and z-directed propagated

RDRAs, when feed probe is given. Compute its resonant frequency

when RDRA dimensions are 5 5 3 mm3 and dielectric constant used

is 20.

Chapter 5

Mathematical Analysis of RDRA

Amplitude Coefcients

(RDRA) have been evaluated. Rigorous theoretical analysis has been developed for

different resonant modes inside RDRA. The resonance phenomenon and its

potential use as radiator have been described. The dielectric polarization P is equal

to the total dipole moment induced in the volume of the material by the electric

eld. The discontinuity of the relative permittivity at the resonator surface allows a

standing electromagnetic wave to be supported in its interior at a particular resonant

frequency, thereby leading to maximum connement of energy within the reso-

nator. Certain eld distributions or modes will satisfy Maxwells equations and

boundary condition. Mathematical solution to get amplitude coefcients Cmnp along

with its phase coefcients has been obtained. These are also known as eigenvector.

Keywords Amplitude coefcients Resonant modes Radiation lobes Fourier

transform Discrete solution PMC (perfect magnetic conducting) PEC (perfect

electrical conducting) Dominant mode Higher-order modes

5.1 Introduction

Rigorous theoretical analysis has been developed for resonant modes in rectangular

DRA (RDRA). RDRA resonance phenomenon and its potential, as a radiator have

been long back described. Accordingly, external electric elds bring the charges of

the molecules of the dielectric into a certain ordered arrangement in space. The

dielectric polarization P is equal to the total dipole moment induced in the volume

of the material by the electric eld. The discontinuity of the relative permittivity at

the resonator surface allows a standing electromagnetic wave to be supported in its

interior at a particular resonant frequency, thereby leading to maximum conne-

ment of energy within the resonator. Certain eld distributions or modes will satisfy

Maxwells equations and boundary conditions. Resonant modes are eld structures

that can exist inside the RDRA. The RDRA prototype is shown in Fig. 5.1.

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_5

104 5 Mathematical Analysis of RDRA Amplitude Coefcients

dielectric RDRA on ground

plane

P jxmnpt

mnp ReCmnp e umnp x; y; z, using orthonormality.

In discrete form,

X

jCmnp jumnp x; y; z cos x xmnp t Wmnp

m;n;p

Z

jxlIdlx2 y2 jxtxp

x2 y2 d2

Ez x; y; d; t Gx; y e c I xejxt dx

2 3=2

4p x y d

2 2

r

X 2 ppd

Resonator current Cmnp sin cosxmnpt Wmnpum;n x; y;

p

d d

Z

jxlIdlx2 y2

Probe current Gx; y 3=2 I xe dx

jkt

4p x2 y2 d2

p

jxtxc x2 y2 d2 wmnp

e um;n x; ydx dy ;

5.2 Amplitude Coefcients Cmnp 105

The probe current must be equal to the resonator current due to principle of

orthonormality.

X

1 n o

E x; y; z; t Re Cmnpejxmnpt wEmnp x; y; z

mnp1

X

1 n o

E x; y; z

Re Dmnpejxmnpt / mnp

mnp1

X

1 n o

H x; y; z; t Re Cmnpejxmnpt wH

mnp

x; y; z

mnp1

X

1 n o

H x; y; z

Re Dmnpejxmnpt / mnp

mnp1

c jxl

E? 2

r? Ez 2 r? Hz x ^z

h h

From duality

c jx

H? r Hz 2 r? Ez x ^z

h2 ? h

mpx 0 mpx npy np mpx npy

Ex a mnp cos sin b0 mnp cos sin ;

a a b b a b

npy 0 mpx npy mp 0 mpx npy

Ey a mnp sin cos b mnp cos sin ;

b a b a a b

and

r

X 2 ppd

jxmnpt

Ez ReCmnpe sin um;n x; y

m;n;p

d d

Z

jxlIdlx2 y2 jxp

x2 y2 d2

Ez e c : I xejkt dx

2 3=2

4p x y d

2 2

Here, I x is the Fourier transform of source current, i.e., I(t) is the probe current

1X

Ix jI mnpj dx xmnpej mnp ej mnpdx xmnp

2 mnp

106 5 Mathematical Analysis of RDRA Amplitude Coefcients

p !!

ldlx2 y2 x2 y2 d2

Ez x; y; z; t 3=2 xmnpjI mnpj sin xmnp t /mnp

4p x2 y2 d2 c

r

2 ppd

Cmnp jumn x; y cosxmnpt wmnp sin

d d

s

bmnp2 amnp2

Hence, Cmnp h p i2 ; amplitude coefcient

p2 sin

ppd

d d

" #

1

amnp cos/mnp bmnp sin/mnp

wmnp tan ; Phase

amnp sin/mnp bmnp cos/mnp

homogeneous medium.

r E jxlH M; a

r H J jxE; b

qv qm

rE ; rH ;

e l

where qv is the electric charge density, and qm is the magnetic charge density.

r J jxer E 0;

i:e: r M jxqm 0; r J jxq 0;

rr E jxlr H r M;

or rr E r2 E jxl J jxE r M;

rq

or r2 k2 E jxlJ r H s electric source; c

e

5.3 RDRA Maxwells Equation-Based Solution 107

Likewise; r r H r J jxer E

or rr H r2 H r J jxejxlH M

rqm

or r2 k 2 H jxeM r J f

l

magnetic source due to probe; d

Hence, H also satises Helmholtz equation with source. Rectangular cavity reso-

nator sidewalls are the perfect magnetic conductors (PMC) and top and bottom

surfaces are the perfect electric conductors (PEC). Applying these boundary con-

ditions, we get the following equation:

Hz 0; where x 0; a or y 0; b

So,

X

Hz x; y; z /mn zumn x; yja; b 5:1

m;n 1

where

2 mpx npy

umn x; yja; b p sin sin 5:2

ab a b

as we know,

Hz0 x; y; z Cmn p sin sin sin

ab a b d

Let

X

fz x; y; z fzmn zumn x; yja; b 5:3

mn 1

X

r2 k2 Hz /00mn z k 2 h2 m; nja; b/mn zumn x; yja; b

mn

5:4

fz ) /00mn z k2 h2 m; nja; b/mn z

fzmn z

m 2 n 2

where k2 x2 l; and h2 x; yja; b p2 a b ;

108 5 Mathematical Analysis of RDRA Amplitude Coefcients

from (1),

Taking Laplace Transform of (5.4);

mn s ^fzmn s

S2 /mn s S/mn 0 /0mn 0 c2z m; n b

So,

b

mn s 2 5:5

s2 cz m; n

2 S c2z m; n

Thus,

Zz

1

b

mn z sin cz m; nz n fzmn ndn

cz m; n

0

C1 sin cz m; nz C2 coscz m; nz

mn 0 mn d 0 ) C2 0;

Zd

1

C1 sincz m; nd nfzmn ndn

cz m; n sincz m; nd

0

So,

Zz

1

mn z sin cz m; nd sin cz m; nz n fzmn ndn

cz m; n sin cz m; nd

0

Zd

sin cz m; nz sin cz m; nd n fzmn ndn

0

2

m 2 n 2 p 2

In the limit k 2 ! p2 a b d ; we have, c2z m; n ! pp

d and we get;

8 z 9

Z R d pp

d < pp ppz sin d n fzmn n dn =

mn z ! sin z n fzmn ndn sin limpp 0 d

pp : d d k! d sinkd ;

0

5:6

5.3 RDRA Maxwells Equation-Based Solution 109

m 2 n 2 p 2

k2 p2

a b d

Let r

X

1

2 rpt

fmn z fzmnr sin sin 5:7

r1

d d

Then

Zd pp

sin d n fzmn ndn

d

0

5:8

r Zd rpn

2X pp

fzmnr sin d n sin dn

d r d d

0

r Zd

p1 2X rpn

1 fzmnr sinkn sin dn

d r d

0

r Zd

p1 2X 1 rp rp 5:9

1 fzmnr cos k n cos k n dn

d r 2 d d

0

r " #

1 2 X sin k rp d sin k rp d

1 p1

fzmnr d d

2 d r k drp

k d

rp

d

Thus,

Zd

1

sinkd n fzmn ndn

sinkd

0

r " #

1 2X 1 sinkd 1r sinkd 1r

1p1 fzmnr

2 d r sinkd k rp

d k rp

d

" # 5:10

1 X 1 1

p 1 p1

1 fzmnr

r

2d r k rp

d k rp

d

2 3

1p1 X 2rp

p 1r fzmnr 4 d

2 5

2d r k rp

2

d

110 5 Mathematical Analysis of RDRA Amplitude Coefcients

Writing k pp

d d d ! 0; we get;

Zd

1

sinkd n fzmn ndn

sinkd

0

1p1 fzmnp

p 1p

2d d

5:11

1

p fzmnp Dominant term

d 2d

Hence,

8 z r 9

< Z X =

d 2 rpn ppz 1

mn z sinkz n fzmnr sin dn sin p fzmnp

pp : r1

d d d d 2d ;

0

5:12

Now

Zz Zz

rpn 1 h rp rp i

sinkz n sin dn cos kz k n cos kz k n dn

d 2 d d

0 0

" rpz #

1 sin rpz sinkz sin d sinkz

d

2 k rpd k rp

d

2 ppz rpz 3

d sin d dz

1 4 sin rpz sin d sin ppz

dz

5

ppr

d

2

d d ppr

d d

5:13

2

2 2

Hence, for k 2 p2 ma nb pp d d

ppz

d 1

mn z p fzmnp sin

pp d 2d d

r 5:14

1 d ppz

fzmnp sin

ppd 2 d

5.3 RDRA Maxwells Equation-Based Solution 111

r2 k 2 Hx fx

X

Hx x; y; z xmn ~umn y; zjb; d

m;n

X

Let, fx ~

x; y; z fxmn x~umn y; zjb; d

m;n

mpy npy

where ~umn y; zjb; d p2

bd

sin b cos b ; orthogonal 2D half wave Fourier basis

function.

Then,

00xmn x k2 h2 m; njb; d xmn x fxmn x

Zx

1

xmn x sincx m; nx nfxmn n

cx m; n

0

C1 coscx m; nxC2 sincx m; nx 5:15

Likewise,

X

Hy x; y; z ymn yumn x; zja; d

m;n

X

fy x; y; z fymn yumn x; zja; d

m;n

with

00ymn y k2 h2 m; nja; d uymn y fymn y

Ex 0 where x 0; a or z 0; d;

Ey 0 where y 0; b or z 0; d;

Zy

1

ymn y sin cy m; nz n fymn ndn

cy m; n 5:16

0

D1 coscy m; ny D2 sincy m; ny

112 5 Mathematical Analysis of RDRA Amplitude Coefcients

cy m; n k 2 h2 m; nja; d1=2

The equation

r H J jxE

gives

jxEy Hx;z Hz;x J

We assume that J on the walls is zero. Then, the boundary conditions yields

Hx;z Hz;x 0; where y 0; b

e e0 1 dp Xe x; y 5:17

l l0 1 dp Xm x; y 5:18

been worked out using perturbation techniques to determine shift in the frequency.

As per Maxwells equation,

r E jxlH

r H jxeE

0 x aW

0 y bL

0 z dh

5.4 RDRA Inhomogeneous Permittivity and Permeability 113

Sidewalls have been taken as PMC (magnetic conductor walls) and top and

bottom as PEC (perfect electrical conductor).

Hx ; Hy 0; when y 0; b

Hy ; Hx 0; when x 0; d

Ex ; Ey 0; when z 0; d

E E x; yecz 5:19

H H x; yecz 5:20

h20 c2 x2 l0 0 ; c2 k 2 ; when k2 x2 l0 0

c jxl

EX Ez0 x 2 Hz0 y 5:21

h2 h

c jxl

EY Ez0 y 2 Hz0 x 5:22

h2 h

c jx

HX Hz0 x 2 Ez0 y 5:23

h2 h

c jx

HY Hz0 y 2 Ez0 x 5:24

h2 h

Ex ; Ey 0; when z 0; d

E E x; yexpcz; H H x; yexpcz

E E ? Ez^z; H H ? Hz^z:

@

r r? ^z r? c^z

@z

114 5 Mathematical Analysis of RDRA Amplitude Coefcients

r? H ? j w Ez^z 5:28

jx c E? r? Hz ^z

5:30

c jxl Z^ H ? r? Ez

jxl c r? Hz Z^

E? c jx r? Ez

^z H ? x2 l c2

c jxl

E? r Ez 2 r? Hz ^z 5:31

h2 ? h

c jxl

^z H ? r Hz ^z 2 r? Ez 5:32

h2 ? h

where h20 c2 x2 l0 e0 ; k2 x2 l0 e0 c2 k 2

vx; y ve x; y vm x; y d ve x; yvm x; y

c jxe

r? Hz 2 ^z r? Ez H ? 5:33

h2 h

c jxl

EX Ez;x 2 Hz;y

h2 h

c jxl

EY Ez;y 2 Hz;x

h2 h

c jx

HX Hz;x 2 Ez;y

h2 h

c jx

Hy Hz;y 2 Ez;x

h2 h

5.4 RDRA Inhomogeneous Permittivity and Permeability 115

c jxl ^

r? r E

? z r H

? z Z jxlHz^z 0 5:34

h2 h2

c jxl

or ^z r? 2 ; r? Ez r? ; 2 r? Hz jxlHz 0

h h

or

jxl h2 h2 ^ c

r2? Hz h H z r?

2

; r ? Hz Z r? ; r ? E z

h2 jxl jxl h2

0

or

l

ck2

r2? h20 Hz d k 2 XHz d1 log 2 r? ; r? Hz r X r E

jxlh2 ?

1 ? z

h

0

5:35

v ve vm

vm vm

2

h2 h0

ck 2 ck2

jxlh 2 jxlh20

2 r? v

r2? h20 Hz d k 2 v Hz r? vm ; r? H z

h20

5:36

ck 2

r ; vr? Ez g 0

jxl0 h20 ?

By duality

E ! H; H ! E; ve $ vm

0 $ l0 ; v$v

116 5 Mathematical Analysis of RDRA Amplitude Coefcients

k2 k2

r2? h20 Ez d k2 vEz r? ve r ? v; E r ? v r v; Ez

h20 ?

z e

h20

5:37

ck 2

r v; r? Hz 0

jx0 h20 ?

Hz 0; x 0; a and Y 0; b Hz 0; Z 0; d

HX 0; Y 0; b HY 0; x 0; a EX 0; x 0; a

EX EY 0; Z 0; d EY 0 y 0; b

Equations (5.28) and (5.29) are the own fundamental equations, let h20 k:

Let

k km;n

0

d k1 0 d2

Ez Ez0 dEz1 0 d2 5:38

Hz Hz0 dHz1 0 d2 5:39

if there is non-homogeneity !

2

m n2

kn;m

0

p 2

a2 b2

0

r2? kn;m Ez0 0

0

r2? kn;m Hz0 0 5:40

By Eqs. (5.36) and (5.37) 0 d0

c 0 jx0

Hx0 Hz;x 0 0

Ez;y

k0m;n km;n

Since

0 0

HZ 0; when Y 0; b Hz;x 0; when y 0; b

5.4 RDRA Inhomogeneous Permittivity and Permeability 117

Then

0

Hx0 0; when y 0; b ! Ez;Y 0; when Y 0; b

Likewise

Ez;0X 0; when X 0; a

Thus,

mpx npy 2

Hz0 C mn sin sin p 5:41

a b ab

mpx npy 2

Ez0 Dmn cos cos p 5:42

a b ab

If z-dependent is taken into account, then Hz0 ; Ez0 must be multiplied by exp

cz

according to Eq. (5.34),

c jxl0

Ex0 Hz;0Y

k0m;n k0m;n

We get Ez;0X 0; when z 0; d then,

Ez0 x; y; z Dmn p cos cos sin 5:43

ab a b d

jpp

c ; p 1; 2; 3:

d

Hz0 x; y; z Cmn p sin sin sin 5:44

ab a b d

Frequency of oscillations:

x xmnp

0

c2 x2 l0 0 kmn

1=2

p m 2 n 2 p 2

or x p :

l0 0 a a d

118 5 Mathematical Analysis of RDRA Amplitude Coefcients

The rectangular cavity has dimensions a, b, and d as shown in Fig. 5.2. Sidewalls

are taken as magnetic conductors (PMC), and top and bottom surfaces are as PEC;

theoretical elds (modes) solution has been worked under boundary conditions with

a square-type feed probe for excitation.

EX ; Ey 0; top and bottom plane being electric walls.

EX ; Ey 0; sidewalls being magnetic walls.

X 2 mpx npy ppz

Hz x; y; z; t Cm; n; p p sin sin sin fcosxm; n; pt

mnp ab a b d

/m; n; p

where m, n, and p are the integers (half wave variations in particular direction, i.e.,

x, y, z directions, respectively); a, b, and d are the dimensions (width, length, and

height) of the RDRA, C m; n; p and /m; n; p are the magnitude and phase

coefcients of Hz and Dm; n; p and wm; n; p for Ez. npy

Let, orthogonal 2D half wave Fourier basis function p2 sin mpx

ab a sin b

umn x; y for convenience.

X 2 mpx npy ppz

Ez x; y; z; t d m; n; p p cos cos cos

mnp ab a b d

fcos xm; n; pt wm; n; pg

npy

Let, p2

ab

cos mpx

a cos b vmn x; y for convenience = orthogonal 2D half wave

Fourier basis function.

From Lorentz Gauge conditions, Ez jxAz @@z/

Therefore, the magnetic vector potential can be given as below in discrete form

after taking Fourier transform of Az.

feed probe inserted in

a b plane

d

a

b

5.5 RDRA with Probe Current Excitation 119

xr

A 4p r

Div A ^ @ A^ Z ; need to be computed.

@z

Now, if we insert this probe at the location dened below into the cavity to nd

the elds pattern, we get:

l2 a a l2 b b

\jx j\ ; \jy j\

2 2 2 2 2 2

^ z l^Idl

@A @ ejkr l^idl cos h jk cos h jkr jx ^

e 2/

@z 4p @z r 4p r2 r c

^

l^Idljc2 1 jk jkr l^idlc2 z jkz jkr

/ cos h 2 e e

4 px r r 4 px r 3 r 2

^ w.r.t. z

Differentiating /

^

@/ l^idlc2 1 3z2 jk 2jkz2 z jkz jkz

5 2 4 3 2 ejkr 5:45

@z 4px r 3 r r r r r r

^

when; E ^ z @ /^ ; substituting

^ z jxA @/ ^z,

in E

@z @z

^ jxl^Idl l^Idlc2 1 3z2 jk 2jkz2 jkz2 k 2 z2

Ez 5 2 4 4 3 ejkr 5:46

4pr 4px r 3 r r r r r

If we take h p2 ; z 0

p

If we take h ; z 0

2

^

^ z xclIdl jk 1 3 d jk 2jkd jkd k d

2 2 2 2 2 2

E

4pk r r3 r5 r2 r4 r4 r3

Also,

q

r x 2 y 2 d2

k2 r2

r 1r , for r k 2p=k:

120 5 Mathematical Analysis of RDRA Amplitude Coefcients

given that d r:

Minimum of r l2 and Maximum of r a; b;

kr 1

Hence,

^ lc2 dl ^

^ z lcIdl

E and I x;

4pkr 3 4pr 3 jx

Zt

dl Qtdl

Ez t; x; yd p 3 Isds 5:47

4pr 3

4p x2 y2 0

Rt

Charge flowing through the resonator is Qt 0 Isds or equivalently

^

^ x;

I x Q

jx

Here,

X

Qtdl ppd

3 Dm; n; pvmn x; y sin cosxm; n; pt wm; n; p

4px2 y2 2 mnp

d

and

Qt X pp

3 C m; n; pvmn x; y cosxm; n; pt cm; n; p

4px2 y2 2

mnp

d

for Hz elds and Cm; n; p and cm; n; p for Ez elds. The Dm; n; p and

C m; n; p are the desired resonant modes. For region,

l2 a a

\jx j\

2 2 2

l2 b b

\jy j\

2 2 2

Z Z X Dm; n; ppp

Q t vmn x; y

3 dxdy cosxm; n; pt wm; n; p

4p x2 y2 2

p

d

5:48

a l2 a l2 b l2 b l2

\x\a; 0\x\ \ \y\b [ 0\y

2 2 2 2

5.5 RDRA with Probe Current Excitation 121

1

hcosxt cosxti

2

hsinxt cosxti 0

Z

Dm; n; ppp 1 vmn x; y

coswm; n; p hQt cosxm; n; pti 3 dxdy

2d 4p x2 y2 2

Z

Dm; n; ppp 1 vmn x; ydxdy

sinwm; n; p hQt sinxm; n; pti 3

2d 4p x2 y2 2

Hence,

Z !

1 vmn x; ydxdy

Dm; n; p 2d=pp sinwm; n; p hQt sinxm; n; pti 3 :

4p x2 y2 2

5:49

Medium

The basic Maxwells theory can be applied with boundary conditions to express

RDRA resonant elds as superposition of these characteristics frequencies. RDRA is

shown in Fig. 5.3. umn depends on input excitation = orthogonal Fourier basis

function, hmn resonant mode (cut off frequency), k propagation constant. The gen-

eration of modes or characteristics frequencies xmnp due to electromagnetic elds

oscillations inside the cavity resonator has been described. Orthogonal Fourier basis

npy

function um;n x; y p2

ab

sin mpx

a sin b ; xmnp is the characteristic frequency

and wmnp is the phase of current applied. The rectangular cavity resonator is

excited at the center with an antenna probe carrying current i(t) of some known

plane

122 5 Mathematical Analysis of RDRA Amplitude Coefcients

frequency xmnp. This generates the eld Ez inside the cavity of the form given

below:

k2 c2mn h2mn

hence,

p 2 p2

k2 h2mn

d2

X Z

Ez x; y; z; t Re Cmnp ejxmnpt umnp x; y; z;

m;n;p

P

or Cmnp umnp x; y; z cosxmnpt wmnp;

m;n;p

Z

jxlIdlx2 y2 jxtxp

x2 y2 d2

Ez x; y; d; t Gx; y 3=2

e c I xejxt dx

4p x2 y2 d2

where G(x,y) are the constant terms associated with the current.

Equating RDRA probe current elds with the antenna-radiated current elds at

z d;

Radiated currents:

r

X 2 ppd

Cmnp sin cos xmnpt /mnp um;n x; y ;

p

d d

Probe currents:

Z p

jxlIdlx2 y2 jxtxc x2 y2 d2 wmnp

Gx; y 3=2 I xe dx e

jkt

um;n x; y dxdy

4p x2 y2 d2

It is clear that these two expressions have to be equal due to energy conservation.

The probe current can be dened as:

1X h i

Ix jI mnpj dx xmnpejmnp ejmnp dx xmnp

2 mnp

The antenna probe current must contain only the resonator characteristics fre-

quencies xmnp. The radiated and input currents are equated as:

5.6 RDRA Resonant Modes Coefcients in Homogeneous Medium 123

r

X 2 ppd

Cmnp sin cosxmnpt /mnpum;n x; y

p

d d

Z p

jxlIdlx2 y2 jxtxc x2 y2 d2 wmnp

Gx; y 3=2

:I xe jkt

dxe um;n x; ydxdy;

4p x2 y2 d2

Hence, we can conclude that modes generation is due to the dipole moment in

cavity resonator, mostly depend on size, dimensions of device, excitation type,

coupling, and point of excitation.

probe length inserted into RDRA resonator at

point of insertion a=2; b=2; d or x a=2; y b=2; d ; where dlength of

insertion.

Hx ; Hy ; Ex ; Ey , transverse elds; (Ez and Hz) longitudinal elds

X

Ez umnp x; y; zRe Cmnp ejxmnp t

mnp

npy ppz

where umnp x; y; z p2 sin mpx

3=2

walls are PMC, rest all four walls are PEC).

Appling boundary conditions on transparent sidewalls (on all four sides of

RDRA or resonator) and top and bottom planes as electrical walls, we get Hz 0,

for magnetic walls; and Ez 0, for electrical walls; elds to be computed are

Ez ; Hz longitudinal elds;

X Z

Ez x; y; z; t Re Cmnp ejxmnpt umnp x; y; z

mnp

P

q

Ez mnp Re Cmnp ejxmnpt d2 sin ppd

d umn x; y; this is the Ez eld in the reso-

nator at z d. It must be equated to the corresponding eld generated by the

antenna probe, i.e., for the above two expressions to be equal, the antenna probe

currents must contain frequencies

onlynpfrom the set fxmnpg:

Where umn x; y p2

sin mp

ab a sin b

124 5 Mathematical Analysis of RDRA Amplitude Coefcients

p

R jk 2 2 2

x y d

Ez will exist little above from z 0 plane; Ez p

jxlIdle

I xdx; where

4p x2 y2 d2

I x is the Fourier transform of i(t)

@/

Ez jxAz

@z

lIdl

divA jk cos hejkr 5:50

4p

kc2 jx

2/

x c

cos jkr

e

@/ lI cos h

jk cos hejkr

@z 4pr 5:51

jklI cos 2h jkr

e

4pr

jxlIdl jkr jxclIdl cos 2h jkr

Ez e e

4pr 4pr

Hence,

Ez will be e

4pr

where

d

cos h p2

x y2 d

2

x2 y2

sin2 h

x 2 y 2 d2

r

X 2 ppd

jxmnpt

Ez jz d ReCmnpe sin um;n x; y 5:52

mnp

d d

Z

jxlIdlx2 y2 jxp

x2 y2 d2

Ez 3=2

e c I xejkt dx 5:53

4p x2 y2 d2

5.7 RDRA Modes with Different Feed Position 125

1X

I x jI mnpjdx xmnpej mnp

2 mnq

ej mnpdx xmnp

1X

Ix jI mnpjdx xmnpej mnp

2 mnq

ej mnpdx xmnp

Z

Ix cosxmnptejxt dt

q

2

When xmnp p2 ma2 nb2 dp2 , probe current magnitude and phase Ix

2 2

P

m;np jI mnpj cosxmnpt /mnp mnp is the phase of current at fre-

quency xmnp.

Ez x; y; z; t

p !!

ldlx2 y2 x2 y2 d 2

3=2 xmnpjI mnpj sin xmnp t /mnp

4p x2 y2 d2 c

r

2 ppd

jCmn jumn x; y cos xmnpt wmnp sin :

d d

5:54

knowledge of antenna radiation behavior, surface current distribution, input

impedance, and its feeding point location. Combinations of feeding conguration

and dimensions can generate or excite various modes. Thus, modes can be effec-

tively used in design control of an antenna. Surface current and geometry of an

antenna give eigenfunctions or eigenvectors. Closed-loop currents of eigenvectors

that present inductive nature are the magnetic elds. Horizontal and vertical

eigenvectors are noninductive are electric elds. These electric elds are produced

by supplied probe currents. Number of lobes in radiation pattern gets increased if

mode number or order of mode is increased and vice versa. The modal excitation

coefcients shall depend on position, magnitude, and phase of the applied probe

current. The effective current is superposition of all modes excited. The eigenvalue

is most important because its magnitude tells effectiveness of radiation or reactive

power and modes are the solution of characteristics equation. Smaller magnitude of

126 5 Mathematical Analysis of RDRA Amplitude Coefcients

mode, and if modes are negative, it stores electric energy. The eigenvalue variation

versus frequency gives information about resonance and radiation nature.

Excitation angle can have impact on antenna quality factor. The excited mode will

adjust the phase of the reflected currents. Orthogonality of modes can be used to

produce circular polarization in the RDRA. Figure 5.4 represents the equivalent

RLC circuit of RDRA, resonant modes excited, and corresponding magnetic

dipoles. Figure 5.5 depicts the even and odd modes generation. Figure 5.6 presents

RDRA HFF model along with its equivalent RLC circuit. Figures 5.7 and 5.8 are

RLC circuits which are used for derivation of resonant frequency and impedance.

5.8 R, L, C Circuits and Resonant Modes 127

with

natural frequencies xc and forced resonance due to excitation eigen-valued xmnp

has been determined along with eigenvector Jmnp . Separation of all frequencies will

be the out come of modes. The second-order differential equation is the general

solution of equivalent antenna (R, L, C) circuit. Fourier solution will provide a

discrete solution of resonance. Je is excitation current or probe current and c is an

propagation constant c a jb. L, C circuit will introduce non-homogeneous or

inhomogeneous matter, x2 will be replaced in this case by x2 l c is replaced by

~c introducing decay. Hzf represents forced resonance mode.

:: q

Lq Rq_ vs t

C

:: 2

where q ddtq2

1

XL jxL; XC

jxC

128 5 Mathematical Analysis of RDRA Amplitude Coefcients

2 1

jxL jxR Qx Vs x

C

Vs x

Q x 2

jxL 1c jxR

Z1

Qxejxt dx

q t

2p

1

R0

1

xL2

xc

1

x p

LC

J se x; y; z Jsx x; ydz d0 ^x Jsy x; ydz d0 ^y

Z

Js dz Jsx x; y; x^x Jsy x; y; x

r H J e r jxE

r E jxlH

r2 E jxlr jxE J e

r2 Ez c2 xEz

When

p

c x jxlr jx ax jbx

r2 Hz c2 xHz

Hz 0; x 0; a; or y 0; b; z 0; d

Ex 0; x 0; a; when z 0; d; Ey 0; y 0; b; z 0; d;

5.8 R, L, C Circuits and Resonant Modes 129

p npx mpy ppd

X 2 2

Hz x; y; z; x p sin sin sin Re C mnp expjxnmpt

abd a b d

r2 c2 x Ez 0

r2 c2 xEz 0

r2 c2 xE ? J e

r2 H r J e r jxjxljH

2

r c2 x H r J e

r2 c2 xE ? J e

2

r ~c2 Hx Jsy d0 z d0

2

r ~c2 Hy Jsx d0 z d0

2

n m2 p2

~c2 x p2 2 2 2 0

a b d

2

2 n m2 p2

jxlr jx p 0

a2 b2 d 2

2

2 n m2 p2

~c xmnp x l jxlr p

2 2

a2 b2 d 2

xmnp xReal mnp jxImg mnp

ejxmnpt ejxReal mnpt exImg mnpt

x2 l ! x2 l jxlr ~c2 x

X

Jsxy x; y x Jsyx x; y xdz d0 Jz n; m; p; xunmp x; y; z

2 2 0

J n; m; p; x Jsxy x; y; x Jsyx x; y; x p sin sin sin dxdy

abd a b d

0 0

5:55

X

r2 ~c2 x Hz Jz mnp; xumnp x; y; z

mnp

X

Hz f Hz mnp; xumnp x; y; z;

mnp

130 5 Mathematical Analysis of RDRA Amplitude Coefcients

n2 m2 p2

Jz nmp; x p2 ~c2 xHz nmp; x

a2 b2 d 2

X Jz nmp; xumnp x; y

Hz f x; y; z x

mnp

~c2 x ~c2 xnmp

where

p npx mpy ppz

2 2

vmnp p cos cos cos Hz

abd a b d

X

Jsy x; y; xd0 z d0 Jy mnp; xvmnp x; y; z

Z a Z b Zd

Jy n; m; p; x JSy x; y; xd0 z d0 vmnp x; y; zdxdydz

0 0 0

^z r? Ez ^zc ^z E? ^z jxl H ?

jxl

r? Ez cE? r? Hz ^z Je r jxeE? :

c

2

c ~c2 x E? jxl r? Hz ^z jxl Je c r? Ez

5.9 Resonant Modes Based on R, L, C Circuits 131

Hence,

jxl jxl c r Ez

E? r Hz ^z 2 Je 2 ?2

c2 ~c2 x ? c ~c2 x c ~c x

Z

di 1

Ii L i dt V

dt c

On differentiating

di2 di 1

L R 0

d2 t dt c

gives the following:

di2 R di 1

2

0

d t L dt LC

R 1

S2 S 0

L LC

q

R2

R L 4 1

LC

S

L 2

r

R2 2 2

p

R L LC

S1

L 2

r

R2 2 2

p

R L LC

S2

L 2

132 5 Mathematical Analysis of RDRA Amplitude Coefcients

Let

R

k

s2L

2 2

R 1

x2

2L Lc

S1 k x 2

S2 k x 1

I A1 es1 t Az es2 t

Now

2

R 1

Care 1 [

2L Lc

2

R 1

Care 2 \

2L Lc

R 2 1

Care 3

2L Lc

Z

Ldi 1

V Ri i dt

dt C

V I s

RI s LsI s

s sC

V

I s

s R Ls sC 1

V

I s

Rs Ls2 C1

5.9 Resonant Modes Based on R, L, C Circuits 133

Since,

q

R 2

R=L

L 4=LC

s1;2

2

V V 1 1

I s

2 Rs 1

L s L LC L s 1 s 2 s s 1 s s 2

1 1

It q es1 t q es2 t

L L2 LC

R 4

L L2 LC

R 4

Zt

dit 1

v R it L it dt

dt C

0

Taking Laplace transform on both the sides gives

v 1

Is R Ls I s i0 I s i0 0

s Cs

v 1

Is R L s

s Cs

1

v Is R s L s2

C

L s2 R s C1 0

134 5 Mathematical Analysis of RDRA Amplitude Coefcients

s

2

R R 1

S

2L 2L LC

s

2

R R 1

Let s1 s

2L 2L LC

s

2

R R 1

and s2

2L 2L LC

R 1

Now; s1 s2 and s1 s2

L LC

v Is Ls2 s s1 s2 s1 s2

v Is Lss s1 s2 s s1

v Is Ls s1 s s2

v 1

Is

L s s1 s s2

v 1 1

Is

Ls1 s2 s s1 s s2

v

it es1 t es2 t

Ls1 s2

r

R 4L

s1 s2 1

L C

v

it q es1 t es2 t

R 1 4L C

v

Let; A1 A2 q

R 1 4L C

it A1 es1 t A2 es2 t

Chapter 6

Mathematical Analysis of Radiation

Pattern of RDRA

Abstract In this chapter, detailed study using mathematical analysis for radiation

pattern of RDRA has been described. RF excitation with proper impedance match

can generate J-current density into surfaces of RDRA, which leads to produce

A-magnetic vector potential and nally E-electric intensity or H-magnetic eld

intensity. Acceleration or deceleration of charge carriers causing current is

mandatory phenomenon for radiations. Wave can only propagate if wave vector

k > kc, where kc is cutoff frequency. The lowest resonance can be termed as

dominant mode and second and third resonances are higher-order modes.

Propagation constant kx np=a, and propagation takes place if kx [ np=a, while

no propagation takes place if kx \np=a. Thus, standing waves inside the resonator

are formed and energy storing will take place. Hence, mode spectrum will result

into corresponding resonant frequency generation. Wave propagation can be well

dened by Helmholtz equation. The Maxwells equations describe the behavior of

electromagnetic elds and form the basis of all EM classical phenomena. Prad

(power radiated) can be evaluated using Parsevals power theorem. The radiated

power is produced by oscillating dipole moments. The current varying in time can

be analyzed by Fourier analysis. If medium is inhomogeneous, wave possesses

exponential growth or decay in some direction. Thus, Poynting vector S shall

give the magnitude and phase of the radiated elds in particular direction.

Keywords Impedance match Current density Magnetic vector potential

Power radiated Poynting vector Persvals power theorem Moat-shaped DRA

6.1 Introduction

RF excitation with proper impedance match can generate J-current density into

surfaces of RDRA, which leads to produce A-magnetic vector potential and nally

E-electric intensity. Acceleration or deceleration of charge carriers causing current

is mandatory phenomenon for radiations. Wave can only propagate if wave vector

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_6

136 6 Mathematical Analysis of Radiation Pattern of RDRA

k > kc, where kc is cutoff frequency and the lowest resonance can be termed as

dominant mode and second and third resonances are higher-order modes.

Propagation constant kx np=a. Propagation takes place if kx [ np=a, while no

propagation takes place kx \np=a. Standing waves inside the resonator are formed

and energy storing will take place. Hence, mode spectrum will result into corre-

sponding resonant frequency generation due to equivalent RLC circuit formation.

Wave propagation can be well dened by Helmholtz equation. The Maxwells

equations describe the behavior of electromagnetic elds and form the basis of all

EM classical phenomenon. Prad (power radiated) can be evaluated using Parsevals

power theorem. The radiated power is produced by oscillating dipole moments. The

current varying in time can be analyzed by Fourier analysis. If medium is inho-

mogeneous, wave possesses exponential growth or decay in some direction. Thus,

Poynting vector S shall give the magnitude and phase of the radiated elds in

particular direction.

Finally, the radiation pattern produced by the surface electric and magnetic

current densities on the RDRA surfaces is computed. PEC walls, the surface electric

current density is Js ^n E.

Then, the far-eld magnetic vector and electric vector potentials are determined

by the usual reactance potential formulae as follows:

Z

l ejkr

Ax; r J s x; r 0 expjk^r r 0 dsr 0 ; 6:1a

4p r

s

and

Z

ejkr

F x; r M s x; r 0 expjk^r r 0 dsr 0 : 6:1b

4p r

s

Lorentz force conditions are applied to determine the far-eld electric scalar and

magnetic scalar potentials as follows:

j

/e x; r div Ax; r

xl

6:2a

k

^r ; Ax; r

xl

j

/m x; r div F x; r

xl

6:2b

k

^r ; F x; r

xl

The far-eld electric and magnetic elds (i.e., up to Order r 1 ) are then

determined as follows:

6.1 Introduction 137

1

E r/e jxA r F; 6:3

1

H r A r/m jxF;

l

jk 2 jk

^r ^r ; A jxA/ ^r F; 6:4

xl

jk

jxA? /m ^r F;

where

A? Ah ^h A/ /;

^

jk

H ? ^r A jxF 1 ;

l

1

S RefE H g:

2

Up to order r12 i.e., value 1/r2 is taken into account from where, the RDRA

radiation resistance is evaluated:

Z

1 2

I Rr lim S ^r r 2 dX;

2 r!1

depends on the frequency.

i(t) and Probe Length dl

!

lI dl ejkr !

A; where A is magnetic vector potential 6:5

4pr

From Helmholtz equation ~

A

E jx ~

A

138 6 Mathematical Analysis of Radiation Pattern of RDRA

2 r

jE j2 x2 ~

A l

; g characteristic impedance:

2g 2g

Z

J r 0 ; xejkjrr j 3 0

0

~ l

A d r ; at source: 6:6

4p jr r 0 j

Volume

We know that radiation pattern can be dened by the electrical eld intensity

Eh ; E/ :

Eh jxAh and Ah ^h A

X

J r 0 ; x J s mnp; r 0 ejxmnpt ; where; r x; y; z 6:7

mnp

jrr0 j

Z

l X J s mnp; r 0 ejxmnp t c

A dsr 0 ; where; ds is surface of RDRA

4p mnp jr r 0 j

Z

l ejkn X 0

J s mnp; r 0 ejxmnp^rr dsr 0

4p jr r 0 j mnp

s

6:8

H/ Eh =g; Hh E/ g:

1 2

Prad j E h j 2 E /

2g

^h ^x cos u cos h ^y sin u cos h ^z sin h

^ ^x sin u ^y cos u

/

l XZ n o

Eh 2

Re Jsx mnp; r0 cos u cos h Jsymnp;r0 sin / cos h Jszmnp;r0 sin h

4pr mnp

s

mnp 0

expjx gx cos / sin h y0 sin / sin h z0 cos h dsr0 ejxmnpt ;

c

6:9a

6.2 Radiation Pattern of RDRA Due to Probe 139

XZ n o xmnp

Eu Re Jsx mnp; r 0 sin / Jsymnp;r0 cos / ej c

mnp

s 6:9b

0 0 0 0 jxmnpt 0

x cos u sin h y sin / sin h z cos h dsr e dsr :

Z

xmnp^r r0

Px ^r jmnp Jsx mnp; r 0 ej c dsr 0 6:10a

s

Z

xmnp^r r0

Py ^r jmnp Jsy mnp; r 0 ej c dsr 0 6:10b

s

Z

xmnprr 0

Pz ^r jmnp Jsz mnp; r 0 ej c dsr 0 6:10c

s

X

Eh Re Px ^r js cos / cos h Py ^r js sin / cos h Pz ^r js sin h ejxst

s

X 6:11

Re Esh ejxst

s

3 2

000

where s mnp 4 001 5 and so on till s 111, similarly

010

X

E/ Re Px ^r js sin / Py ^r jscos/ejxst 6:12

s

2 ( )

jEh j2 E/ 1 X

Esh ejxst Es/ ejxst

2g 2 s

( ) 6:13

1 X jxst

X

jxst

Hsh e Hs/ e

2 s s

140 6 Mathematical Analysis of Radiation Pattern of RDRA

!

1 X X

Es Hm ejxs xm t Es Hm ejxm xs t

4 s s

1X

Esh Hs Es Hsh

4 s

1 X

Re Esh Hs 6:14

2 s

1 ^ E Es ^

^ sh ^

Esh h E / h

2 g g

X jEsh j2 jEs j2

^r ^r ;

s

2g 2g

Poynting vector is dened as radiated power flux per unit solid angle or power

radiated in particular direction in specied angular zone.

H rA

E r ddAt ; scalar and magnetic vector potential from Lorentz gauge

conditions.

S E H ; S is Poynting vector (energy flow or flux).

Prad

Z Input impedance

jIj2

1 Xh

S ^r xs2 jPx ^r js cos / cos h Py ^r js sin / cos h Pz ^r js sin h2

2g mnp

i

xs2 jPx ^r js sin / Py ^r js cos /2 6:15

1 X

S ^r r; h; / xmnp2 fjPx h; /jmnp cos / cos h

2g mnp

6:16

Py h; /jmnp sin / cos h Pz h; /jmnp sin h2

jPx h; /jmnp sin / Py h; /jmnp cos /2

6.4 Moat-Shaped RDRA Radiation Pattern 141

Moat-shaped RDRA is shown in Fig. 6.1a with x, y, and z coordinates, and feed is

given at a/2 position.

In Fig. 6.1b, rectangular moat-shaped RDRA is covered with r copper plate to

reduce resonant frequency.

E(t, x, y, z) is electric eld intensity of RDRA to be computed in time domain

and E(, x, y, z) in frequency domain having a, b, and d dimensions, excited with

feed probe at a2 ; a2 ; 0 point by I0 cos xt RF current.

A Az^z (due to RF excitation current I0 cos xt along length d inserted into the

RDRA).

Hence, magnetic vector potential can be written as follows:

Zd

lI0 ejkjra=2^xb=2^yn^zj

Az x; x; y; z dn; 6:17

4p jr a=2^x b=2^y n^zj

0

Fig. 6.1 a Moat-shaped RDRA. b RDRA moat cover with rectangular copper plate to reduce

resonant frequency

142 6 Mathematical Analysis of Radiation Pattern of RDRA

Let C 4p

l

; k x=c and n = variable probe length.

9

Zd expfjk x a= 2 y a= 2 z n2 > =

2 2

Az CI0 1=2 >dn 6:18

x a=22 y a=22 z n2 ;

0

CI0 ejkr

Az Ph0 ; 0 ; 6:19

r

Here, it is assumed that probe is very small as compared to RDRA.

x a=22 y a=22 z n2

2

x a=22 y a=2 z2 2zn

2

x a=22 y a=2 z2 d 2

1=2

r x a=22 y a=22 z2

where

r = distance from the points (x, y, z) in the center of the feed probe

a= ; a= ; 0

2 2

1=2

x a=22 y a=22 z2

r2 2zn1=2 r1 zn r0 2 r zn=r0 :

Hence, magnetic vector potential due to source inside RDRA can be computed

as follows:

Zd

CI0 ejkr0 jkzn

Az exp dn; where I0 probe RF current:

r0 r0

0

jkr0 exp jkzn=r

CI0 e 0 n d

n 0 i.e., variable probe length:

r0 jkz=

r0

6:20

jkzd= 1

CI0 jkr0 exp r0

e

r0 jkz=

r0

exp jkzd= 2j sin kzd=

CI0 jkr0 2r0 2r0

e

r0 jkz=

r0

6.4 Moat-Shaped RDRA Radiation Pattern 143

n o sin kzd=2r

Az 2CI0 exp jk r0 zd=2r0

0

kz

where, z r cos h:

jkd cos h0

cos h0 sin kd

Az x; x; y; z CI 0 expjkr0 exp 2 2

6:21

kr0 cos h0

here, (r, , ) are spherical polar coordinates of (x, y, z) so as to relate a=2; a=2; 0 ,

the probe insertion point. Hence, magnetic vector potential can be expressed as

follows:

x 2 2 a

Et; x; y; z 3

jPh0 jsinxt kr0 Wh0 xa=2 y a=2 x z^x

r0 2

x y =2 za

jPh0 jsinxt kr0 Wh0 ^y

r0 3

6:22

k x a2

Bt; x; y; z jPh0 jsinxt kr0 Wh0 ^y

r2

0 a

k y2 6:23

jPh0 jsinxt kr0 Wh0 ^x

r0 2

Ph0 a a

k sin xt kr 0 W h 0 y ^

x x ^y

r0 2 2 2

1

S RefE H g

2

1

Up to O r2 from where the radiator resistance is evaluated as

Z

1 2

I Rr lim S ^r r 2 dX

2 r!1

Hence, this completes the solution for radiation pattern of RDRA.

144 6 Mathematical Analysis of Radiation Pattern of RDRA

The quality factor Q of the RDRA can be evaluated by comparing the power

radiated Prad 12 I 2 Rr with the average electromagnetic energy (W) stored with the

RDRA as follows:

Z

1

W x E; E lH; H dx dy dz 6:24

4

0;a0;b0;c

The average energy stored per unit cycle with the RDRA is

W x x

Px W x 6:25

2p= 2p

x

2xW x

Qx ;

jI xj2 Rr x

The quality factor of a resonant mode measures how sharp its resonance is. As

per conservation of energy,

Z Z

jEj2 dv jH j2 dv

(time) average magnetic energy will be equal to electric energy inside the resonator.

The time-averaged energy dissipated in the walls of RDRA in unit time can be

calculated as of energy into walls from the electromagnetic elds in the cavity

normal component of energy based on the boundary conditions as energy flux

density as follows:

C

S ReE H 6:27

8p

I

c

RejH j2 df

8p

p

The resonant frequency is reduced by l

p

If x ! x l

6.5 Quality Factor of RDRA 145

x0 0 00

2jx00 j = quality factor (Q), x is real frequency, and x is imaginary frequency.

0 00

Complex freq x x jx

Z Z

Ea Eb dv Ha Hb dv 0

Resonator lled with non-absorbing dielectric, for which and l differ from

p

unity by replacing x by x l and E by E, and H by lH.

The (time) average energy flux through surface is

c

S Re Et Ht 6:28

8p

c

where S E H .

4p

If Q of heat evolved per unit time and volumes

x 00 2

Q E l00 H 2 6:29

4p

Bar denotes time-average exciting frequency, must be exactly equal to the chosen

resonance frequency, and is required to establish eld conguration inside res-

onator. This results in dissipation of energy in the cavity walls and dielectric lling

of the cavity resonator. A measure of the sharpen of response of the cavity to

external excitation is quality of the cavity. This is dened as 2p times the ratio of

the time-averaged energy stored in the cavity to the energy dissipated.

stored energyfWxg

Q x0 6:30

power lossI I Rr

moat

146 6 Mathematical Analysis of Radiation Pattern of RDRA

where x0 Resonant frequency oscillations of elds are damped and time depen-

dent. Change in frequency Dx to occur based on superposition of frequencies:

x x0 Dx

1 1 Z

E t p E xejxt dx

2p 1

shown in Fig. 6.2 and also compute quality factor of a RDRA having dimensions

10 10 10 mm3 with dielectric constant 10 and probe current 10 mA.

Chapter 7

Rectangular DRA Higher-Order Modes

and Experimentations

Abstract In this chapter, rectangular DRA higher-order modes have been realized

by mathematical modeling. Resonant modes are seen with experimentations in

anechoic chamber. These resonant modes impart physical insight into the radiating

phenomenon of the antenna. Knowledge of modes can be boon to the antenna

designer. If antenna resonant modes are known, radiation parameters can be steered.

There are two types of modes and they are dominant and higher-order modes. The

dominant mode corresponds to the lowest resonant frequency. These higher-order

modes can be generated either by increasing electrical length of RDRA or by

applying higher excitation frequency. The resonant frequencies of the modes are

represented by eigenvalues and currents by eigenvectors. Radiating behavior of the

antenna can be predicted by modes. They can also help to determine input exci-

tation point. Moreover, having in mind the current distribution of the modes, the

geometry of the antenna can be modied. The aspect ratio is the important

parameter in RDRA. Devising control on aspect ratio can alter resonant frequency,

gain, and bandwidth. RDRA has two fold design flexibility because of two aspect

ratios. The resonant modes of RDRA can be described with an equivalent

sequential RLC circuit having different sequential LC values. Thus, they form many

series-tuned resonant circuits. The superposition of these modes generally give rise

to resulting or weighted resonant frequency. The top-loading RDRA has been

completely modeled. Antenna gain and bandwidth enhancement techniques

have been worked out with examples.

Keywords Aspect ratio RLC circuit Tuned cavity Weighted sum Resonant

frequency Eigen frequency Design flexibility Top-loading RDRA Gain and

bandwidth enhancement

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_7

148 7 Rectangular DRA Higher-Order Modes and Experimentations

Resonant modes impart physical insight into the radiating phenomenon of the

antenna. Knowledge of modes can be boon to the antenna designer. If antenna

resonant modes are known, radiation parameters can be steered. Any of the

antennas have two types of modes. They are dominant and higher-order modes. The

dominant mode corresponds to the lowest resonant frequency. Other than dominant

frequency, all higher resonant frequencies are higher-order modes. These

higher-order modes can be generated either by increasing electrical length of RDRA

or by applying higher excitation frequency.

The resonant frequencies of the modes are represented by eigenvalues and

currents by eigenvectors. Radiating behavior of the antenna can be predicted by

modes. They can also help to determine input excitation point. Moreover, having in

mind the current distribution of the modes, the geometry of the antenna can be

modied. The aspect ratio is the important parameter in RDRA. Devising control on

aspect ratio can alter resonant frequency, gain, and bandwidth. RDRA has two fold

design flexibility because of two aspect ratios.

The resonant modes of RDRA can be described with an equivalent sequential

RLC circuit having different sequential LC values. Thus, they form many

series-tuned resonant circuits. The superposition of these modes generally gives rise

to resulting or weighted resonant frequency. The modes are dened as E and H elds

pattern inside a device, whose EM wave propagation is governed by Maxwells

equations under certain boundary conditions. RF input excitation currents get dis-

tributed on RDRA surfaces. Thus, weighted sum of eigen currents or superposition

of all these currents inside device is the resultant mode at any instant of time.

The resonant modes are state of excited elds at any instant inside the device,

generally classied as transverse electric (TE), transverse magnetic (TM) and

hybrid electromagnetic (HEM), dominant modes and higher modes. TE modes will

have only Hz component as propagating elds. TM modes will have only Ez

component as propagating elds. These propagating elds are longitudinal elds.

HEM has hybrid mode and will have both Ez and Hz components simultaneously as

propagating elds at any instant of time. These eld perturbations form a particular

excited resonant mode in the device.

In the literature, stacking of the RDRA has been used for enhancement in the

directivity of the antenna by Petosa [1]. This can be achieved by devising proper

control on higher-order modes. Higher modes correspond to higher resonant fre-

quency and higher antenna gain. RDRA higher-order modes and hybrid modes are

useful and provide design space to antenna designers, but congure complex elds

structure. The generation of higher modes mainly depends on RF excitation, device

dimensions, dielectric material, perturbation, and coupling used in RDRA.

An aperture-coupled microstrip slot feed RDRA is discussed in this chapter. This

has the advantage of isolating the feeding network from the radiating element. Aspect

ratio can be changed by changing the RDRA dimensions a, b, and d. This will have

impact on resonant modes, and thus, change in resonant frequency will take place.

7.1 Introduction to Higher Modes 149

Fig. 7.1 a RDRA higher modes. b RDRA mode generated. c RDRA mode control due to dipole

moment. d S11 of RDRA with mode merging. e Higher-order modes eld conguration

the operating mode. Once the dimensions of RDRA are xed, modes can also be

altered by excitation frequency. Operating frequency of RDRA has inverse rela-

tionship with permittivity of the material. The device size can be minimized by using

higher permittivity material. Figure 7.1 shows the resonant mode structure.

RDRA shown in Fig. 7.1a has been made with alternate layers of RT-Duroid and

FR-4 dielectric materials having permittivity 10.2 and 4.4. These dielectric mate-

rials are easily available. The fabrication is also simple. The dimensions of these

sheets are 6 6 10 mm3 and 6 6 0.8 mm3, respectively. Figure 7.1b shows

resonant modes. Figure 7.1c shows how these dipoles are broken by introducing air

between these stacking layers. Figure 7.1d presents mode merging due to proper

dipole control, i.e., merging these modes by removing one stacking layer and

creating airspace between these dipoles. This has enhanced the antenna gain.

Figure 7.1e shows higher-order modes in RDRA. Figure 7.2 presents prototype

RDRA with VNA feed probe. Figure 7.3 shows modes or eld pattern. Figures 7.4

and 7.5 show odd and even modes. Figure 7.6 represents structure, and Fig. 7.7

represents gain plot. Figure 7.8 shows microstripline used in RDRA feed.

Figure 7.9 shows RDRA with top loading. Figure 7.10 shows H and E elds

150 7 Rectangular DRA Higher-Order Modes and Experimentations

7.1 Introduction to Higher Modes 151

Lumped port

W (slot)

Fig. 7.10 a H eld distribution inside RDRA and b E eld distribution inside RDRA

152 7 Rectangular DRA Higher-Order Modes and Experimentations

-4.00 0.00

dB(GainTotal)

-6.00

dB(S(1,1))

-10.00

Name X Y

m2 30.0000 5.1616

-20.00 Curve Info

-10.00 dB(GainTotal)

Setup1 : LastAdaptive

Freq='12GHz' Phi='0deg'

-30.00 dB(GainTotal)

-12.00

Setup1 : LastAdaptive

Freq='12GHz' Phi='90deg'

-14.00 -40.00

8.00 10.50 13.00 15.50 18.00 -200.00 -100.00 0.00 100.00 200.00

Freq [GHz] Theta [deg]

Fig. 7.11 a Return loss at f = 11 GHz and b radiation pattern showing 5.16 dB gain of the antenna

at f = 12 GHz

(a) (b)

-2.00

(c) (d) 10.00

m2

-4.00 5.00

m1

dB(GainTotal)

-6.00

-0.00

dB(S(1,1))

-8.00

-5.00

-10.00 Name X Y

-12.00 m2 30.0000 5.3275

Curve Info

-15.00 dB(GainTotal)

-14.00 Setup1 : LastAdaptive

Freq='10GHz' Phi='0deg'

-16.00 -20.00 dB(GainTotal)

Setup1 : LastAdaptive

Freq='10GHz' Phi='90deg'

-18.00 -25.00

8.00 10.50 13.00 15.50 18.00 20.00 -200.00 -100.00 0.00 100.00 200.00

Freq [GHz] Theta [deg]

Fig. 7.12 TE111 mode at frequency 10 GHz. a H eld distribution, b E eld distribution, c return

loss, d gain

distributions inside RDRA. Figure 7.11 shows the plot of return loss at f = 11 GHz

and radiation pattern. Figure 7.12 shows TE111 mode at a frequency of 10 GHz.

Figures 7.13, 7.14, 7.15 and 7.16 show resonant modes and Fig. 7.17 TE113 mode

at a frequency of 10 GHz. Figure 7.18 shows TE115 mode at a frequency of 12 GHz.

Figure 7.19 depicted TE117 resonant mode at frequency 15 GHz. Figures 7.20, 7.21,

7.1 Introduction to Higher Modes 153

(a) (b)

-5.00

5.00

dB(GainTotal)

dB(S(1,1))

-7.50

0.00

-10.00

-5.00

Name X Y

-10.00 Curve Info

dB(GainTotal)

Setup1 : LastAdaptive

-15.00 -15.00 Freq='12GHz' Phi='0deg'

dB(GainTotal)

Setup1 : LastAdaptive

Freq='12GHz' Phi='90deg'

-17.50 -20.00

8.00 10.00 12.00 14.00 16.00 18.00 20.00 -200.00 -100.00 0.00 100.00 200.00

Freq [GHz] Theta [deg]

Fig. 7.13 TE113 mode at frequency 12 GHz. a H eld distribution, b E eld distribution, c return

loss, d gain

(a) (b)

(c) 10.00

m1

m2

5.00

0.00

dB(GainTotal)

-5.00

-10.00

Name X Y

m1 -20.0000 6.0721

-15.00 m2 20.0000 5.3982

Curve Info

-20.00 dB(GainTotal)

Setup1 : LastAdaptive

Freq='15GHz' Phi='0deg'

-25.00 dB(GainTotal)

Setup1 : LastAdaptive

Freq='15GHz' Phi='90deg'

-30.00

-200.00 -100.00 0.00 100.00 200.00

Theta [deg]

Fig. 7.14 TE115 mode at frequency 15 GHz. a H eld distribution, b E eld distribution, c gain

154 7 Rectangular DRA Higher-Order Modes and Experimentations

frequency plot

5.00

0.00

dB(GainTotal)

-5.00

-10.00

Curve Info

dB(GainTotal)

-15.00 Setup1 : Sweep

Phi='0deg' Theta='0deg'

-20.00

8.00 10.00 12.00 14.00 16.00 18.00 20.00

Freq [GHz]

7.22, 7.23, 7.24, 7.25, 7.26, 7.27, 7.28, 7.29, 7.30, 7.31 and 7.32 show parameter

measured in anechoic chamber and HFSS simulated results. Working is mentioned

below each gures.

Resonant modes take the real orthogonal basis for currents on the antenna

surfaces.

In this chapter, mechanism for mode generation and their possible control in

RDRA are discussed. These are validated with simulated and experimental results

using prototype models. Figure 7.2 shows the prototype of RDRA. VNA probes are

connected in order to take measurements. Top-loading RDRA is used for genera-

tion of higher-order modes. These higher-order modes are of even and odd types. If

RDRA design is isolated even as well as odd modes will be available, i.e., both

even as well as odd modes will be present in isolated DRA case. RDRA once

Fig. 7.16 Spacing adjustment between short magnetic dipoles placed at the center of each mode

7.1 Introduction to Higher Modes 155

(a) (b)

0.00

-10.00

dB(GainTotal)

dB(S(1,1))

-10.00

-20.00

Name X Y

Curve Info

dB(GainTotal)

-30.00 Setup1 : LastAdaptive

-30.00 Freq='10GHz' Phi='0deg'

dB(GainTotal)

Setup1 : LastAdaptive

Freq='10GHz' Phi='90deg'

-40.00 -40.00

8.00 10.50 13.00 15.50 18.00 -200.00 -100.00 0.00 100.00 200.00

Freq [GHz] Theta [deg]

Fig. 7.17 TE113 mode at frequency 10 GHz. a H eld distribution, b E eld distribution, c return

loss, d gain

extended with ground plane, only odd modes will be generated, because even

modes get grounded. Thus, ground plane canceled out even modes. The E and

H elds patterns are shown in Fig. 7.3.

Higher-order even and odd modes are shown in Figs. 7.4 and 7.5. These modes

can further be identied as TE/TM/HEM.

If H eld is propagating, then it is TE mode. By contrary, if E eld is propa-

gating, then it is TM mode. Also, when both types of elds, E and H, are excited

simultaneously, then it is HEM mode. HEM modes are most advantageous but have

complex structure. The detailed analysis of hybrid modes is described later in

Chap. 10. Resonant modes can be shifted, merged, and independently controlled by

different techniques. Increasing RDRA electrical length and input excitation fre-

quency can generate higher-order modes into RDRA.

156 7 Rectangular DRA Higher-Order Modes and Experimentations

(a) (b)

(c) 5.00 m1 m2

0.00

dB(GainTotal)

-5.00

Name X Y

m1 -40.0000 4.2889

-10.00 m2 40.0000 4.0314

Curve Info

dB(GainTotal)

Setup1 : LastAdaptive

-15.00 Freq='15GHz' Phi='0deg'

dB(GainTotal)

Setup1 : LastAdaptive

Freq='15GHz' Phi='90deg'

-20.00

-200.00 -100.00 0.00 100.00 200.00

Theta [deg]

Fig. 7.18 TE115 mode at frequency 12 GHz. a H eld distribution, b E eld distribution, c gain

The structure of the antenna is shown in Fig. 7.6. Slot and microstrip are shown

in Figs. 7.7 and 7.8. The feed is aperture-coupled. The substrate rectangular plane

of 50 50 mm with a thickness of 0.6 mm was used. FR4 was used as RDRA

substrate, and RDRA with a dielectric constant (permittivity) of 10.2 was placed on

top of the substrate. The width of the microstrip used was 1.15 mm. Slot dimen-

sions were 6 mm in length and 1 mm in width. RDRA dimensions were

6 6 5 mm3.

7.2 Resonant Frequency and RDRA Structure 157

(a) (b)

dB(GainTotal)

m1

m2 Setup1 : LastAdaptive

Freq='12GHz' Phi='0deg'

5.00 dB(GainTotal)

Setup1 : LastAdaptive

dB(GainTotal)

Freq='12GHz' Phi='90deg'

0.00

-5.00

-10.00

-15.00 Name X Y

m1 -20.0000 6.1242

m2 20.0000 5.7701

-20.00

-200.00 -100.00 0.00 100.00 200.00

Theta [deg]

Fig. 7.19 TE117 mode at frequency 15 GHz. a H eld distribution, b E eld distribution, c gain

frequency plot at h = 15 mm

7.50

5.00

2.50

dB(GainTotal)

0.00

-2.50

-5.00

Curve Info

-7.50

dB(GainTotal)

Setup1 : Sweep

-10.00 Phi='0deg' Theta='0deg'

-12.50

8.00 10.50 13.00 15.50 18.00 20.00

Freq [GHz]

158 7 Rectangular DRA Higher-Order Modes and Experimentations

(a) (b)

m1

(c) 0.00 (d) 10.00

5.00

-10.00 0.00

dB(GainTotal)

dB(S(1,1))

-5.00

-20.00 -10.00

Name X Y

Curve Info

dB(GainTotal)

-30.00 -20.00 Setup1 : LastAdaptive

Freq='9.5GHz' Phi='0deg'

-25.00 dB(GainTotal)

Setup1 : LastAdaptive

Freq='9.5GHz' Phi='90deg'

-40.00 -30.00

8.00 10.00 12.00 14.00 16.00 18.00 20.00 -200.00 -100.00 0.00 100.00 200.00

Freq [GHz] Theta [deg]

Fig. 7.21 TE115 mode at frequency 9.5 GHz. a H eld distribution, b E eld distribution, c return

loss, d gain

TE11 resonant mode in rectangular DRA the elds can be dened using dielectric

waveguide model depending upon given boundry conditions

kx2 ky2

Hx coskx x cosky y coskz z 7:1

jxl0

ky kx

Hy sinkx x sinky y coskz z 7:2

jxl0

kz kx

Hz sinkx x cosky y sinkz z 7:3

jxl0

Ex 0

7:4

Ey kz coskx x cosky y sinkz z

7.2 Resonant Frequency and RDRA Structure 159

(a) (b)

m1

(c) 10.00

5.00

0.00

dB(GainTotal)

-5.00

-10.00

Name X Y

Curve Info

dB(GainTotal)

-20.00 Setup1 : LastAdaptive

Freq='10.5GHz' Phi='0deg'

-25.00 dB(GainTotal)

Setup1 : LastAdaptive

Freq='10.5GHz' Phi='90deg'

-30.00

-200.00 -100.00 0.00 100.00 200.00

Theta [deg]

Fig. 7.22 TE117 mode at frequency 10.5 GHz. a H eld distribution, b E eld distribution, c gain

q

kx d

kx tan er 1k02 kx2 7:6

2

The resonant frequency and propagation constant can be determined from the

transcendental equation.

The characteristic equation is as follows:

So, the resonant frequency can be obtained for grounded RDRA as follows:

r

m2 n2 p 2

c

fmnp p : 7:8

2 er a b 2d

160 7 Rectangular DRA Higher-Order Modes and Experimentations

(a) (b)

m1 m2 m1 -30.0000 5.7498

5.00 m2 30.0000 6.1212

-0.00

dB(GainTotal)

-5.00

-10.00

Curve Info

-15.00 dB(GainTotal)

Setup1 : LastAdaptive

Freq='12.55GHz' Phi='0deg'

-20.00 dB(GainTotal)

Setup1 : LastAdaptive

Freq='12.55GHz' Phi='90deg'

-25.00

-200.00 -100.00 0.00 100.00 200.00

Theta [deg]

Fig. 7.23 TE1,1,11 mode at frequency 15 GHz. a H eld distribution, b E eld distribution, c gain

height

0.00

Gain [dB]

-10.00

-20.00

-30.00

-40.00

-200.00 -100.00 0.00 100.00 200.00

Theta [deg]

7.2 Resonant Frequency and RDRA Structure 161

r

m2 n2 p2

c

fmnp p ; 7:9

2 er a b d

constant k0 can be determined in terms of kx. Transcendental equation can be solved

for kz0 using kx. This solution can be obtained using MATLAB for xed value of n,

p, and d. kx will now contain a in place of a. a is the extended electrical length due

to fringing effects. Hence, kx is the complete solution of transcendental equation.

Example 1 Let us determine the resonant frequency for dominant and higher-order

modes of RDRA with given dimensions and dielectric constant:

162 7 Rectangular DRA Higher-Order Modes and Experimentations

7.2 Resonant Frequency and RDRA Structure 163

164 7 Rectangular DRA Higher-Order Modes and Experimentations

7.2 Resonant Frequency and RDRA Structure 165

RDRA and design dimensions

1 Ground plane 20 30

2 Substrate 20 30 0.8

3 RDRA 4.6 9 10.8

4 Width of microstrip 2.4

5 Length of stub and microstrip 18.693

6 Ground slot (l w) 3.743 0.404

r

m2 n2 p2

c=2

fr p

er a b d

s

3 108 1 1 1

p 106

2 10 10 10 10

r p p

3 108 3 3 3 1011 3 3

p 10 3

1010

2 10 10 2 10 2

fr 2:598 1010 Hz 25:98 GHz

s

3 108 1 1 9

Higher-order mode: fr113 p 106

2 10 10 10 10

p

3 108 103 p 3 1011 p 3 11

p 11 11 1010

2 10 2 10 2

fr 4:9749 1010 Hz 49:749 GHz

in the last two decades due to several attractive characteristics, such as design

flexibility, high gain, and wide bandwidth. RDRA has two different aspect ratios (b/

a, d/a), high radiation efciency, light weight, and low prole. In contrast, patch

antenna has low gain, less bandwidth, and design flexibility.

The resonant modes are E and H elds pattern. They can be expressed as

follows:

X

Ez umnp x; y; z Re C mnp ejxmnp t ; where; Cmnp are amplitude coefficients:

mnp

7:10

where

166 7 Rectangular DRA Higher-Order Modes and Experimentations

umnp x; y; z p cos cos sin ; orthogonal Fourier basis function:

abd a b d

7:11

v

u

ubmnp2 amnp2

Cmnp u

t hp ppdi2 ; amplitude coefficient: 7:12

p2

sin d

d

" #

amnp cos/mnp bmnp sin/mnp

1

wmnp tan ; Phase: 7:13

amnp sin/mnp bmnp cos/mnp

RDRA as shown in Fig. 7.9, for generating higher-order modes. The even as well as

odd modes can be generated even with ground plane. It has also been studied that

even modes (in z-direction) were short-circuited, when RDRA was placed on a

ground plane. Short magnetic dipoles are basis for generation of these resonant

modes. Nearly, /3, /2, and heights of the dielectric resonator generated TE11,

TE13 and TE15 (0 < 1) modes. The gain is found to be increasing in higher

modes. This is also evident from the ndings that gain of RDRA starts decreasing or

reducing even in higher modes, when magnetic dipoles start overlapping. This

overlapping of dipoles can be seen when the wavelength used is very small.

Top loading excited both even and odd modes. Simulations have shown that 1st

and 3rd resonances got shifted toward 2nd resonance, when the space s between

top and bottom RDRAs varies. Merging of neighboring resonance modes could be

done using this method. This is an excellent phenomenon, which can be used for

bandwidth enhancement. This merging of bands helped to increase the antenna

bandwidth. Thus, existing patch antenna gain and bandwidth can also be increased

by using the concept of higher-order modes. Blocking or shifting of any modes has

become possible in RDRA.

E and H elds perturbations in RDRA can be introduced by carrying out small

changes in the structure, or this can be obtained by input excitation currents. This

perturbation gets converted into eigenvector or eigenvalues. The perturbations are

proportional to eigenvector and resonant mode. The number of modes is directly

related to number of lobes occuring in radiation patterns. There are two ways in which

the number of modes can be increased in RDRA: One is by increasing the RDRA

dimensions a, b, and d, and the other is by increasing the input excitation frequency.

7.5 Simulated HFSS Results 167

From Fig. 7.10, it is evident that single resonant mode, as one half-wave variation

take place in x-direction.

The gain of the antenna got enhanced due to increase of RDRA height. TE111

mode resonant frequency is 10 GHz, TE113 mode frequency is 12 GHz and TE115

frequency is 15 GHz. Figure 7.12 shows the magnetic and electric eld distributions

inside RDRA. The return loss and radiation pattern are shown in that gure.

It has been seen from Fig. 7.13 that gain has been enhanced. Figure 7.14 shows

TE115 mode, which has higher resonant frequency. Figure 7.15 shows gain versus

frequency plot. Figure 7.16 depicts the spacing between short magnetic dipoles.

The above results obtained from the analysis of RDRA carried out revealed that

higher-order modes offer high gain until dipole overlapping does not take place.

The decrease in gain due to overlapping of short magnetic dipoles takes place. This

will happen when there is a less spacing between two short magnetic dipoles.

Hence, minimum spacing between short magnetic dipoles must be equal to 0.4. If

the spacing between short magnetic dipoles is less than this limit, then the gain will

be reduced. This is depicted by simulations in Fig. 7.14. TE113 gain has been

reduced even at TE115 as shown in Fig. 7.18. Now, if we obtain TE115 mode with

increase in RDRA height, then more gain can be obtained. This is the reason why

gain at TE115 is less than TE113 as shown in Figs. 7.14 and 7.15.

At h k=2

When h = 15 mm, three modes got generated, i.e., TE113, TE115, and TE117 cor-

responding to 10, 12, and 15 GHz, respectively. It is clear that gain has been

decreased at higher modes due to the reason explained earlier that spacing between

short dipoles placed at the center of the eld is less than 0.4 . At frequency 10 GHz

inside RDRA, there is proper spacing between these dipoles; hence, the gain is

maximum. Various excited modes are shown in Figs. 7.16, 7.17, 7.18, 7.19, 7.20,

7.21, 7.22 and 7.23 are excited, i.e., resonant modes in RDRA. Their results have

also been taken on S11 plots.

hk

Here, the height of RDRA has been chosen as h = 30 mm, and mode was

operating at 10.5 GHz. The highest gain was due to the same concept of spacing of

short magnetic dipoles. From Figs. 7.17, 7.18, 7.19 and 7.20, very important fact is

noticed that when spacing between short magnetic dipoles was reduced, then the

order of mode becomes high, while the power of main lobe was distributed to side

lobes. Thus, the gain of the antenna was reduced at higher mode if the spacing is

less than 0.4 .

In the above gures, generation of higher modes, limitation, and their effect on

antenna gain have been clearly shown.

168 7 Rectangular DRA Higher-Order Modes and Experimentations

The comparison between three RDRAs of different heights have been made. It was

noticed that RDRA having less height and operating at lower mode offers less gain

but higher bandwidth. On increasing the height of RDRA, the gain of the antenna is

found to be higher along with directivity, thus narrowing the beam width.

Figure 7.24 shows that gain is increasing, when the height of RDRA is increased.

RDRA of height /3, /2, and operating around 1115 GHz consists of TE111,

TE113, and TE115 modes. This fullls the requirement of separation of magnetic

elds by spacing 0.4 . But when RDRA with same height operates at higher

frequency, then the spacing between dipoles is reduced. The power of main lobe is

distributed to the side lobes, which creates the loss of antenna power and gain.

Hence, any desired resonant modes inside the device can be excited for desired

radiation pattern at known resonant frequency. The higher modes amplitude coef-

cients equation has been developed. Modes can be used to visualize corresponding

radiation pattern and polarization of the antenna. Modes give physical insight into

eigenvalue for determining resonant frequency and feeding point for 50

impedance.

Prototype RDRA was made, and it was tested inside the anechoic chamber using

VNA. Results for radiation pattern and other antenna parameters have been taken

and are shown in Figs. 7.26, 7.27, 7.28, 7.29, 7.30, 7.31 and 7.32. Each gure is

captioned below for the results.

It was seen in RDRA of particular height, more number of higher-order mode

can be excited by applying another excitation on the top loaded RDRA as shown in

Fig 7.3. The reduced spacing s between top and bottom RDRAs, merged even

modes, thus increased bandwidth of the antenna. The RDRA under top loading

converted few odd modes to nearest even mode. Thus, both even and odd modes

were made available due to top loading. Thus, spacing s seems to control band-

width of RDRA. High gain, miniaturization, high bandwidth, directive antenna can

be designed by having proper control or maneuvering resonant modes.

The design of this antenna offers wide scope of achieving wide bandwidth along

with high gain. The application of this antenna includes satellite tracking, air trafc

control Wi-, Wi-max, and mobile communication.

7.8 Prototype and Anechoic Chamber Experimentations 169

By developing control on modes, we can control beam width of antenna and can

restrict the reception of signal to a particular area and hence it can be used for

military applications. Presently, we face the problem of TV signal reception during

rainy season, due to the absorption of signal by rain drops due to signal being single

polarized either TE or TM. This can be minimized by application of dual polarized

or hybrid antenna. The other application could be miniaturization of antenna. By

keeping the dimensions of antenna xed, the mode of antenna can be changed by

changing the permittivity of RDRA and thereby changing the frequency. To

automate the mode generation, microcontroller-based lookup table can generate

possible combinations of bandwidth, gain, and frequency.

Applications

The merging of adjacent bands or neighboring modes of RDRA can be used for

enhancement of the bandwidth. By varying the aspect ratio, three resonant bands

can be obtained for useful operation as shown in Fig. 7.33.

The dimensions of RDRA are given in Table 7.2. Figure 7.33b shows the return loss

of the antenna with three bands resonating at 2.89 GHz at dominant mode, i.e., TE111

mode and at 3.61 GHz for TE121 mode and at 4.6 GHz for TE131. Figure 7.34 shows H

and E elds distributions inside RDRA. The direction of the electric eld is indicated

by arrow.

Figure 7.35 depicted that the lower gain at lower mode and high gain at higher

modes.

The effect of the air gap between ground plane and RDRA is shown in Fig. 7.36.

Table 7.3 shows the variation in the resonant bands of the antenna. Effect of

introducing the gap between RDRA and ground plane is depicted in Fig. 7.36.

Results obtained by simulation along with the results obtained by calculations

clearly indicate the effect of air gap. The modes are spreading as the frequency of

the modes is shifted in forward direction with respect to increase in the gap.

Tables 7.1, 7.2, 7.3, 7.4, 7.5 and 7.6 are the results tabulated for various

simulations.

170 7 Rectangular DRA Higher-Order Modes and Experimentations

0.00

(a) (b)

-2.50

-5.00

dB(S(1,1))

-7.50

-10.00

-12.50 m1 m2

-15.00

Name X Y

m1 2.8963 -13.4462

-17.50 m2 3.6187 -13.2545 m3

m3 4.6120 -19.2167

-20.00

2.50 3.00 3.50 4.00 4.50 5.00

Freq [GHz]

calculated resonant

frequencies of the modes in TE111 Simulated 2.89 2.83 2.88

DR Calculated 2.84 2.85 2.87

TE112 Simulated 3.67 3.83 Omitted

Calculated 3.69 3.71 3.74

TE113 Simulated 4.64 4.72 4.81

Calculated 4.62 4.69 4.73

When two asymmetrical wells are inserted, then these modes start to merge together

and all the bands are shifted. This is shown by return loss graph in Fig. 7.38.

Comparison between asymmetrical wells and without asymmetrical wells has been

made. This is to note that by adding wells, higher frequency bands get shifted more

as compared to lower frequency bands as shown in Fig. 7.40. Dimensions of the

structure are shown in Table 7.4. The results were taken for various values of width

of well as 0.5, 1.1, 2.0 mm respectively are shown in Fig. 7.38.

In this way of merging of modes takes place when air gap is inserted in the structure

of RDRA as shown in Fig. 7.41 then we get broader bandwidth. Plot for reflection

coefcient in Fig. 3.9 shows clearly the effect of moat in the structure. All the

dimensions of the structure are shown in Table 7.3.

7.13 Effect of a/b and d/b Aspect Ratio 171

Fig. 7.34 Magnetic and electric eld distributions inside DR. a H eld TE111 mode, b E eld

TE111 mode, c H eld TE112 mode, d E eld TE112 mode, e H eld TE113 mode, f E eld TE113

mode

The effect of length and width of RDRA is such that if we increase the dimensions

then there can be large number of modes generated. The effect of a/b and d/b ratio

has been speculated in the manner such that when the ratio a/b is increased, the

modes come closer to each other and merged, and when the ratio d/b is increased,

resonant frequencies of all modes are diverged. Further, if the ratio d/b is reduced,

then the modes are merged. So here, we increased the a/b ratio and reduced the d/

b ratio, and then we pointed aristocratically that intensied changes like modes have

been merged to increase the bandwidth of the device. The important thing to note is

that mode TE112 has been merged, and there are the resonant frequencies of modes

TE111 and TE113 only. Figure 7.31 reflects the effect of the overall process.

172 7 Rectangular DRA Higher-Order Modes and Experimentations

of the antenna for frequency

m1

range 5.00

0.00

dB(GainTotal)

-5.00

-10.00

-15.00

-20.00

Name X Y

-25.00

m1 2.8930 4.9158

m2 4.6572 8.2124

-30.00

m3

between RDRA and ground

plane m2

-10.00

m1

-20.00

S11 [dB]

-30.00

-40.00

-50.00

2.50 3.00 3.50 4.00 4.50 5.00

Freq [GHz]

Table 7.3 Modes and their Mode Resonant frequency (GHz) Gain (dB)

resonant frequencies

TE111 4.56 5.2

TE112 4.96

TE113 5.56 8.78

Figures 7.35, 7.36, 7.37, 7.38, 7.39, 7.40, 7.41, 7.42, 7.43 and 7.44 presented effect

on change in aspect ratios. Tables 7.3 and 7.4 indicated design parameters.

Tables 7.5 and 7.6 show resonant modes.

7.13 Effect of a/b and d/b Aspect Ratio 173

Dimension

Parameter/structure Aperture-coupled DR with two DR after increasing in DR with moat

DR asymmetrical a/b and decreasing in

wells d/b ratio

Ground plane and 70 70 and FR_4 70 70 and 70 70 and FR_4 70 70 and

substrate (mm) and epoxy (4,4) FR_4 epoxy epoxy (4,4) FR_4 epoxy

permittivity (4,4) (4,4)

DR (mm) 28 9 10 28 9 10 30 19 4 30 19 4

Inner

dimensions:

16.5 10 4

Moat gap

(mm):

Gi = 0.1,

G2 = 5.3,

G3 = G4 = 0.2

Microstrip (mm) 35 1.15 38 1.15 37 1.15 40 1.15

Slot (mm) 10 2 10 2 71 13.5 2

Ds (mm) 8 7 12 13

Ls (mm) 6 6 4 5

Comparison between all the structures

Parameter/structure Aperture-coupled DR with two DR after DR with Moat

DR Asymmetrical Increasing in

Wells a/b and

decreasing in

d/b ratio

Modes and TE111(2.92), TE111(2.92), Modes are Modes are

frequency (GHz) TE112(3.70), TE112(3.70), merged with merged with

TE113(4.64) TE113(4.64) each other, each other, gain

gain is is enhanced, and

enhanced, and the bandwidth is

the bandwidth increased

is increased drastically

Gain (dB) 1.7 2.1 3 4.8

Bandwidth (GHz) Less Less Bandwidth is Bandwidth is

enhanced by enhanced and is

merging the larger, and

modes modes have the

closer resonant

frequency to

each other

174 7 Rectangular DRA Higher-Order Modes and Experimentations

Height Excitation at top of Generated Frequency Gain Bandwidth

h (mm) the DR Mode f (GHz) (dB) (GHz)

5 No TExd11 11.3 5.6 10.112.0

Yes TExd11 11.8 4.8 9.517.7

TExd12 15 4.5

10 No TExd11 10 5 9.2512.1

5 12.8114.85

TExd15 13.7 7 16.517.1

Yes TExd14 13 6 9.416.8

15 No TExd13 10 9.1 9.4811.4

TExd17 16 5.6 15.816.4

6 18.418.6

Yes TExd14 11.7 5 9.317.6

TExd16 13.7 4.3

TExd18 16.7 4.5

30 No TExd;1;5 9.8 8.2 9.3011.85

TExd;1;11 12.56 6 12.4512.70

TExd;1;15 15.95 5.8 15.7016.20

TExd;1;17 17 5.9 16.9017.25

6 18.2018.50

Yes TExd;1;6 10.83 5 9.3517.1

TExd;1;12 13.58 5.1

TExd;1;14 14.88 4.7

TExd;1;16 15.7 5.1

7.13 Effect of a/b and d/b Aspect Ratio 175

of addition of wells

-5.00

m2

-10.00

m3

-15.00

S11 [dB]

-20.00 m1

-25.00

wells

-40.00

2.50 3.00 3.50 4.00 4.50 5.00 5.50

Freq [GHz]

176 7 Rectangular DRA Higher-Order Modes and Experimentations

Fig. 7.39 Merging of second band in rst and eld distributions. a Hx eld at f = 2.9 GHz, b E

eld at f = 2.9 GHz, c Hx eld at 3.6 GHz, d E eld at 3.6 GHz

of wells m1

5.00

0.00

dB(GainTotal)

-5.00

-10.00

-15.00

Name X Y

-20.00

m1 2.9314 5.4961

m2 5.0686 7.8804

-25.00

-30.00

2.50 3.00 3.50 4.00 4.50 5.00 5.50

Freq [GHz]

7.13 Effect of a/b and d/b Aspect Ratio 177

Fig. 7.41 Structure of the antenna after insertion of moat inside RDRA and eld distribution at

frequency 4.56 GHz

178 7 Rectangular DRA Higher-Order Modes and Experimentations

-2.50 2.50

-5.00

dB(S(1,1))_1

0.50

dB(GainTotal)

-7.50

-10.00 -1.50

-12.50

-3.50

-15.00

-17.50

-5.50 Name X Y

-20.00 m1 4.7559 2.9000

-22.50 -7.50

4.00 4.50 5.00 5.50 6.00 6.50 7.00 4.00 4.25 4.50 4.75 5.00 5.25 5.50 5.75 6.00

Freq [GHz] Freq [GHz]

Fig. 7.42 a Frequency response showing return loss. b Gain of the antenna over frequency

-2.00

0.00

-4.00

-5.00

-6.00

dB(S(1,1))

S11 [dB]

-8.00 -10.00

-10.00

-15.00

-12.00

-20.00

-14.00

- - - - with moat

-16.00 -25.00

4.00 4.50 5.00 5.50 6.00 4.00 4.25 4.50 4.75 5.00 5.25 5.50 5.75 6.00

Freq [GHz] Freq [GHz]

Fig. 7.43 Return loss and comparison with the structure with moat

increasing a/b ratio and

decreasing d/b ratio m2

5.00 m3

dB(GainTotal)

0.00

- 5.00

Name X Y

m1 5.4983 8.7898

-10.00 m2 4.5686 5.2020

m3 4.9699 4.0093

-15.00

4.00 4.50 5.00 5.50 6.00

Freq [GHz]

7.13 Effect of a/b and d/b Aspect Ratio 179

applications. The aspect ratio plays very important role in this phenomenon.

Excitation applied at top of the RDRA converted odd modes into nearby even

modes.

Reference

dielectric resonator antennas theory and experiment. In: IEEE international symposium on

Antennas Propagation Society, Seattle, WA, pp 764767, June 1994

Chapter 8

RDRA Angular Excitation Mathematical

Model and Resonant Modes

RDRA. The shift in radiation pattern and resonant modes have been realized based

of angular shift in input. Slot is the source of input voltage to RDRA. Slot size and

orientation effects loading of RDRA. The resonant characteristics of a RDRA are

dependent on shape, DRA volume and excitation. The excitation current can be

dened in terms of magnetic vector potential A based on applied current densities

J. This A can be expressed in terms of E and H elds or as S Poynting vector.

Keywords Slot Angular variation Change in radiation pattern Resonating

modes Power flux HFSS VNA Hardware model Anechoic chamber

8.1 Introduction

Slot is the source of input to RDRA. Slot size and orientation is responsible for

loading of RDRA. The angular orientation of slot has been investigated in this

chapter with simulations and experimentation. The resonant characteristics of a

RDRA depend upon the shape and size of the (volume) dielectric material along

with feeding style. It is to be appreciated that in a RDRA, it is the dielectric material

that resonates when excited by the feed. This phenomenon takes place due to

displacement currents generated in the dielectric material. The excitation current

can be dened in terms of magnetic vector potential A based on the current

densities J inside the resonator, at any far-eld point. This A can be expressed

in terms of E and H elds. Later, this is expressed as S Poynting vector. Now the

flux described can be treated with boundary conditions to nd Radiated power Prad

into space. Figure 8.1 presented RDRA excited at slot angle. Figures 8.1 and 8.2 are

HFSS model of RDRA. In Fig. 8.3, slot is shifted with certain amount of angle. If

two slots are placed at 90, circular polarization will take place. If one slot area is

larger than the other, then LHCP (left-hand circular polarization) or RHCP

(right-hand circular polarization) will take place. Figure 8.4 RDRA is excited at 45

angle. Figures 8.5, 8.6 and 8.7 presented radiation pattern at slot angles. Using two

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_8

182 8 RDRA Angular Excitation Mathematical Model

cross slots circular polarization can be integrated. If two slot of different lengths are

used then due to differential signal LHCP and RHCP can be generated. This

indicates that a mechnism for polarization cantrol can become possible if these

slots are arranged in a particular manner.

8.1 Introduction 183

0 Curve Info 0 Curve Info

rETotal rETotal

-30 30 Setup1 : Sweep -30 30 Setup1 : Sweep

12.80 Freq='8.451903808GHz' Phi='0deg' 9.60 Freq='8.451903808GHz' Phi='0deg'

rETotal rETotal

Setup1 : Sweep Setup1 : Sweep

9.60 7.20

Freq='8.451903808GHz' Phi='90deg' Freq='8.451903808GHz' Phi='90deg'

-60 60 -60 60

6.40 4.80

3.20 2.40

-90 90 -90 90

-180 -180

0 Curve Info 0 Curve Info

rETotal rETotal

-30 30 Setup1 : Sweep -30 30 Setup1 : Sweep

12.00 Freq='12.81062124GHz' Phi='0deg' 12.00 Freq='12.81062124GHz' Phi='0deg'

rETotal rETotal

Setup1 : Sweep Setup1 : Sweep

9.00 9.00

Freq='12.81062124GHz' Phi='90deg' Freq='12.81062124GHz' Phi='90deg'

-60 60 -60 60

6.00 6.00

3.00 3.00

-90 90 -90 90

-180 -180

HFSSDesign1 HFSSDesign1

Radiation Pattern 8 Curve Info Radiation Pattern 8 Curve Info

0 rETotal 0 rETotal

Setup1 : Sweep Setup1 : Sweep

-30 30 Freq='16.64328657GHz' Phi='0deg' -30 30 Freq='16.64328657GHz' Phi='0deg'

14.00 rETotal 14.00 rETotal

Setup1 : Sweep Setup1 : Sweep

Freq='16.64328657GHz' Phi='90deg' Freq='16.64328657GHz' Phi='90deg'

10.50 10.50

-60 60 -60 60

7.00 7.00

3.50 3.50

-90 90 -90 90

-180 -180

RDRA radiates from the fringing elds. The resonator acts as tuned sequential RLC

circuits having different values of LC or resonant cavity with an electric eld

perpendicular to the resonator, that is, along the Z-direction. The magnetic eld has

184 8 RDRA Angular Excitation Mathematical Model

0 Curve Info 0 Curve Info

rETotal rETotal

-30 30 Setup1 : Sweep -30 30 Setup1 : Sweep

11.20 Freq='8.451903808GHz' Phi='0deg' 11.20 Freq='8.451903808GHz' Phi='0deg'

rETotal rETotal

Setup1 : Sweep Setup1 : Sweep

8.40 8.40

Freq='8.451903808GHz' Phi='90deg' Freq='8.451903808GHz' Phi='90deg'

-60 60 -60 60

5.60 5.60

2.80 2.80

-90 90 -90 90

-180 -180

0 Curve Info 0 Curve Info

rETotal rETotal

-30 30 Setup1 : Sweep -30 30 Setup1 : Sweep

12.00 Freq='12.81062124GHz' Phi='0deg' 12.00 Freq='12.81062124GHz' Phi='0deg'

rETotal rETotal

Setup1 : Sweep Setup1 : Sweep

9.00 9.00

Freq='12.81062124GHz' Phi='90deg' Freq='12.81062124GHz' Phi='90deg'

-60 60 -60 60

6.00 6.00

3.00 3.00

-90 90 -90 90

-180 -180

0 Curve Info 0 Curve Info

rETotal rETotal

-30 30 Setup1 : Sweep -30 30 Setup1 : Sweep

12.00 Freq='16.64328657GHz' Phi='0deg' 14.00 Freq='16.64328657GHz' Phi='0deg'

rETotal rETotal

Setup1 : Sweep Setup1 : Sweep

9.00 10.50

Freq='16.64328657GHz' Phi='90deg' Freq='16.64328657GHz' Phi='90deg'

-60 60 -60 60

6.00 7.00

3.00 3.50

-90 90 -90 90

-180 -180

vanishing tangential components at the four side walls. The four extended edge

surfaces around RDRA serve as the effective radiating apertures. These fringing

elds extend over a small distance around the side walls and can be replicated as

elds Ex that are tangential to the substrate surface. The only tangential aperture

eld on these walls is E = Ez, because the tangential magnetic elds vanish by the

8.1 Introduction 185

boundary conditions. The ground plane can be eliminated using the image theory,

resulting in doubling the aperture magnetic currents, that is, J = n E. Hence, the

effective tangential elds can be expressed in terms of the eld Ez. Now, radiated

power pattern can be compared with the modes generated inside the resonator. The

surface current density can be the main source of EH elds pattern when applied

with boundary conditions inside the resonator. This can be correlated with far-eld

pattern. The physics of this radiation is based on the fringing effect due to dipole

moments. First derivative is velocity elds, and then, the second derivative on

dipole moments can be termed as acceleration, which is main source of radiations.

Hence, steering of the resonant modes mainly depends on excitation. Ez, Hz, or both

Ez and Hz elds at any instant of time can dene TM, TE and HEM modes.

Let aperture-coupled microstrip with slot and stub (feed) is situated in xy plane of

RDRA at bottom part and slot placed at an angle hi ; /i as shown in Fig. 8.1. The

resonator modes and radiation pattern generated have been investigated as follows:

1. Hz ; Ez elds are longitudinal. These have been expressed in terms of

orthonormality with signals umnp x; y; z and vmnp x; y; z at frequency xmnp

based on the Maxwells equations with given boundary conditions of RDRA.

2. At

z = 0; surface (x, y)excitation is applied with slot and surface current density

Jsx x; y; t; Jsy x; y; t is developed into RDRA.

3. The surface electric current density is equated with generated magnetic elds

into RDRA:

Js x; y; d Jsx ; Jsy ^n H Hy ; Hx ;

Cmnp and Dmnp .

4. Equating tangential component of Ez at boundary, i.e., Ey jz0 to zero, the

amplitude coefcients Dmnp for Hz and Cmnp of Ez are expressed.

5. Feed position in xy plane can be dened as follows:

x0 ; y 0 / 0 ; h0

8 9

<1 x0 x dllength =

6. f x; y j yj W width

: ;

0 otherwise

7. Excitation current in time domain can be expressed as:

Js x; y; z; t

186 8 RDRA Angular Excitation Mathematical Model

where is the

angle of variation in excitation.

Here, we apply excitation through slot dl at some specic angle. Later, shift in the

position of slot is provided. Change in radiation pattern or resonant modes is inves-

tigated with mathematical equation, simulations and experimentations on RDRA.

!

lI dl ejkr ~

A; where A is magnetic vector potential 8:1

4pr

Z

A V E dl

E jx ~

A

Radiated power

2 r

jE j2 x2 ~

A l

; g impedance: 8:2

2g 2g

Z

J r 0;x ejkjrr j 3 0

0

~ l

A d r ; at source: 8:3

4p jr r 0 j

Volume

Eh jxAh and Ah ^h A

X

J r 0 ; x J s mnp; r 0 ejxmnpt 8:4

mnp

jrr0 j

Z

l X J s mnp; r 0 ejxmnp t c

A dsr 0 ; ds is surface

4p mnp jr r 0 j

Z 8:5

l ejkn X 0

0

J s mnp; r 0 ejxmnp^rr dsr 0

4p jr r j mnp

s

Eh ; E/ ; H/ Eh =g; Hh E/ g:

8.2 Angular Shift in Excitation 187

1 2

2g

(

^h ^x cos u cos h ^y sin u cos h ^z sin h

/^ X ^ sin u Y^ cos u

l XZ

Eh Re fJsx mnp; r 0 cos u cos h

4pr 2 mnp

s

o

Jsymnp; r0 sin / cos h Jszmnp; r0 sin h

mnp

exp jx X 0 cos / sin h Y 0 sin / sin h Z 0 cos hdsr 0 ejxmnpt

c

8:7

XZ n o

Eu Re Jsx mnp; r 0 sin / Jsymnp; r0 cos /

mnp

xmnp

ej c X 0 cos u sin h Y 0 sin / sin h z0 cos hdsr 0 ejxmnpt dsr 0 8:8

Z

xmnp^r r 0

Px ^r j mnp Jsx mnp; r 0 ej c dsr 0 8:9a

s

Z

xmnp^r :r 0

Py ^r j mnp Jsy mnp; r 0 ej c dsr 0 8:9b

s

Z

xmnp^r:r 0

Pz ^r j mnp Jsz mnp; r 0 ej c dsr 0 8:9c

s

^r h; / X

Let

s mnp for convenience; then

Xn o

Eh Re Px ^rjs cos / cos h Py ^r js sin / cos h Pz ^rjs sin hgejxst

X

s

8:10

Re Esh ejxst

s

188 8 RDRA Angular Excitation Mathematical Model

3 2

000

where s mnp 4 001 5 and so on till s 111: Similarly

010

X

E/ Re Px ^r js sin u Py ^r js cos /ejxst 8:11

s

Hence, Radiation Pattern of RDRA: Power flux per unit solid angle will describe

the pattern. Power radiation pattern can be dened as follows:

2 ( !)

jEh j2 E/ 1 X

jxst jxst

Esh e Es/ e

2g 2 s

( !)

1 X X

jxst jxst

Hsh e Hs/ e

2 s s

!

1 X X

Es Hm ejxs xm t Es Hm ejxm xs t

4 s s

1X

8:12

Esh Hs Es Hsh

4 s

1 X

Re Esh Hs

2 s

1 ^

E E

Esh h E ^ sh

^ s ^h

u

2 g g

X jEsh j2 jEs j 2

^r ^r

s

2g 2g

Poynting vector

1 Xh

S ^r xs2 jPx ^r js cos / cos h Py ^r js sin u cos h Pz ^r js sin hj2

2g

2 i

xs2 jPx ^rjs sin u Py ^r js cos u 8:13

1 X

S ^r r; h; / xmnp2 fjPx h; /jmnp) cos / cos h

2g mnp

8:14

Py h; /jmnp sin / cos h Pz h; /jmnp sin h2

jPx h; /jmnp sin / Py h; /jmnp cos /2

8.3 Radiation Pattern Based on Angle 0 ; /0 Variation in xy Plane 189

in xy Plane

^

~ mnp pp Dmnp ejwmnp

Let D

d h2mn

~ mnp ej/mnpp=2 C

C ^ mnp

h2mn

" #

X X n o

H? ~ mnp ejxmnpt gr? ~umnp x; y; z

Re fD Re C~ mnp ejxmnpt r? ~

umnp x; y; z

mnp mnp

8:15

Probe orientation

^

^n H Js

jpp

c ; for all wave guide and

d

d d

let c ; jx ; for all cavity resonator

@z dt

Ez E?

Hz H?

Hx Jsy

Hy Jsx

H? Hx^x Hy^y

X

Ez Re C mnp umnp x; y; z expjxmnp t 8:16

mnp

c jx

H? r? Hz r? Ez ^z 8:17

h2mnp h2mnp

190 8 RDRA Angular Excitation Mathematical Model

jppz

Hz umn x; y exp 8:18

d

Hz umn x; yCmnpejxmnpt d

jppz

n o

Hz Re Cmnp umn x; ye jxmnpt d

jppz

n jppz

o

Hz Re C1 mnpumn x; yejxmnpt d

n jppz

o 8:19

Re C2 mnpumn x; yejxmnpt d

" #

1 d @

H? 2 r? Hz x; y; z; t 2 r? Ez x; y; z; t 8:20

hmn dz hmn @t

Hence

X

Ez x; yz; t Cmnp umnp x; y; z cosxmnpt umnp 8:21a

mnp

X

Hz x; yz; t Dmnp umnp x; y; z cosxmnpt wmnp 8:21b

mnp

(Ls, Ws) at an Angle h0 ; /0

We replace excitation probe to slot umnp and vmnp by ~umnp and ~vmnp

p mpx

npy

qpz

2 2

umnp x; y; z p sin

~ sin sin 8:22a

abd a b d

p mpx

npy

qpz

2 2

~vmnp x; y; z p cos cos cos 8:22b

abd a b d

0

Imnp eg xmnpt

Js x; y; d ; where g0 is an angle

x

X n o

H ? jzd Re D~ mnp ejxmnpt r?~vmnp x; y; z

mnp

X n o 8:23

Re C~ mnp ejxmnpt r? ~umnp x; y; z

mnp

8.4 Replacing Probe with Slot of Finite Dimensions (Ls, Ws) at an Angle h0 ; /0 191

^n H

Ex ; Ey ; Hx ; Hy are the elds in terms of surface current density due to applied probe

current at an angle ; Jsx ; Jsy can be expressed as current density using Fourier

coefcients; Cmnp and Dmnp Hy Hx elds can be computed from Ex Ey fields; prop-

agation terms h2mn c2 k 2

Imnp jgmnp

H? e f^x sin/0 cos/0 ^yggx; y

xwidth

p n mpx

npy

o

X

~ jxmnpt 2 2

H? RefDmnp e p r? cos cos : 8:24

mnp abd a b

Jsy x; y; t Jsx x; y; t

X n o mp

2p2 mpx

npy

Jsy x; y; t ~ mnp e

Re D jxmnpt

p sin cos

a abd a b

Za Zb mpx

npy

2

Jsy x; y; t p sin cos dxdy

ab a b

0 0

n o mp

r

2 mpx

~

Re Dmnp e jxmnpt

sin

a d a

Hence,

n o

Jsy x; y; t Re fy x; yejxmnpt

Za Zb r

mpx

npy

2 2 mp ~

fy x; y p ; Dmnp sin cos dxdy 8:25

ab d a a b

0 0

Now

r Za Zb mpx

npy

~ mnp 2 a 2

D fy x; y p sin cos dxdy 8:26

d mp ab a b

0 0

Hy Jsx

192 8 RDRA Angular Excitation Mathematical Model

Jsx x; y; t H y jz0

X n o

2p2 mpx

npy

mnp

b abd a b

8:27

n o

Jsx x; y; t Re fx x; yejxmnpt

r Za Zb mpx

npy

~ mnp 2 2 b

D fx x; y cos cos dxdy

ab d np a b

0 0

X

Jsx x; y; t Reffx fx; yjmnpgejxmnpt 8:28

mnp

X

Jsy x; y; t Reffy fx; yjmnpgejxmnpt 8:29

mnp

hi ; /i Slot Positions

Results of angular excitation on radiation pattern have been evaluated on HFSS

and shown in Fig. 8.5.

Angular shifts in excitation at 45 and 30 right side and radiation pattern are

shown in Fig. 8.7 and summarized are placed in Table 8.1.

Results of Radiation pattern when angular excitation is given to the right side

have been shown in Fig. 8.7.

Frequency S11 in S11 in dB S11 in dB S11 in dB Gain Gain Gain Gain

in GHz dB slot slot position cross-slot in dB in dB in dB (cross-slot)

position 45o right at left right in dB

at 45o left 45o 45o

8.45 11.2 11.6 10.8 11.5 1.5 3.75 2.7 1.5

12.81 13.3 12.4 13.4 12.2 1.1 1.0 1.1 1.2

16.64 14.7 18.1 18.0 15.9 1.5 1.8 1.6 3.0

8.5 HFSS Computed Radiation Pattern with Shifted hi ; /i 193

Table 8.1 described results of antenna parameters in tabular form. It has been

observed that angular variation in excitation has direct impact on the radiation

pattern as well as number of modes generated. This has been veried by the plots

given above. We have taken measurements of radiation pattern on varying slot of

feed at 30 and 45 to left and right from its original position. These plots have been

veried at two different frequencies. This completes the solution.

8.6 Experimentations

Figures 8.8, 8.9, 8.10, 8.11, 8.12, 8.13, 8.14, 8.15 and 8.16 present the experimental

results of RDRA. Their signicance is placed below each gure. The RDRA made

from acrylic glass sheets having dimensions of 9, 6 and 3 cm. The silicon oil having

e 2:2 was used as RDRA dielectric material. The resonant frequency of RDRA

was measured to 4.55 GHz. The measurements were taken at various angular

positions of the slot. Aperture-coupled feed RDRA is shown in Fig. 8.1. The feed

position was shifted to investigate RDRA S11 using VNA 40 GHz. The results are

shown in Figs. 8.2, 8.3, 8.4 and 8.5.

194 8 RDRA Angular Excitation Mathematical Model

Fig. 8.10 S11 RDRA with shifted slot resonant frequency 3.67

8.6 Experimentations 195

196 8 RDRA Angular Excitation Mathematical Model

8.6 Experimentations 197

198 8 RDRA Angular Excitation Mathematical Model

The results obtained with VNA have clearly shown shift in resonant frequency

due to feed orientation. It indicated that resonant modes are changing based on the

slot orientation. Hence, it is clearly evident that radiation pattern can be steered with

slot position in RDRA. If these results can be placed in look up table, then

microcontroller-based orientation can result into automated antenna. This auto-

mated antenna can be very useful for military applications. These cross slot can be

arranged in such a manner that circular polarization becomes possible in RDRA.

Also by varying lengths of cross slots, left hand or right hand polarization can be

achieved. The circular ploarization makes the signal robust and help to reduce

electromagnetic polution.

Chapter 9

Sensitivity Analysis of Rectangular DRA

and a, b, and d dimensions. These dimensions decides resonant frequency of

RDRA. The resonant modes are formed when realized with excitation. The resonant

frequency solution is worked with MATLAB and HFSS software. When these

dimensions are changed, resonant frequency of RDRA also changes. Variance

method has been tried out to evaluate error.

Ground plane RDRA Resonant frequency

Sensitivity analysis Variance Error minimization

been analyzed for frequency and resonant modes. RDRA is shown in Fig. 9.1.

These have been solved based on MATLAB and HFSS. Figure 9.2 presented

rectangular DRA with a, b, and d dimensions. Table 9.1 has shown RDRA

dimensions and their corresponding resonant frequencies. Figure 9.3 indicated

resonant modes with RDRA height. Plot of frequency versus length a variation is

shown in Figs. 9.4, 9.5 and 9.6. HFSS simulated modes in RDRA with S11

parameters are shown in Figs. 9.8 and 9.9 (Fig. 9.7).

da; db; dd

are (small change in length) random variables, and computed func-

tions are f dmnp and xmnp . The variance functions are ra ; rb ; rd . These are mainly

dependent on a, b, and d. Taylors expansion is restricted to second-order variable.

Hence, da; db; dd are mapped in terms of ra ; rb ; rd using diagonal matrix. Cmnp ;

Dmnp ; are amplitude coefcients which depend on the RDRA a, b, or d.

Frequency relationship can be determined based on a, b and d length variation

as given below:

d d dxmnpja; b; d

; ;

dd db da

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_9

200 9 Sensitivity Analysis of Rectangular DRA

b

a

x (mm) y (mm) z (mm) r Material used

RDRA a=7 b=7 d = 10 10 Sapphire

a=6 b=6 d = 15

a=5 b=5 d = 30

Substrate 20 30 0.5 3.38 Arlon25N(tm)

Ground plane 20 30

Microstrip feed line 15 1.11

DRA dimensions (mm) Resonant frequencies (GHz) simulated

a b d

7 7 10 f1 = 13.46

6 6 15 f2 = 13.85

5 5 30 f3 = 14.21

9 Sensitivity Analysis of Rectangular DRA 201

202 9 Sensitivity Analysis of Rectangular DRA

Fig. 9.7 RDRA having three different heights for increasing resonant modes

X

Ez x; y; z; t Re ejxmnpt Cmnp umnp x; y; z 9:1a

mnp

X

Hz x; y; z; t Re ejxmnpt Cmnp vmnp x; y; z 9:1b

mnp

At z = 0, Ez eld

r

X 2

jxmnpt

Ez t; x; y; 0 Re e Cmnp umn xy 9:2

mnp

d

and

Hx ; Hy Jsy ; Jsx

9 Sensitivity Analysis of Rectangular DRA 203

0.00 9.5741 -23.9108

m1 Curve Info

dB(S(1,1))

Setup1 : Sweep

-5.00

-10.00

dB(S(1,1))

-15.00

-20.00

m1

-25.00

7.50 8.75 10.00 11.25 12.50 13.75 15.00

Freq [GHz]

Name X Y

0.00 9.4238 -21.4417

XY Plot 1 HFSSDesign1

m1 Curve Info

m2 10.0251 -20.5547 dB(S(1,1))

m-2.50

3 14.5190 -17.9202 Setup1 : Sweep

-5.00

-7.50

-10.00

dB(S(1,1))

-12.50

-15.00

-17.50 m3

-20.00 m2

m1

-22.50

7.50 8.75 10.00 11.25 12.50 13.75 15.00

Freq [GHz]

Name X Y

m0.00

1 9.5741 -23.0345

XY Plot 1 HFSSDesign1

Curve Info

m2 14.0080 -12.2318

dB(S(1,1))

m3 16.0170 -16.3397 Setup1 : Sweep

m4 17.5701 -15.7949

m-5.00

5 17.5952 -15.7757

-10.00

m2

dB(S(1,1))

-15.00 m3 m4

-20.00

m1

-25.00

7.50 10.00 12.50 15.00 17.50 20.00

Freq [GHz]

204 9 Sensitivity Analysis of Rectangular DRA

We have computed

p Z mpx npy

2 2

Cmnp p Jsx X; Y sin sin dxdy 9:3

abd a b

Now if

a ! a da a is increased to a da

b ! b db b is increased to b db

d ! d dd d is increased to d dd

We need to compute

Cmnpja da; b db; d dd

and similarly

xmnpja da; b db; d dd

N normal distribution with mean zero

variance function

To compute error

2 3 0 2 2 31

da ra 0 0

4 db 5 N @0; 4 0 r2b 0 5A

dd 0 0 r2d

dCmnp dCmnp dCmnp

Cmnpja da; b db; d dd Cmnpja;b;d da db dd

da db dd

2

1 d Cmnp 2 d2 Cmnp 2 d2 Cmnp 2

da db dd

2 da2 db2 dd 2

d2 Cmnp d2 Cmnp d2 Cmnp

2 dadb 2 dadd 2 dbdb

dadb dadd dbdd

9:4

dCmnp 2 D E

hCmnpja da; b db; d dd C2mnpja; b; d i jdaj2

da

E 2 D

dC mnp 2 D E

jdbj2 dCmnp jdd j2

db dd

2 2

dCmnp 2

r2a r2 dCmnp r2 dC mnp

da b

db d

dd

9:5

9 Sensitivity Analysis of Rectangular DRA 205

hxmnpjada;bdb;ddd xmnpja;b;d i

Error value

D E

xmnp dxmnp jdxj2

Error or variance:

dxmnp 2 2 dxmnp 2 2 dxmnp 2 2

r r r

da a db b dd d

r

m2 n2 p2

xp

a2 b2 d 2

z direction is given below:

m p 2

@x

q

a3

9:6a

@a p2

a2 b2 d 2

m 2 n2

n p 2

@x

q

b3

9:6b

@b p2

a2 b2 d 2

m2 n2

p2 p

@x

q

d3

9:6c

@d p2

a2 b2 d 2

m 2 n2

This gives the complete solution of RDRA sensitivity analysis. The higher-order

modes of a rectangular DRA were used to produce radiation patterns with enhanced

gain. The advantage of this approach is for enhancing gain. The maximum

achievable gain on mode m = 1, n = 7 to increase Directivity to 13.7 dBi.

Such DRA designed at 11 GHz with height 35 mm, this investigation focused on

rectangular DRAs, for excitation of the appropriate higher-order modes in RDRAs.

206 9 Sensitivity Analysis of Rectangular DRA

clear all;

clc;

close all;

c=3*10^8;

m=1;

n=1;

p=1;

E=10;

a=5*10^-3:.1*10^-3:30*10^-3;

b=10*10^-3;

d=15*10^-3;

for i=1:length(a)

f(i)=c/(2*pi)*sqrt(E)*sqrt((m*pi/a(i))^2+(n*pi/b)^2+(pp*pi/(2*d)^2));

end

a1=15*10^-3;

b1=10*10^-3:.1*10^-3:40*10^-3;

d1=20*10^-3;

9.1 MATLAB Simulation 207

for k=1:length(b1)

f1(k)=c/(2*pi)*sqrt(E)*sqrt((m*pi/a1)^2+(n*pi/b1(k))^2+(pp*pi/(2*d1)^2));

end

a2=10*10^-3;

b2=5*10^-3;

d2=10*10^-3:.1*10^-3:50*10^-3;

for t=1:length(d2)

f2(t)=c/(2*pi)*sqrt(E)*sqrt((m*pi/a2)^2+(n*pi/b2)^2+(pp*pi/(2*d2(t))^2));

end

This gives the complete solution for RDRA sensitivity analysis.

It has been observed that resonant modes have been increasing based on increase

in dipole moment, i.e., modes are proportional to the height of RDRA.

208 9 Sensitivity Analysis of Rectangular DRA

See below Figs. 9.10, 9.11, 9.12, 9.13 and Table 9.2.

Fig. 9.11 Return loss versus frequency with dimensions a = 5 mm, b = 5 mm, d = 30 mm shows

Return loss 18 dBi at f = 10.95 GHz

9.2 HFSS Simulations 209

Fig. 9.12 Return loss versus frequency with dimensions a = 6 mm, b = 6 mm, d = 15 mm shows

return loss 24 dBi at f = 10.95 GHz

x (mm) y (mm) z (mm) r Material used

RDRA a=7 b=7 d = 10 10 TMM10i

a=6 b=6 d = 15

a=5 b=5 d = 30

Substrate 20 30 0.5 3.38 Arlon25N(tm)

Ground plane 20 30

Microstrip feed line 19.2 1.1672

Lumped element 1.1672 0.5

210 9 Sensitivity Analysis of Rectangular DRA

Chapter 10

Hybrid Modes in RDRA

Abstract In this chapter, new kind of resonant mode, i.e., hybrid mode in RDRA

(rectangular dielectric resonator antenna), is described using mathematical model-

ing. RDRA is excited by inserting RF feed probe or microstripline having nite

dimensions, carrying electric and magnetic currents at a given frequency. The

charge conservation equations then imply the presence of electric charge densities

and magnetic charge densities within the resonator. From Maxwells equations, we

derive vector Helmholtz equations for the electromagnetic elds. The vector

sources provide electric charge which gets converted into magnetic charge. In one

of the models, sidewalls of the resonator are perfect magnetic conductors, and top

and bottom surfaces are perfect electric conductors. Thus, the boundary conditions

on the elds are such that the tangential components of the magnetic eld vanish on

sidewalls and the normal components of the magnetic eld vanish at top and bottom

surfaces. The normal components of the electric eld vanish on sidewalls. Hz can

therefore be expanded as linear combinations of sin functions in xy direction along

with z-component of the source. For the Hz, Helmholtz equation can be expanded

in terms of sin functions (assuming that these sources vanish on boundary), with

z-dependent coefcients.

Mathematical model

Normal component

Tangential component Conservation equation Magnetic energy Electrical

energy Field diversity Fourier basis function

10.1 Introduction

probe or microstripline having nite dimensions, carrying electric and magnetic

currents at a given frequency. The charge conservation equations then imply

the presence of electric charge densities and magnetic charge densities within the

resonator at that particular frequency. To completely solve for the elds with the

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_10

212 10 Hybrid Modes in RDRA

resonator, we therefore set up four Maxwells equations taking into account mag-

netic and electric currents and charge densities. From these equations, we derive

vector Helmholtz equations for the electromagnetic elds with vector sources

determined from gradient and curl of the electric charge, magnetic charge, and

current densities. The sidewalls of the resonator are perfect magnetic conductors,

and top and bottom surfaces are perfect electric conductors. Thus, the boundary

conditions on the elds are such that the tangential components of the magnetic

eld vanish on sidewalls and the normal components of the magnetic eld vanish at

top and bottom surfaces. The normal components of the electric eld vanish on

sidewalls. Hz can therefore be expanded as linear combinations of sin functions in

xy direction along with z-component of the source. For the Hz, Helmholtz equation

can be expanded in terms of sin functions (assuming that these sources vanish on

boundary), with z-dependent coefcients.

On substitution of these expressions into the Helmholtz equation for Hz, source

then gives us a second-order linear differential equation for the coefcient functions

of z in Hz with a source term. This is solved, and the solution consists of a

superposition of a source (particular solution or inhomogeneous solution) term and

a homogeneous term (i.e., general solution of the homogeneous part). In hybrid

modes, total solution is developed, i.e., homogeneous and inhomogeneous. Two

constants in the homogeneous part are determined by applying the vanishing

boundary conditions on Hz at top and bottom surfaces, i.e., at z = 0, d. Likewise

applying boundary conditions on the vanishing of the normal component of the

E eld on the sidewalls, expressions are determined for Hx and Hy. Then, resonance

is seen, i.e., the electromagnetic eld inside the resonator is proportional to 1d, where

d is the frequency perturbations determined from Dirac delta functions. This

completely solves the problem of RDRA modes.

Hybrid modes can be generated by superposition of TE and TM modes inside

RDRA. In this case of RDRA, hybrid modes have been generated by using a probe

of nite dimension (d) is inserted into z direction and excitation of this probe

(d length) current is given to rectangular copper plane (x, y) as shown in Fig. 10.1

The current density can be determined based on KAM (KolmogorovArnold

Moser) time-averaging method and using dDirac delta function. The principle of

orthogonality is nally applied to determine Cmnp and Dmnp amplitude coefcients

of hybrid modes with elds in homogeneous and particular case. Here, particular

case will have inhomogeneous medium with source applied. Hz and Ez elds have

been computed simultaneously to generate hybrid modes and their coefcients.

Fig. 10.1 RDRA with copper x, y rectangular plane and feed d length

10.1 Introduction 213

Figure 10.1 indicates RDRA for generating hybrid modes. Figure 10.2 shows

wall conguration of rectangular DRA required for hybrid modes. The hybrid modes

offer high efciency and polarization diversity. Frequency bandwidth can be con-

served by using polarization diversity. Maxwells equation is applied to nd solution

of RDRA. The eigen functions are obtained by solving Helmholtz equation. The

transverse components of Ex , Ey , Hx , and Hy have been expressed in terms of

longitudinal components Ez and Hz . The RDRA has been excited with a RF feed

probe of nite radius and small length inserted through ground plane into RDRA

along z-axis. Surface current density

on walls of resonator is produced due to

excitation given at feed point a2 ; b2 ; z of rectangular resonator. The azimuthal

component of magnetic elds inside the resonator is introduced, which is also

equivalent to z-component of surface current density. The modal longitudinal

coefcients are Ez and Hz . The radiation pattern or power distribution among these

different eigen modes is controlled by current distribution inside the resonator. The

inner product or reaction term of eigen function will be equal to corresponding eigen

mode. It is because magnetic currents are equal to electric currents in an antenna, due

to orthonormality principle or conservation of energy methods. Some of these power

coefcients can be made zero by canceling a particular resonant mode or blocking a

particular eigen function. This is possible for TE and TM modes. The same fre-

quency is introduced inside the guide with phase opposite to each other. Extracting a

particular resonant mode is also possible if surface current density of that mode is

made large enough by input excitation. When we apply input excitation frequency

matching to the desired mode, weighted magnitude of that particular mode coef-

cient becomes large and corresponding mode gets excited in the RDRA. The mode

merging can also be made possible by introducing shift in more than two modes

toward a common desired point. Equal weightage of TE and TM at same fre-

quency with opposite phase can cancel the mode. Higher-order modes can also be

generated in RDRA as shown in Fig. 10.1. Higher modes can provide higher gain

and high directivity to prevent EM pollution in microwave devices.

Figure 10.3 shows circular polarization of EM waves. The transverse components

Ex , Ey , Hx , Hy are the components determined in terms of longitudinal components

Ez and Hz . These transverse elds satisfy Helmholtz equations, are expressible in

terms of umnp ejxmnpt and vmnp ejxmnpt i.e., fourier basis functions. Ex , Ey , Hx ,

Hy elds are also expressible in terms partial derivatives of umnp, vmnp and

214 10 Hybrid Modes in RDRA

polarization

vmnp for Ez , Hz , then the same coefcients appear in Ex ; Ey ; Hx ; Hy . These

amplitude coefcients C mnp, Dmnp can be determined by matching Hx , Hy at

z = 0, to the surface current density of RDRA when feed is at z = 0. If the surface

excitation at z = 0 has frequency component other than xmnp, say x, then the eld

amplitude components corresponding to this excitation are determined from given

below terms. The source frequency, which is other than xmnp, introduces decay in

the resonator due to nite conductivity of the medium. Using the KAM theory of

averaging, the resonator extracts out only xmnp frequencies with amplitude.

Equivalently, if source contents are switched on for a nite duration and then

switched off, the only the dominant xmnp frequencies will be present in the

resonator. This situation is analogous to connecting a voltage source to an LC

oscillator for a nite duration and then switching it off. If, however, the source is not

switched off, then the other (non-dominant frequencies) will also be present and

these can be computed based on above-mentioned methods. The composite structure

having combination of PMC and PEC walls can generate hybrid modes (HEM).

The HEM can be further classied HE as odd hybrid modes and EH as even hybrid

modes. The applications for higher modes generation, mode shifting, mode merging,

and mode control can be made in antenna design. They can impart wide design space

in the eld of antenna. These designs can be used in beam control and regulation.

Maxwells equations:

For magnetic elds:

1 0

0 1

Hx r J x x; y; z; x

r2 k @ Hy A @ r Jy x; y; z; x A;

2

10:1

Hz r Jz x; y; z; x

10.2 Mathematical Model 215

J current density

wn function

a n, b n amplitude coefcients

umnp, vmnp Fourier basis function

hmn cut off frequency

0 1 2P P 3 2 3 2 3

Ex an wn bn un wn un

B C 6P P X X

0 0 7 6 07 6 07

@ Ey A 4 an wn bn un 5 an 4 wn 5 bn 4 u n 5

P P

Ez an w00n bn u00n w00n u00n

0 1 10:2

Ex

B C 1 @ 2 umnp

@ Ey A C mnp ;

hmn2 @z@x

Ez

0 P Cmnp @ 2 umnp P lDmnp @ 2 vmnp

1

hmn2 @x@z

hmn2

jxm; n; p @y@z

B mnp C

B C

E B P 1 C @ 2 umnp P l jxm; n; pD @ 2 vmnp Cejxmnpt ; 10:3

@ hmn2 mnp @y@z hmn2 mnp @x@z A

mnp

Cmnp umnp

0 E 1

wmnpx x; y; z

X B E C jxmnpt

C mnpB C

@ wmnpy x; y; z A e

mnp

wEmnp x; y; z 10:4

0 Ez 1

umnpx x; y; z

X B E C jxmnpt

DmnpB C

@ umnpy x; y; z A e ;

mnp

umnpz x; y; z

E

with duality:

0 H 1

wmnpx x; y; z

X B H C jxmnpt

CmnpB C

@ wmnpy x; y; z Ae

mnp

wHmnpz x; y; z 10:5

0 H 1

/mnpx x; y; z

X B H C jxmnpt

DmnpB C

@ /mnpy x; y; z Ae ;

mnp

/Hmnpz x; y; z

216 10 Hybrid Modes in RDRA

^

n H Js surface current density on walls;

Hsx x; y; 0 Jsy x; y

Hsy x; y; 0 Jsx x; y:

Hence,

X

jxmnpt

CmnpwH mnpx x; y; 0e

mnp 10:6

X

jxmnpt

DmnpuH

mnpx x; y; 0e ;

mnp

X

jxmnpt

Jsx x; y; dt CmnpwH

mnpy x; y; 0e

mnp

X 10:7

jxmnpt

DmnpuH

mnpy x; y; 0e :

mnp

from Linear combinations of sine and cosine terms given below:

wH a

wH b

uH c

uH d

Amplitude

D coefcients can E be determined from principle of orthonormality:

C mnp wmnpy x; y; 0; wmnpy inner product or reaction terms can be written as

H H

follows:

D E D E

mnpx ; wmnpx Dmnp /mnpx ; wmnpx

C mnp /H H H H

1 T

Z

D E 10:8

lim Jsy x; y; t; wH

mnpx x; y; 0 ejxmnpt dt;

T!1 2T

T

10.2 Mathematical Model 217

and

D E D E

C mnp /H

mnpy ; wmnpy Dmnp /mnpy ; wmnpy

H H H

1 T

Z

10:9

jxmnpt

lim hJsx x; y; t; wH

mnpy x; y; 0i e dt:

T!1 2T

T

These are the solutions of amplitude coefcients Dmnp and Cmnp using

time-averaging KAM method (KolmogorovArnoldMoser).

These are the solutions of hybrid modes.

Helmholtz Equations

2

r2 x2 w ^ x; y; z; x 0; Helmholtz equation in frequency domain

c

^ x; y; z; x X xY yZ z; separation of variables will be used

w

Solutions for RDRA on application of boundary conditions are as follows:

r2 k2 Ez 0; TM mode

r2 k2 Hz 0; TE mode

Hz 0; on all walls

Ex ; Ey 0; z 0; d;

Ex 0; x 0; a;

Ey 0; x 0; b;

1 @ l @

E ? 2 r? Ez 2 r? Hz ^z ; Hybrid mode electric fields; 10:10

h @z h @t

1 @ l @

H ? 2 ; r? Hz 2 r? Ez ^z ; Hybrid mode electric fields: 10:11

h @z h @t

1 @ 2 Ez l @ 2 Hz

Ex ;

h2 @x@t h2 @y@t

1 @ 2 Ez l @ 2 Hz

Ey 2 2 ;

h @y@z h @x@t

218 10 Hybrid Modes in RDRA

@ 2 Ez

0; x 0; a;

@x@z

@Ez

0; x 0; a;

@x

@Ez

0; y 0; b:

@y

There are three types of resonant modes at any known frequency xmnp:

@ 2 Ez

0; z 0; d;

@x@z

@Ez

0; z 0; d:

@z

X

Ez C mnpejxmnpt umnpx; y; z; t;

X

Hz Dmnpejxmnpt vmnpx; y; z; t:

X X

Ehom x; y; z; t C mnpejxmnpt wEmnp r Dmnpejxmnpt wEmnp r ;

10:12

where r x; y; z

X X

H hom x; y; z; t C mnpejxmnpt wH

mnp

r Dmnpejxmnpt wH

mnp

r;

10:13

J ! Js x; ydz

Js x; y; x

10.3 Modes in Homogeneous Medium with Source Terms 219

Maxwells equation:

r H J jxE

r E jxlH

E or H can be calculated

r2 H r J x2 lH

r2 k2 H r J with source;

H Hhom Hpart

J s x; y; z J sy x; ydz d Jy;z

r J Jz;y Jy;z

r2 k2 Hx Jsy x; yd0 z d

r2 k2 Hy Jsx x; yd0 z d

2

r k2 Hz Jsyx Jsxy d0 z d

Hx ; Hy ; Hz x; y; z; x

vmnp sinx siny sinz

umnp cosx cosy cosz

Hz from E ? ; H ? equations:

X 10:14

Dsource m; n; p; xvmnp r Csource m; n; p; xumnp r ;

!

X xmnp2 x2

r k Hz

2 2

2 Czs m; n; p; xvmnpr : 10:15

mnp

c2 c

magnitude of coefcients for non-resonant frequency, and x and fxmnpg: x are

non-resonant terms.

X

Jsx x; y x2 xmnp2 C s mnp; xvmnpx; y; d ;

mnp

220 10 Hybrid Modes in RDRA

c2 Jsy;x x; y Jsx;y x; y

Z0 Z0

10:16

c2 Jsy;x x; y; x Jsx;y x; y; xvmnpx; y; d dx dy;

a b

Z

1

Czs mnpx Jsy;x x; y; x Jsx;y x; y; xvmnpx; y; d dx dy:

x2 xmnp2

10:17

Similarly, we can compute Cxs Cys terms which are the desired solutions of

hybrid resonant modes.

E, H, elds:

r E jxlH

r H jxE J

r2 E jxljx

rHJ

E

Jx

2 R

hence, Czs mnp x2 xc2 mnp Jsy;x x; y; x Jsy;y x; y; xvmnp x; y; d dx dy is the

solution of hybrid modes. Similarly, other hybrid mode coefcients can be worked

out.

Cys mnp dys mnp

Czs mnp dzs mnp

10.6 Mathematical Modeling of Hybrid Modes 221

n H Js on each wall

n E Ms on each wall

based on solution of Maxwells equation.

First, we develop solution of rectangular waveguide and switch to resonator. The

waveguide solution is very simple. Figure 10.4 indicates eld conguration inside

RDRA. These waveguide equations will have both the elds Hz and Ez as given below:

jxl

r? Ez cE ? r? Hz ^z jxE ? 10:23

c

222 10 Hybrid Modes in RDRA

2

jxl x l

r? E z r? Hz ^z c E?

c c

c2 x2 l h2

c jxl

E? r? Ez 2 r? Hz ^z 10:24

h2 h

c jxl

H? 2

r? Hz 2 r? Ez ^z 10:25

h h

h i

Resonator equations are obtained by simply replacing c by ddz :

1 d l d

E? r? Ez 2 r? Hz ^z 10:26

h2 dz h dt

1 d l d

H? 2

r? Hz 2 r? Ez ^z 10:27

h dz h dt

2 Ez

r h2 0; Helmholtz equation

Hz

Boundary conditions in RDRA have been taken as, top and bottom walls of

resonator are PEC other four sides walls are PMC.

Hz 0; at x 0; a; and y 0; d; z 0; d;

Ex Ey 0; at z 0; d;

Hx 0; at y 0; b;

Hy 0; at x 0; a:

X

Hz Re Dmnpejxmnpt vmnp r 10:28

X

Ez Re Cmnpejxmnpt umnp r 10:29

p mpx mpy mpz

2 2

vmnp p sin sin sin 10:30

abd a b d

10.6 Mathematical Modeling of Hybrid Modes 223

2 2

umnp p cos cos cos 10:31

abd a b d

Equations (10.30) and (10.31) have been obtained from expansion of Helmholtz

equation by separation of variables method:

r2 h2 Hz 0

m2 2

Hence, h2 h2mn p2 a2 nb2 ; this gives the resonant frequency of RDRA.

Tensor product of linear combination can appear as given below:

n mpx npx npx mpx

Hz L cos cos ; cos sin ;

a a

mpx a a o

npx npx mpx

sin cos ; sin sin

a a a a

where L denotes linear combinations. It turns out that depending on the nature of

wall or surface (PEC or PMC), four possible linear combinations can appear

cos sin; sin cos; and sin sin; cos cos. Also,

x2 l c2 h2mn

Hence,

Hz 0; when y 0; against cos terms are ruled out from y:

mpx npy

Hz sin sin C1 ecmnz C2 ecmnz

a b

Hz 0; when z 0; d

C1 C2 0

ecmnd ecmnd 0; b

C1 C2 ; sincmnd 0

cmn jbmn

bmnd pp

Hence,

pp

bmn

d

mpx npy ppz

Hz sin sin sin 10:32

a b d

224 10 Hybrid Modes in RDRA

pp2

x2 l d h2mn ; hence, resonant frequency can be determined as follows:

s

p m 2 n2 p2

x l p ;

a2 b2 d 2

Here, we note that resonant frequency in hybrid mode is same for TE and TM

modes.

Now

1 @ @Hz l @ @Ez

Hx 2 ;

h2 @z @x h @t @y

Hx 0; at y 0; b;

dHx

0; at y 0; b;

dx

dEz

0; at y 0; b;

dy

Ex 0;

Ey 0; z 0; d;

Hence,

1 @ @ l @ @Hz

Ex Ez 2 ;

h2 @z @x h @t @y

@ 2 Ez

0 at z 0; d

@x@z

@Ez

0 at z d

@t

z-dependence of Ez is cos ppz

d ; Ex 0; when x 0; a;

mpx npy ppz

Ez cos cos cos 10:33

a b d

separation of variables.

10.7 General Solution of Hybrid Modes (HEM) 225

regarding the electromagnetic elds to vary with z-axis, i.e., these are exploited into

the Maxwell curl equations, then manipulating them to express the transverse

components of the elds in terms of partial derivatives of the longitudinal com-

ponents of the elds w.r.t. x and y axis (i.e., the transverse coordinates). Waveguide

models of four different rectangular DRAs with specied boundary conditions lled

with homogeneous material having linear permittivity have been mathematically

developed and realized to determine TE and TM modes propagating elds. These

have resulted in different sinecosine combinations. Propagation of these elds

have been split as inside and outside of the RDRA with an interfacing surface

having two different permittivity. The solution is transcendental equation which

purely characterizes rectangular DRA resonant frequency and propagating elds.

The amplitude coefcient of these elds Cmnp and Dmnp inside the DRA can be

determined by comparing time-averaged magnetic energies equal to time-averaged

electrical energies by KAM method based on principle of orthonormality. The

transverse components Ex , Ey , Hx , Hy are the components determined in terms of

longitudinal components Ez , Hz . These transverse elds satisfy Helmholtz equa-

tions, are expressible in terms of umnp ejxmnpt and vmnp ejxmnpt Fourier basis

function. Ex , Ey , Hx , Hy elds are also expressible in terms partial derivatives of

umnp, vmnp, and hence if C mnp, Dmnp denotes the linear combinational

coefcients of umnp; vmnp for Ez , Hz , then the same coefcients appear in Ex ,

Ey , Hx , Hy . These coefcients Cmnp, Dmnp can be determined by matching Hx ,

Hy at z = 0 to the surface current density of RDRA, when feed is at z = 0. If the

surface excitation at z = 0 has frequency component other than xmnp, say x, then

the eld amplitude components corresponding to this excitation are determined.

Both the elds Ez and Hz will remain excited at any instant of time in resonator, and

then, these modes can be termed as hybrid modes. Our solution is developed based

on homogeneous medium in the resonator.

X n o

Ez Re Cmnpejxmnpt umnp r 10:34

X n o

Hz Re Dmnpejxmnpt vmnp r 10:35

X 1 n o@

jxmnpt

E? Re Cmnp e r? umnp r

h2mn @z

n o 10:36

l

2 Re jxmnpDmnpejxmnpt r? vmnp r ^z

hmn

226 10 Hybrid Modes in RDRA

E? ^x Ex ^y Ey ;

H? ^x Hx ^y Hy ;

@ @

r? ^x ^y ;

@x @y

0 1

Ex X X

E @ Ey A Cmnp ejxmnpt WEmnp r Dmnpejxmnpt /Emnp r 10:37

Ez

0 1

Ex Xh i

H @ Ey A C mnp ejxmnpt WH

mnp r Dmnpe

jxmnpt H

/mnp r 10:38

Ez

0 1

1 @2

2 @x@z umnp

B h1mn C

WEmnp r @ 2 @2 A

hmn @x@z umnp

umnp

where WEmnp is the electric eld vector coming from the z-components of electric

eld, i.e., TM mode.

And /Emnp is the electric eld vector coming from the z-components of magnetic

eld, i.e., TE mode.

Similarly, magnetic eld vectors /H

mnp

r and WH

mnp r can be solved.

0 1

1 @2

h2mn @x@z vmnp

B C

/H r B

@ 1 @2 C

A

mnp h2mn @x@z vmnp

vmnp

0 jlxmnp 1

@

h2mn @y vmnp

B C

/Emnp r B jlxmnp

@ 2 @ C

A

hmn @x vmnp

0

10.7 General Solution of Hybrid Modes (HEM) 227

and,

0 jlxmnp

1

@

@y umnp

2

B hmn C

mnp r

WH @ jlxmnp A

@

@x umnp

h2mn

0

Xh i

E CmnpWEmnp r Dmnp/Emnp r ejxmnpt 10:39

Xh i

H C mnp WH

mnp r D mnp /H

mnp

r ejxmnpt 10:40

Solution of the RDRA can be developed by using these above two equations.

For this, we insert a probe of d length having R radius into rectangular DRA. This is

pointing toward z-axis.

Figure 10.5 gives a clear picture of RDRA with feed associated, and Fig. 10.6

shows the structure of cylindrical probe.

a

x R cos /;

2

b

y R sin /;

2

z 0:

228 10 Hybrid Modes in RDRA

^

n H J s : This is based on boundary conditions inside the RDRA.

^ J

^ Hz^z H/ /

q s

^ J

H/^z Hz / s

and

a b

H/ R cos /; R sin /; z Jsz /; z 0\z\d; 0\/\2p

2 2

a b

Hz R cos /; R sin /; z Js //; z

2 2

X

a b

Js //; z; t Dmnpvmnp R cos /; R sin /; z ejxmnpt

2 2

a b

H/ R cos /; R sin /; z Hx sin / Hy cos /

2 2

Jsz /; z 0\z\d; 0\/\2p

Js/ /; z

X jlxmnp @vmnp a

b

Jsz /; z; t sin / cmnp R cos /; R sin /; z

h2mn @y 2 2

dmnp @ 2 vmnp a b

2 R cos /; R sin /; z ejxmnpt

hmn @z@x 2 2

X

jlxmnp @

a b

cos / Cmnp umnp R cos /; R sin /; z

h2mn @x 2 2

Dmnp @ vmnp a

2

b

2 R cos /; R sin /; z ejxmnpt ;

hmn @z@x 2 2

10:41

X

Js/ /; z; t Cmnp ejxmnpt Xmnp /; z;

mnp

where,

a b

Xmnp /; z vmnp R cos /; R sin /; z ; 10:42

2 2

10.7 General Solution of Hybrid Modes (HEM) 229

X

1 2

Jsz /; z; t Cmnp gmnp /; z Dmnp gmnp /; zejxmnpt

mnp

where,

1 jxmnpt l sin / @ a b

gmnp /; z u mnp R cos /; R sin /; z

h2mn @y 2 2

2 sin / @ 2 vmnp a b e mnp /; z

gmnp /; z R cos /; R sin /; z ejxmnpt X

h2mn @x@z 2 2

Z

1 ~ mnp /; zejxmnpt dt d/ dz

Cmnp lim J s/ /; z; tX 10:43

T!1 2T

jtj\T

0\/\2p

0\z\d

Z

Xe mnp /; z2 d/ dz

Z 2 Z

1 1 2

Cmnp gmnp /; z d/ dz Dmnp gmnp /; z gmnp /; z d/ dz

10:44

Z

1 1

lim gmnp /; zJsz /; tejxmnpt d/ dz dt

T!1 2T

jtj T

Z Z

1 2 2 2

Cmnp gmnp /; z gmnp /; z d/ dz dt Dmnp jgmnp /; zj d/ dz

Z

1 2

lim Jsz /; z; t gmnp /; zejxmnpt d/ dz dt

T!1 2T

jtj\T

10:45

If we keep Js/ 0, from Eqs. (10.23) and (10.24), we get C(mnp) and D(mnp).

The study of electric and magnetic elds for maxima and minima inside RDRA

introduces us to dene mode number. By applying perturbations, higher modes can

be excited. The increase in the electrical length of the antenna on higher-order mode

causes higher antenna gain. Short and open boundaries are the basis of modes. The

half-wavelength resonant modes with odd numbers only will be excited when

ground plane is used as even modes get short-circuited due to ground plane. The

polarization of even and odd modes is opposite. The higher modes will have higher

resonant frequency. A number of higher modes also modify the radiation patterns,

230 10 Hybrid Modes in RDRA

i.e., the modes number will be equal to number of lobes in nal radiation pattern.

Highly directive patterns can be obtained at higher modes. Bandwidth of

higher-order modes will be decreased. HEM 1, 3, 5, 7 are odd modes that can be

written as HE. Similarly HEM 2, 4, 6, 8, are even modes or EH mode. The care

must be taken to select this hybrid number n because it has a direct relationship with

the radiation pattern of far elds or beam shape. Gain of antenna can also decrease

abruptly due to dispersion at higher modes. This is introduced when the dipole

moment starts overlapping. Based on various solutions, hybrid modes can be

memorized for any particular mode with desired radiation patterns. Automated

applications using microcontroller can generate lookup table for desired radiation

patterns or beam pattern for any desired frequency as well as gain. Thus, hybrid

modes can be used for automated RDRA recongurability.

Figures 10.7, 10.8, 10.9, 10.10, 10.11, 10.12 and 10.13 are results of simulated and

experimentations of RDRA. Type of result is mentioned in each picture.

Fig. 10.7 The excitation is given by TE and TM modes at the same time

Fig. 10.8 The excitation is given by TE and TM modes at the same time

10.8 HFSS Results 231

HFSSDesign1

Name Theta Ang Mag Radiation Pattern 2

m1 360.0000 -0.0000 1.8149 Curve Info

0

m2 60.0000 60.0000 0.6570 m1 GainTotal

-30 30 Setup1 : Sweep

1.60 Freq='10.05511022GHz' Phi='0deg'

GainTotal

1.20 Setup1 : Sweep

-60 60 Freq='10.05511022GHz' Phi='90deg'

0.80

0.40 m2

-90 90

-120 120

-150 150

-180

Radiation Pattern 8 HFSSDesign1

Curve Info

m1 360.0000 -0.0000 12.1194 0

dB(rELHCP)

-30 30 Setup1 : Sweep

20.00 Freq='15.34673367GHz' Phi='0deg'

m1 dB(rELHCP)

10.00 Setup1 : Sweep

-60 60 Freq='15.34673367GHz' Phi='90deg'

0.00

-10.00

-90 90

-120 120

-150 150

-180

232 10 Hybrid Modes in RDRA

HFSSDesign1

Name X Y XY Plot 4 Curve Info

80.00

m1 70.0000 4.3036 dB(AxialRatioValue)

Setup1 : Sweep

Freq='13.34170854GHz' Phi='0deg'

70.00

dB(AxialRatioValue)

Setup1 : Sweep

dB (AxialRatioValue)

50.00

40.00

30.00

20.00

10.00 m1

0.00

0.00 125.00 250.00 375.00

Theta [deg]

Chapter 11

Inhomogeneous Permittivity, Permeability,

and Conductivity Solution in Rectangular

DRA

ductivity in rectangular DRA was found. The solution is very complex. This is

based on the solution of rectangular waveguide lled with inhomogeneous per-

mittivity, permeability, and conductivity material. These amplitudes are called the

waveguide modes and are of the form cossin. Depending on the nature of wall

surfaces (PEC or PMC), four possible linear combinations can appear (cossin, sin

cos, sinsin, and coscos). The discrete modes enable us to visualize the resonator

as collection of L, C oscillators with different L, C values.

Keywords Inhomogeneous Permittivity Permeability Linear combinations

Sequential RLC circuits Discrete modes Complex solution

11.1 Introduction

and conductivity of the medium was found. These amplitudes are called the

waveguide modes and are of the form cossin and sincos which denotes linear

components. It turns out that depending on the nature of wall surfaces (PEC or PMC),

four possible linear combinations can appear (cossin, sincos, sinsin, and coscos).

In a rectangular DRA, weve got to applying in additional boundary conditions on top

and bottom surfaces to be the linear combinations as compared to waveguide. They

have two possible linear combinations of sin cos. Thus, the possible frequencies

obtained by solving them comes out to be an equivalent, but computationally simpler

way to pass on from waveguide physics to resonator physics is to just replace byin all

the waveguide formulae that express the tangential eld components in terms of the

longitudinal components. This is done after solving the full 3-D Helmholtz equations

using separation of variable in x, y, z. The discrete modes enable us to visualize the

resonator as collection of L, C oscillators with different L, C values. The outcome of all

these analyses enables us to write down the elds inside the resonator, as superposition

of four or three vector-valued basis functions.

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_11

234 11 Inhomogeneous Permittivity, Permeability, and Conductivity

x; y 0 1 d ve x; y 11:1a

lx; y l0 1 d vm x; y 11:1b

Er Ex; yexpcz

H r H x; yexpcz 11:1c

Curl Er jxlx; yH r

Curl H r r jx1 x; yE r

jrx; y 11:1d

jx1 x; y 1 E r jxx; yE r

x1 x; y

We dene

jrx; y

x; y 1 x; y 1 0 1 d ve x; y

x1 x; y

and by substituting into the Maxwell equations, it gives the following equations:

only. We can arrange the equations [i.e., (11.3a), (11.3b), (11.4a) and (11.4b)] as

follows:

c jxl Ey Ez;y

;

jx c Hx Hz;x

jx c Ex Hz;y

:

c jxl Hy Ez;x

11.2 Mathematical Model 235

Thus,

Ey 1 c jxl Ez;y

2 ;

Hx h jx c Hz;x

Ex 1 jxl c Ez;y

;

Hy h2 c jx Hz;x

h2 h2 x; y c2 x2 l

c2 x2 0 l0 1 dve 1 dvm

h20 k2 dve vm k2 d2 ve vm

where

h20 c2 x2 0 l0 c2 k2 ; k 2 x2 0 l0

Then,

1

Ex cEz;x jxlHz;y ;

h2

1

Ey 2 cEz;y jxlHz;x ;

h

1

Hx 2 cHz;x jxEz;y ;

h

1

Hy 2 cHz;y jxEz;x

h

Ez;y lHz;x Ez;x lHz;y

c 2 ;X jx ;X c 2 ;Y jx ;Y jxlHz 0 11:5a

h h2 h h2

Hz;y Ez;x Hz;x Ez;y

c ;X jx ;X c ;Y jx ;Y jxEz 0 11:5b

h2 h2 h2 h2

jxl n l l o

DH z jxlH z jx ; H

2 X z;x

; H

Y z;y

h2 h h2

11:6a

1 1

c ;Y Ez;x 2 ;X Ez;y 0

h2 h

and,

236 11 Inhomogeneous Permittivity, Permeability, and Conductivity

jx n o

2

DEz jxEz jx 2

;X Ez;x 2 ;Y Ez;y

h h h 11:6b

1 1

c ;X Hz;y 2 ;Y Hz;x 0

h2 h

or equivalently,

h2 n o

D h2 Ez ; X Ez;x ; Y E z;y

h2 h2

11:7a

ch2 1 1

; X H z;y ; Y H z;x 0

jx h2 h2

h2 n l l o

D h2 H z ;X H z;x ;Y Hz;y

l h2 h2

11:7b

ch2 1 1

2

;Y Ez;x 2 ;X Ez;y 0

jxl h h

ve vm v1 x; y; ve vm v2 x; y 11:7c

Ez v1 0 Ez v2 0 Ez

D h2 k2 d k 2 d2

Hz 0 v1 Hz 0 v2 Hz

v2 0 Ez @

k 2 d2 log 2 ; x

0 v2 Hz h @x

@ c @ c @ c @

log 2 ;y ; log h2 ; x log h2 ;y log h2 ;y

h @y jx @y jx @x jxl @x

@ l @ l @ E

z

log h2 ;x ; log 2 ;x log 2 ;y 0

@y h @x h @y Hz

p

We write c h20 h2 in (11.8) and then (11.8) can be expanded as:

Ez v 0 Ez v2 0 Ez

D h2 k2 d 1 k 2 d2

Hz 0 v1 Hz 0 v2 Hz

11.2 Mathematical Model 237

@ @ @ @

f11 x; y; k; d g11 x; y; k; d ; f12 x; y; k; d g12 x; y; k; d

@x @y @x @y

@ @ @ @ Ez

f21 x; y; k; d g21 x; y; k; d ; f22 x; y; k; d g22 x; y; k; d 0

@x @y @x @y Hz

where k h20 ,

f11 x; y; k; d log ; x g 11 x; y; k; d log ;y

h2 h2

c

f12 x; y; k; d log h2 ;y

jx

c

g12 x; y; k; d log h2 ;x

jx

c

f21 x; y; k; d log h2 ;y

jxl

c

g21 x; y; k; d log h2 ;x

jxl

l

f22 x; y; k; d log 2 ;x

hl

g22 x; y; k; d log 2 ;y

h

!

f11 @@x g11 @@y f12 @@x g12 @@y

lk; d

f21 @@x g21 @@y f22 @@x g22 @@y

where fab ; Iab are the functions of x, y, k; d: It is early to see that for small d; fkb and

gkb can be expanded in power of d with the series state from d0 .

In other words,

fab x; y; k; 0 0;

gab x; y; k; 0 0:

Writing therefore

and likewise,

238 11 Inhomogeneous Permittivity, Permeability, and Conductivity

We have

@ @

lk; d F x; y; k; d Gx; y; k; d

@x @y

@ @

d F1 x; y; k G1 x; y; k

@x @y

@ @

d2 F2 x; y; k G2 x; y; k Od3

@x @y

where

F x; y; k; d fab x; y; k; d j1 a;b 2j

Gx; y; k; d gab x; y; k; d j1 a;b 2j

F1 fab1 ; G1 fab1 ;

F2 fab2 ; G2 fab2 ;

Note that

F dF1 d2 F2 Od3

G dG1 d2 G2 Od3

Thus,

where

@ @

l1 k F1 x; y; k G1 x; y; k

@x @y

@ @

l2 k F2 x; y; k G2 x; y; k

@x @y

We dene

v1 k l1 k k 2 v1 x; yI2 ;

v2 k l2 k k 2 v2 x; yI2 ;

11.2 Mathematical Model 239

then

D kwx; y dv1 k d2 v2 k wx; y 0

With neglecting higher order terms of O d3 ; where

Ez x; y

wx; y

Hz x; y

Hy 0; where y = 0, b. Using the expression for Hx Hy in terms of Ez ; Hz and the

boundary conditions on Ez , it follows that the boundary conditions on H can be

replaced by Hz;x 0, where x = 0, a and Hz;y 0, where y = 0, b.

Where d 0; (the homogeneous case), w satises D kw 0:

We denote the solution to this by wmn 0

x; y since application of the boundary

conditions leads to (after separation of variables)

Am; numn x; y

wx; y

Bm; nvmn x; y

2 mpx npy

umn x; y p sin sin

ab a b

2 mpx npy

vmn x; y p cos cos

ab a b

Note that,

Za Zb

humn ; ukl i umn x; yukl x; ydxdy

0 0

dmk dnl

humn ; ukl i dmk dnl

humn ; ukl i 0

D kw0 0; k h20

240 11 Inhomogeneous Permittivity, Permeability, and Conductivity

0 Am; numn x; y

w0 x; y wx;y x; y

Bm; nvmn x; y

mp 2 np 2

k kmn0

h2 m; n

a b

are given as follows:

1

X

Ez t; x; y; z Am; numn x; y

expcmn z

Hz t; x; y; z m;n1 Bm; nvmn x; y

X 1 h i

wm;n

0

x; y exp cm;n z

m;n1

given as follows:

( )

Ez t; x; y; z X

0

Re wmn x; y expfjxt cmn zg

Hz t; x; y; z

mn

the x and y components of the electromagnetic elds are easily obtained for d 0:

Fields are easily obtained for d 0:

X 1

Ex x; y; z cmn Am; num;n ; xx; y

m;n h0 m; n

2

jxl0 Bm; nvm;n Xx; y expcm;n z ;

X 1

Ey x; y; z cmn Am; num;n ; xx; y

m;n h20 m; n

jxl0 Bm; nvm;n Yx; y expcm;n z ;

X 1

Hx x; y; z jx0 Am; num;n ; xx; y

m;n h20 m; n

cmn Bm; nvm;n Xx; y expcm;n z ;

X 1

Hy x; y; z jx0 Am; num;n ; xx; y

m;n h20 m; n

cmn Bm; nvm;n Yx; y expcm;n z ;

11.2 Mathematical Model 241

where

q

cmn h20 m; n k2

q

0

cmn kmn k2

Consider now d [ 0 and d is small. Then, we have to solve the following equation:

D kw dV1 k d2 V2 k1 w 0

Let

k k0 dk1 d2 k2 O d3 ;

w w0 dk1 d2 k2 O d3

O d0 D k0 w0 0;

O d1 D k0 w1 V1 k0 w0 k1 w0

O d2 D k0 w2 V1 k0 w1

V1 k0 w0 k1 V10 k0 w0 k1 w1 k2 w0 0

The solution to O d0 case has already been obtained:

For each m and n, we have two orthogonal solutions:

umn x; y 0

w01 and w02 :

mn 0 mn vmn x; y

Z

0K 2

wmn dxdy 1; k 1; 2 where knk2 nT n;

n 2 D2

0\x\a; 0\y\b;

mp 2 np 2

wmn

0

a b

242 11 Inhomogeneous Permittivity, Permeability, and Conductivity

0

in az and also

umn 0

wmn

0

A B

0 vmn

Awmn

01

Bwmn

02

To get

0

D kmn w1 k1 wmn

0 0

V1 kmn 0

wmn 0

mn ; k 1; 2 gives the following equation:

D E

k1 A wmn

01

; V1 kmn

0

wmn

0

0;

D E

k1 B wmn

02

; V1 kmn

0

wmn

0

0:

0D ED E1

wmn

01

; V1 kmn

0

wmn

01

wmn

01

; V1 kmn

0

wmn

02

@D ED EA

wmn

02

; V1 kmn

0

wmn

01

wmn

02

; V1 kmn

0

wmn

02

A 1 A

x k

B B

split into two eigenvalues

0 1

kmn dk .

Where k1 can be any one of the two eigenvalues of the secular matrix (2 2)

D E

wmn

0a

; V1 kmn

0

wmn

0b

1 a;b 2

Aa m; n

Let ; a 1; 2 be the corresponding normalized eigenvalues:

Ba m; n

jAa m; nj2 jBa m; nj2 1.

We denote the corresponding eigenvalues by kmn 1

a; a 1; 2.

The principal normalized eigenfunction of D corresponding to the eigenvalue

kmn

0

is split into kmn

0

dkmn

1

a; a 1; 2; and

Aa m; numn x; y

kmn

0a

; a 1; 2;

Ba m; nvmn x; y

respectively.

11.2 Mathematical Model 243

(dynamic or moving probe).

The antenna orientation at time t is specied by the vector

where

R/; h; w Rz /; Rx h; Rz w

i.e., /; h; w are the Euler angles. The antenna is a rigid body (like a top) that

carries a current density. Jt; r in its initial conguration. So, after some time the

volume current density within the antenna body is given by

Jb t; r J t; Rt1 r ; r RtB

where B is the antenna body space at the time t = 0, we wish to control the

orientation angles /t; ht; wt, 0 t T so that the radiation pattern of the

antenna is as close as possible to a given pattern. Let us say that the pattern is

specied by the vector potential Ad t; r in space.

Then, the vector potential produced by the rotating antenna is as follows:

Z Jb t jrr0 j ; r 0

l c

Ad t; r d 3 r0

4p jr r 0 j

R3

Z Jb t jrr0 j ; R1 t jrr0 j r 0

l c c

d 3 r0

4p jr r 0 j

RtB

Z J t jrRtnj ; R1 t jrRtnj r 0 Rtn

l c c

d3 n

4p jr Rtnj

B

Z

l r ^r ; Rtn 1 r

At; r J t ; R tn d 3 n

4pr c c c

B

in a region of space in the far-eld zone.

244 11 Inhomogeneous Permittivity, Permeability, and Conductivity

from the origin).

Then,

Z

jAt; r Ad t; r j2 dt sin h0 dh0 d/0

0;TX0

l

r

our time or region by c; the quantity to be minimized is taking 1

4p

2

Z Z

rAd t; r Jt R1 t^r ; n ; R1 tnd 3 n sin h0 dh0 d/0

c

0;T X0 B

where

r; h0 ; /0 r; h0 ; /0 ^r

Ad t; r Ad t; r; h; / r

constant

Once the optimal trajectory f/d t; hd twd tg has been determined by opti-

mizing this highly nonlinear functional, we decide how to apply machine torques

fsd t; sd tsd tg; 0 t T.

The rigid body carrying current so that the time fsd t; sd tsd t; 0 t T g is

as close the desired trajectory as possible.

The kinetic energy of the top is given in terms of its principal moments of inertia

by (golds term classical mechanism)

1

T I1 x21 I2 x22 I3 x23

2

where

x2 h_ sin w h_ cos w sin h;

x3 w_ h_ cos h:

11.2 Mathematical Model 245

We get,

1 1 2

T I1 h_ 2 /_ 2 sin2 h I3 w_ /_ cos h

2 2

Then, the Lagrangian after taking machine torque into account is given as:

where V mgl cos h; l being the distance between the CM of the top and the origin.

The equation of motion is given as:

d @L @L d @L @L d @L @L

; ; give

_

dt @ / @/ dt @ h_ @h dt @ w_ @w

d _

I1 / sin2 h I3 cos h w_ h_ cos h sw t;

dt

d _

I1 h I3 /_ sin hw_ /_ cos h I1 /_ 2 sin h cos h mgl sin h sh t;

dt

d _

I3 w h_ cos h sw t

dt

We dene

Zt

F / t s/ sds;

0

Zt

F w t sw sds;

0

I1 /_ 2 sin2 h I3 cos h w_ h_ cos h F/ t 11:8

I3 w_ h_ cos h Fw t 11:9

I1

h I3 / sin h w_ h_ cos h I1 /_ 2 sin h cos h mgl sin h s/ t 11:10

246 11 Inhomogeneous Permittivity, Permeability, and Conductivity

ZT

W s/ t/_ t sh th_ t sw tw_ t dt

0

Z

F_ / t/_ t sh th_ t F_ w tw_ t dt

This must be minimized subject to the constraints robotic of the equation of motion

nal orientation dened

(11.8)(11.10) and that the by /t; ht; wt is given.

Here, we calculate F/ t; Fw t; Fh t ; 0 t T and hence, fsd t; sd t;

sd t; 0 t Tg by putting

/t /d t

h t hd t

wt wd t

This gives us an algorithm for calculating the machine torque to be supplied over

the range [0, T], so that the top antenna follows a desired trajectory that will lead to

a radiation pattern that matches a given radiation pattern as closely as possible in a

given solid angle element X0 over a given time interval [0, T]. Another way to

design the machine torque is to minimize a weighted combination of the error

energy between the desired trajectories f/d ; hd ; wd g and the actual trajec-

tory, and the total work done by the torques over the duration [0, T] is minimized as

follows:

ZT ZT 2

Sa F_ / t/_ t sh th_ t F_ w tw_ t dt b /t /d t2 ht hd t

0 0

2

wt wd t dt

write the equation as:

ZT

_ /;

S n /; /; h; h;

_ h; w; w;

_ w dt

0

Euler trajectories equations:

11.2 Mathematical Model 247

d @n d2 @n @n

;

dt @ /_ dt2 @ /

@/

d @n d2 @n @n

;

dt @ h_ dt2 @ h @h

d @n d2 @n @n

:

dt @ w_ dt2 @ w

@w

Now support that the electromagnetic eld generated by an antenna falls on the

aperture z = 0, 0 < x < a, 0 < y < b, of a rectangular waveguide.

We wish to complete the elds inside the guide. Assuming the guide to have

constant permittivity and permeability, we get for the phase elds inside the guide

at a given frequency x,

Ez x; y

D h20 0

Hz x; y

h20 c2 x2 0 l0 c2 k2

We have

1

Ex 2

cEz;x jxlHz;y

h0

1

Ey 2 cEz;y jxlHz;x

h0

1

Hx 2 cHz;x jxEz;y

h0

1

Hy 2 cHz;y jxEz;x

h0

The general solution for the elds within the given satisfying boundary condi-

tions is (as weve seen earlier) given by

1

X

Ez x; y; z Am; numn x; y

ecmn z

Hz x; y; z Bm; nvmn x; y

m;n1

2 mpx npy

umn x; y p sin sin ;

ab a b

2 mpx npy

vmn x; y p cos cos

ab a b

q

cmn cmn x h20 m; n x2 l0 0

m2 2

where h20 m; n p2 a2 nb2 :

248 11 Inhomogeneous Permittivity, Permeability, and Conductivity

When anelectromagnetic

eld is incident on the surface z = 0 of the guide, let

E0x x; y; E0y x; y ; be the incident electric eld (tangential components to the

surface). Then by continuity of the tangential components of the electric eld, we have

Ex x; y; 0 E0x x; y; Ey x; y; 0 E0y x; y;

0 x a; 0 y b;

where

X 1

Ex x; y; 0 cmn Am; numn x; y jxl0 Bm; nvmn x; y

m;n h20 m; n

and

X 1

Ey x; y; 0 cmn Am; numn x; y jxl0 Bm; nvmn x; y

m;n h20 m; n

Thus

X c 2 mp mpx npy

E0x x; y mn

A m; n p cos sin

m;n h20 m; n ab a a b

jxl 2 np mpx npy

2 0 p Bm; n cos sin

h0 m; n ab b a b

np mpx npy

2 X cmn

E0y x; y p A m; n sin cos

ab m;n h20 m; n b a b

jxl mp mpx npy

2 0 Bm; n sin cos

h0 m; n a a b

or equivalently

!

X 2cmn mp 2jxl0 np mpx npy

E0x x; y p A m; n p B m; n cos sin

m;n h20 m; n ab h20 m; nb ab a b

and

!

X 2cmn np 2jxl0 mp mpx npy

E0y x; y p Am; n 2 p Bm; n sin cos

m;n h0 m; nb ab

2 ah0 m; n ab a b

11.2 Mathematical Model 249

Thus, we get

Z mpx npy

2

p E0x x; y cos sin dxdy

ab a b

0\x\a; 0\y\b

mpcmn jxl0 np

A m; n B m; n

ah20 m; n h20 m; nb

These are the two simultaneous linear equations for the two variables

Am; n and Bm; n which are easily solved. 2 2 matrix notation supports the

incident electric eld is E0 x; y; z; t: Then, we take

Z

^ 0x x; y; x

E E0x x; y; 0; tejxt dt

R

Z

^ 0y x; y; x

E E0y x; y; 0; tejxt dt

R

Dene

Z mpx npy

2

/x m; n; t p E0x x; y; 0; t cos sin dxdy

ab a b

D

Z mpx npy

2

/y m; n; t p E0y x; y; 0; t sin cos dxdy

ab a b

D

Then,

Z

^ m; n; x ,

/ /x m; n; tejxt dt

x

R

Z

2 ^ 0x x; y; x cos mpx sin npy dxdy

, p E

ab a b

D

Z

^ m; n; x ,

/ /y m; n; tejxt dt

y

R

Z

2 ^ 0y x; y; x sin mpx cos npy dxdy

, p E

ab a b

D

250 11 Inhomogeneous Permittivity, Permeability, and Conductivity

RDRA Can Be Used for Polarization Diversity

S11

Gain

Radiation pattern

VSWR

Dispersion

Polarization

Permittivity and permeability

Bandwidth

S21

Isolation

Efciency

Directivity

Resonant frequency

Propagation constants

Axial ratio

Resonant mode

Dominant mode

Higher-order modes

Chapter 12

Case Studies

Abstract This chapter deals with case studies, where implemented cases have been

discussed. Various type of antennas have been fabricated. Their dominating para-

meters are shown. These cases have either been developed using HFSS simulations

or hardcore experimentations. The case study is based on the rectangular DRA

using ceramics such as eccostock-500. Nomenclature and parameters obtained have

been mentioned below the each gure. Emphasis is also given on geometry of

antennas and their experimental results. The experimental results have been

obtained under specic environmental conditions i.e. anechoic chambers.

Keywords Rectangular DRA Simulated and experimental results Isolated RDRA

and RDRA with ground plane Single feed Double feed Anechoic chamber

measurements Radiation pattern Gain S11, VSWR, Z11, and E and H elds

distribution RF absorbers Test set for measurement Prototype Azimuth and

elevation pattern Manganesemanganese material Bandwidth enhancement

Higher order resonant modes Variable DRA height Smith chart Group

delay Rectangular wells LHCP RHCP Circular polarization Phase distortion

S21 Ferrite RDRA Slot variation Permittivity variation effects Hardware

implementation VNA calibration Aperture coupled RDRA and probe fed RDRA

The case study is based on the rectangular DRA where the various designs of DRA

have been presented and the nomenclature is given below each gure. These

antennas have been simulated and fabricated. The results obtained have been pre-

sented graphically here. These antennas also have been placed inside anechoic

chamber to minimize external effects that come during measurements. The gures

indicate various measurement steps involved in this process. For simulated results,

Ansoft HFSS 13.0 has been used. These graphs of radiation pattern, gain, S11,

VSWR, Z11, and eld distribution have been presented and their domains are

mentioned below each gure (Figs. 12.1, 12.2, 12.3, 12.4, 12.5, 12.6 and 12.7).

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3_12

252 12 Case Studies

Fig. 12.1 a Model of electronic band gap (EBG) structure cavity rectangular dielectric resonator.

b Diagrammatic representation of RDRA with top-loading DRA

Fig. 12.2 Positioning of RDRA antenna ready for test procedure setup

12.1 Structure and Hardware Experimentations 253

Fig. 12.4 RDRA antenna between RF absorbers inside chamber for gain testing

Fig. 12.5 View of RDRA antenna under test setup on sliding table

Fig. 12.6 View of RDRA antenna under test setup on sliding table

Fig. 12.7 RDRA antenna on sliding table with variation in position of RDRA

See Figs. 12.8, 12.9, 12.10, 12.11, 12.12, 12.13, 12.14, 12.15, 12.16, 12.17, 12.18,

12.19, 12.20, 12.21, 12.22 and 12.23.

254 12 Case Studies

Fig. 12.8 E-plane radiation pattern at 13.7 GHz of simulated RDRA for electric eld distribution

Fig. 12.9 H-plane radiation pattern of simulated RDRA for magnetic eld distribution

12.1 Structure and Hardware Experimentations 255

Fig. 12.10 VSWR measurement for measurement of reflected eld strength of simulated RDRA

Fig. 12.11 H-plane radiation pattern at 16.8 GHz of simulated RDRA for magnetic eld distribution

256 12 Case Studies

as Dielectric

been presented. This dielectric material shows the various effects on the parameters

of the developed antenna. The bandwidth enhancement techniques have been

implemented using two wells. The results obtained have been presented graphically

here. These antennas also have been placed inside anechoic chamber to minimize

external effects that come during measurements. The gures indicate various

measurement steps involved in this process. These antennas have been simulated

and fabricated. For simulated results, Ansoft HFSS 13.0 has been used. These

graphs of radiation pattern, gain, S11, VSWR, Z11, and eld distribution have been

12.2 RDRA with ManganeseManganese Material as Dielectric 257

Fig. 12.13 Radiation pattern at 13.7 GHz for radiated eld pattern of simulated RDRA

258 12 Case Studies

Fig. 12.14 Return loss S11 of higher-order modes excited inside simulated RDRA

Fig. 12.15 Gain of simulated RDRA at various heights (with the excitation of higher-order modes)

Fig. 12.16 Excitation of higher mode at RDRA 10 mm height with variable frequency of 10, 12,

15 GHz

12.2 RDRA with ManganeseManganese Material as Dielectric 259

Fig. 12.17 RDRA height 15 mm, excitation frequency variable, generated higher modes. Even

modes excitation with top excitation a TE114 at 11.7 GHz b TE116 at 13.7 GHz c TE118 at

16.7 GHz in RDRA

260 12 Case Studies

Fig. 12.20 Prototype RDRA under test setup for measurements with VNA

12.2 RDRA with ManganeseManganese Material as Dielectric 261

Fig. 12.21 Experimental gain of antenna of 16.624 dBi at 16.8 GHz RDRA inside microwave

anechoic chamber

Fig. 12.22 E-plane radiation pattern at 16.8 GHz dBi inside microwave anechoic chamber

amplitude versus theta

262 12 Case Studies

Fig. 12.23 H-plane radiation pattern at 16.8 GHz dBi inside microwave anechoic chamber

amplitude versus theta

presented and their domains are mentioned below each gure. The phase versus

frequency plots indicate distortions in the developed RDRA. The group delay,

forward power has also been indicated in simulated results (Figs. 12.24, 12.25,

12.26, 12.27, 12.28, 12.29, 12.30, 12.31, 12.32, 12.33, 12.34, 12.35, 12.36, 12.37

and 12.38).

12.2 RDRA with ManganeseManganese Material as Dielectric 263

m1 Curve Info

dB(S(1,1))

Setup1 : Sweep

-6.00

-8.00

dB(S(1,1))

-10.00

-12.00

-14.00

-16.00

m1

-18.00

10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00

Freq [GHz]

Fig. 12.26 Return loss curve S11 of simulated DRA at 15.5 GHz

0 Curve Info

rETotal

-30 30 Setup1 : Sweep

14.00 Freq='15.34673367GHz' Phi='0deg'

rETotal

Setup1 : Sweep

10.50

Freq='15.34673367GHz' Phi='90deg'

-60 60

7.00

3.50

-90 90

-120 120

-150 150

-180

264 12 Case Studies

50.00

Curve Info

dB20(Z(1,1))

Setup1 : Sweep

45.00

40.00

dB20(Z(1,1))

35.00

30.00

25.00

20.00

10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00

Freq [GHz]

m2 56.0000 56.0000 5.0286 GainTotal

-30 30 Setup1 : Sweep

4.80 Freq='15.34673367GHz' Phi='0deg'

GainTotal

Setup1 : Sweep

3.60

Freq='15.34673367GHz' Phi='90deg'

-60 m2 60

2.40

1.20

m1

-90 90

-120 120

-150 150

-180

Fig. 12.29 Gain plot of 4.80 dBi of simulated DRA at 15.5 GHz

Substrate e 2:2 20 30 0:8 mm3

DRA material e 12:2 12 manganesemanganese

12.2 RDRA with ManganeseManganese Material as Dielectric 265

Name X Y

XY Plot 4 HFSSDesign1 ANSOFT

80.00

m1 70.0000 4.3036 Curve Info

dB(AxialRatioValue)

Setup1 : Sweep

70.00 Freq='13.34170854GHz' Phi='0deg'

dB(AxialRatioValue)

Setup1 : Sweep

60.00 Freq='13.34170854GHz' Phi='90deg'

dB(AxialRatioValue)

50.00

40.00

30.00

20.00

10.00

m1

0.00

0.00 125.00 250.00 375.00

Theta [deg]

dB(rELHCP)

-30 30 Setup1 : Sweep

20.00 Freq='15.34673367GHz' Phi='0deg'

m1 dB(rELHCP)

Setup1 : Sweep

10.00

Freq='15.34673367GHz' Phi='90deg'

-60 60

0.00

-10.00

-90 90

-120 120

-150 150

-180

266 12 Case Studies

dB(rERHCP)

-30 30 Setup1 : Sweep

19.00 Freq='15.34673367GHz' Phi='0deg'

m1 dB(rERHCP)

Setup1 : Sweep

13.00 Freq='15.34673367GHz' Phi='90deg'

-60 60

7.00

1.00

-90 90

-120 120

-150 150

-180

12.2 RDRA with ManganeseManganese Material as Dielectric 267

Fig. 12.34 E eld pattern when single feed along y-axis of simulated DRA

Fig. 12.35 E eld pattern when single feed along x-axis applied of simulated DRA

268 12 Case Studies

Fig. 12.36 Group delay measurements in RDRA dual feed of simulated DRA

Fig. 12.37 Phase versus frequency plot (phase distortion) of simulated DRA

12.3 Dual-Feed RDRA Hardware and Measurements 269

In this case, the designing of RDRA dual-feed mechanism has been implemented

for circular polarization. Ferrite DRA has been used for bandwidth enhancement

using magnetization concept. The results obtained have been presented graphically.

These antennas also have been placed inside anechoic chamber to minimize

external effects that come during measurements. The gures indicate various

measurement steps involved in this process. These antennas have been simulated

and fabricated. For simulated results, Ansoft HFSS 13.0 has been used. These

graphs of radiation pattern, gain, S11, VSWR, Z11, eld distribution have been

presented and their domains are mentioned below each gure. The impedance

versus frequency has been presented. The hardware results using VNA for return

loss S11 have been also included (Figs. 12.39, 12.40 and 12.41; Table 12.1).

Optimization of the feed position for impedance match to have maximum gain

(Figs. 12.42, 12.43, 12.44, 12.45, 12.46, 12.47, 12.48, 12.49, 12.50, 12.51, 12.52,

12.53, 12.54, 12.55, 12.56, 12.57, 12.58, 12.59, 12.60, 12.61, 12.62 and 12.63).

270 12 Case Studies

30mm

20mm

S. No. Element Dimension (mm)

1 Ground plane 20 30

2 Substrate er 2:2 20 30 0.8

3 DRA er 12:2 4.6 9 10.8

4 Width of micro-strip 2.4

5 Length of stub and micro-strip 18.693

6 Slot (l w) 3.743 0.404

12.3 Dual-Feed RDRA Hardware and Measurements 271

position for maximum gain in 10

gain at dominant mode

RDRA 8

6

4

2

0

-1 -2 -3 -4 -5 1 2 3 4

slot position (mm) ---------->>

S11 for single and double 0

feed of RDRA

-5

<<<----------- s11(db)

-10

-15

-20

single feed

Double feed

-25

10.71

11.51

12.31

13.11

13.91

14.71

15.52

16.32

17.12

17.92

18.72

19.52

7.50

8.30

9.10

9.90

frequency(GHZ) -------->>

for single and double feed in 60

single feed

RDRA double feed

50

Impedance(ohm) --- >>

40

30

20

10

0

7.50

8.13

8.75

9.38

10.01

10.63

11.26

11.88

12.51

13.14

13.76

14.39

15.02

15.64

16.27

16.89

17.52

18.15

18.77

19.40

Frequency(GHZ) ----->>

272 12 Case Studies

in permittivity on return loss 5.00

er 9.8 er 20 er 40

S11 of RDRA 0.00

-5.00

<<<------S11(db)

-10.00

-15.00

-20.00

-25.00

-30.00

-35.00

-40.00

7.50

8.30

9.10

9.90

10.71

11.51

12.31

13.11

13.91

14.71

15.52

16.32

17.12

17.92

18.72

19.52

Frequency(GHz)------->>>

in permittivity on return loss 0

S11 of RDRA

-5

er 12.9

-10

<<---- S11(db)

-15

er 20

-20

-25

er 30

-30

-35

-40

7.50

8.75

10.01

11.26

12.51

13.76

15.02

16.27

17.52

18.77

Frequency(GHz) ----->>

12.3 Dual-Feed RDRA Hardware and Measurements 273

0

m1 GainTotal

-30 30 Setup1 : Sweep

2.00 Freq='8.251503006GHz' Phi='0deg'

GainTotal

Setup1 : Sweep

1.50

Freq='8.251503006GHz' Phi='90deg'

-60 60

1.00

0.50

-90 90

-120 120

-150 150

-180

m2 60.0000 60.0000 0.6570 m1 GainTotal

-30 30 Setup1 : Sweep

1.60 Freq='10.05511022GHz' Phi='0deg'

GainTotal

Setup1 : Sweep

1.20 Freq='10.05511022GHz' Phi='90deg'

-60 60

0.80

0.40 m2

-90 90

-120 120

-150 150

-180

274 12 Case Studies

m2 54.0000 54.0000 7.1650 GainTotal

-30 30 Setup1 : Sweep

8.00 Freq='18.79759519GHz' Phi='0deg'

GainTotal

Setup1 : Sweep

m1 6.00 Freq='18.79759519GHz' Phi='90deg'

-60 m2

60

4.00

2.00

-90 90

-120 120

-150 150

-180

Fig. 12.50 Front and rear view of hardware implemented of dual-feed RDRA

Fig. 12.51 Short, open, and load termination for calibration of VNA

12.3 Dual-Feed RDRA Hardware and Measurements 275

Fig. 12.52 Top and side view of single- and double-feed aperture couple feed of RDRA

double-feed aperture couple

feed of RDRA

RDRA

276 12 Case Studies

slot in ground plane of

single-feed RDRA

12.3 Dual-Feed RDRA Hardware and Measurements 277

278 12 Case Studies

Fig. 12.60 Fabricated model of RDRA under test with single feed and slot

12.4 Isolated and Grounded RDRA 279

Fig. 12.63 Top view with single feed and SMA connector of fabricated RDRA

Design of Isolated DRA: Isolated and grounded RDRA has different lengths due to

image theory. Isolated RDRA is shown in Fig. 12.64. The RDRA is excited by a

coax feed. Ground plane is absent in the rst design. The rectangular DRA height

can be reduced to half if we use ground plane of nite dimensions (Table 12.2).

Return loss of isolated DRA is shown in Fig. 12.65. It has resonant frequency of

3.99 GHz with 41.74 return loss.

280 12 Case Studies

isolated DRA

Dimension of DRA in Y-direction = 18.62 mm

Dimension of DRA in Z-direction = 4.6 mm

Permittivity of DRA = 37.84

0.00 Curve Info

dB(St(coaxpin_T1,coaxpin_T1))

dB(St(coaxpin_T1,coaxpin_T1))

Setup1 : Sweep1

-5.00 Name X Y

m1 3.9900 -41.7414

m2 3.9133 -9.7575 m2 m3

-10.00

m3 4.0772 -9.8660

-15.00

-20.00

-25.00

-30.00

-35.00

-40.00 m1

-45.00

3.50 3.63 3.75 3.88 4.00 4.13 4.25

Freq [GHz]

12.4 Isolated and Grounded RDRA 281

Gain plot is shown in Fig. 12.66. It shows that isolated DRA has 4.7 dB gain at

resonant frequency.

Impedance plot of isolated DRA is shown in Fig. 12.67. This has the real imped-

ance nearly 50 at resonant frequency.

Ansoft

NameLLC Theta Ang Mag Radiation Pattern 1 HFSSDesign1

m3 90.0000 90.0000 4.7037 0 Curve Info

dB(GainTotal)

-30 30 Setup1 : Sweep1

0.00 Freq='3.99749499GHz' Phi='0deg'

dB(GainTotal)

Setup1 : Sweep1

-10.00

Freq='3.99749499GHz' Phi='90deg'

-60 60

-20.00

-30.00

m3

-90 90

-120 120

-150 150

-180

62.50 Curve Info

re(Zt(coaxpin_T1,coaxpin_T1))

Name X Y Setup1 : Sweep1

m8

50.00 m8 3.9930 50.7358 im(Zt(coaxpin_T1,coaxpin_T1))

m9 3.9915 -0.2234 Setup1 : Sweep1

37.50

25.00

Y1

12.50

m9

0.00

-12.50

-25.00

3.50 3.63 3.75 3.88 4.00 4.13 4.25

Freq [GHz]

282 12 Case Studies

Image theory has been applied to reduce the height of the antenna. HFSS model of

DRA with ground plane has been developed using excitation with coaxial feed.

Here, the height of DRA has been reduced to half as compared to isolated DRA

(Fig. 12.68; Table 12.3).

Simulated return loss of DRA with ground is shown in Fig. 12.69. It has resonant

frequency 4.18 GHz with 28 dB return loss.

Gain plot is shown in Fig. 12.70. It shows that antenna radiates in the end re

direction and holds the value of gain 4.62 dB at resonant frequency.

DRA with ground

DRA with ground plane

Dimension of DRA in Y-direction = 9.31 mm

Dimension of DRA in Z-direction = 4.6 mm

Permittivity of DRA = 37.84

12.4 Isolated and Grounded RDRA 283

0.00

Name X Y

dB(St(coaxpin_T1,coaxpin_T1))

-5.00 m2 4.1343 -9.7735 dB(St(coaxpin_T1,coaxpin_T1))

Setup1 : Sweep1

m3 4.2345 -10.1310

m2 m3

-10.00

-15.00

-20.00

-25.00

m1

-30.00

3.00 3.25 3.50 3.75 4.00 4.25 4.50 4.75 5.00

Freq [GHz]

Ansoft

NameLLC Theta Ang Mag Radiation Pattern 1 HFSSDesign1

m3 90.0000 90.0000 4.6291 0 Curve Info

dB(GainTotal)

-30 30 Setup1 : Sweep1

0.00 Freq='4.186372745GHz' Phi='0deg'

dB(GainTotal)

Setup1 : Sweep1

-10.00

Freq='4.186372745GHz' Phi='90deg'

-60 60

-20.00

-30.00

m3

-90 90

-120 120

-150 150

-180

Impedance plot of DRA with ground is shown in Fig. 12.71. This has the real

impedance nearly 48.2 and reactive part is 0.17 at resonant frequency.

284 12 Case Studies

50.00 m6

Curve Info

Name X Y

re(Zt(coaxpin_T1,coaxpin_T1))

m6 4.1984 48.1805

Setup1 : Sweep1

37.50 m7 4.1824 0.1700

im(Zt(coaxpin_T1,coaxpin_T1))

Setup1 : Sweep1

25.00

12.50

Y1

m7

0.00

-12.50

-25.00

-37.50

3.00 3.25 3.50 3.75 4.00 4.25 4.50 4.75 5.00

Freq [GHz]

From the comparison in Table 12.4, it is seen that resonant frequency and gain for

isolated DRA and DRA with ground are approximately same but return loss and

impedance bandwidth are better in isolated DRA (Fig. 12.72).

Parameter Isolated With ground

Resonant frequency (GHz) 3.99 4.16

Gain (dB) 4.7 4.629

B.W. (GHz) 0.17 0.1

Return loss (dB) 41.04 28

150.00 Curve Info

Name X Y im(Z(1,1))

Setup1 : Sweep1

125.00 m1 5.1303 49.9819

re(Z(1,1))

m2 5.1363 -2.4335 Setup1 : Sweep1

100.00

75.00

m1

50.00

Y1

25.00

0.00 m2

-25.00

-50.00

-75.00

4.00 4.50 5.00 5.50 6.00 6.50 7.00

Freq [GHz]

12.4 Isolated and Grounded RDRA 285

Figure 12.73 shows the view of a simple structure of rectangular DRA. The rect-

angular DRA of length L, width W, and height H is placed over a slot, cut at the

center of ground plane of size 50 50 mm2. The micro-strip line of length Lm and

width We is placed on the other side of the ground plane. The dielectric material

used for substrate is having permittivity, er 10:2 and thickness 0.64 mm. The

dielectric material used for DRA is Rogers RT/Duroid 6010/6010LM having per-

mittivity er 10:2. All other the dimensions have been shown in Table 12.5

(Fig. 12.74).

L W H Ls Ws Lm Wm Er er H (sub)

(mm) (mm) (mm) (mm) (mm (mm) (mm) (sub) (dra) (mm)

10.6 6 9.6 7.2 1.2 28 0.6 10.2 10.2 0.64

286 12 Case Studies

0.00 Name X Y Curve Info

m5 6.9000 -24.4423 dB(S(1,1))

m6 6.5800 -10.2532 Setup1 : Sweep1

k='28mm'

m7 7.3400 -9.8973

-5.00

dB(S(1,1))

m7

-10.00 m6

-15.00

-20.00

m5

-25.00

5.00 5.50 6.00 6.50 7.00 7.50 8.00

Freq [GHz]

Return loss of antenna 1 is shown in Fig. 12.75. Resonant frequency is 6.89 GHz

with return loss 24 dB.

12.4 Isolated and Grounded RDRA 287

300

250

im(Z(1,1))

200

re(Z(1,1))

150

Impedance

100

50

0

-50

-100

-150

5.0 5.4 5.8 6.2 6.6 7.0 7.4 7.8

Frequency [GHz]

8.796 dBi (Figs. 12.76 and 12.77).

Annexure-1

1. MgOSiO2 6.3 Countis Laboratories

(CD-6) 12295 Charles Dr, Grass Valley, CA 95945,

2. MgOSiO2TiO2 9.5 United States

(CD-9) +1 530-272-8334

tcountis@countis.com

3. MgOTiO2SiO2 13.0

(CD-13)

4. MgOTiO2 15.0

(CD-15)

5. MgOTiO2 16.0

(CD-16)

6. MgoCaOTiO2 18.0

(CD-18)

7. MgoCaOTiO2 20.0

(CD-20)

8. MgoCaOTiO2 30.0

(CD-30)

9. MgoCaOTiO2 50.0

(CD-50)

10. MgoCaOTiO2 100.0

(CD-100)

11. MgoCaOTiO2 140.0

(CD-140)

(continued)

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3

290 Annexure-1

12. Boron nitride 4.0 Emerson & Cuming Microwave

(ECCOSTOCK@) Products N.V.

13. Beryllium oxide 6.0 A unit of Laird Technologies

(ECCOSTOCK@) Hong Kong Holdings (4) Ltd.

Unit 2507-8, 25/F, Ofce Tower, Langham

14. Magnesium oxide 9.0

Place,

(ECCOSTOCK@)

8 Argyle Street, Mongkok

15. Magnesium titanate 10.0 Kowloon, Hong Kong

(ECCOSTOCK@) Tel: +852-2923 0600 Call: +852-2923 0605

16. Zirconia 20.0 Email: sales@hk.eccosorb.com

(ECCOSTOCK@)

17. Titanium dioxide (rutile) 50.0

(ECCOSTOCK@)

18. Strontium titanate >100.0

(ECCOSTOCK@)

19. Magnesium manganese 9.2 Hiltek Microwave Limited

Aluminum iron ferrite 15200 Shady Grove Road Suite 350

20. Magnesium titanate 16.0 Rockville, Maryland 20850

United States

21. Lithium ferrite 20.0

(301) 670-2833

22. Zirconium tin titanate 37.0 (301) 670-2831 Fax

23. Titania ceramic 80100 www.hiltek.com

24. MgSi (Steatile) 6.0 Morgan advanced Materials

(D6) 150 Kampong ampat

25. CaMgTi (Mg, Ca 20.0 05-06a

titanate) Ka Centre

(D20) Singapore

368324

26. ZrTiSn (Zr, Sn titanate) 37.0

t +65 6595 0000

(D36)

F +65 6595 0005

27. BaSmTi (Ba, Sm 76.5 asia.mc@morganplc.com

titanate)

(D37)

(continued)

Annexure-1 291

28. Titanate with other 6.5 Pacic Ceramics, Inc.

ingredients Advanced Microwave Ceramic Materials

(PD-6) 824 San Aleso Ave Sunnyvale, CA 94085

29. Titanate with other 9.5 USA (408) 747-4600

ingredients info@pceramics.com

(PD-9)

30. Titanate with other 12.0

ingredients

(PD-12)

31. Titanate with other 13.0

ingredients

(PD-13)

32. Titanate with other 15.0

ingredients

(PD-15)

33. Titanate with other 16.0

ingredients

(PD-16)

34. Titanate with other 18.0

ingredients

(PD-18)

35. Titanate with other 25.0

ingredients

(PD-25)

36. Titanate with other 38.0

ingredients

(PD-38)

37. Titanate with other 50.0

ingredients

(PD-50)

(continued)

292 Annexure-1

38. Titanate with other 98.0

ingredients

(PD-100)

39. Titanate with other 160.0

ingredients

(PD-160)

40. Titanate with other 270.0

ingredients

(PD-270)

41. Zr Sn Ti oxide 37.0 Temex Components & Temex Telecom,

(E2000) USA

42. E3000 34.0 Supplier 1

SM CREATIVE

43. Ba Zn Ta oxide 30.0

No 845, 2nd Cross, 7th Main HAL 2nd

(E4000)

Stage

44. Ba Sm Ti oxide 78.0 Indiranagar, Bangalore, 560 038

(E5000) India

45. Ti Zr Nb Zn oxide 45.0 sundar@smcel.com

(E6000) +91 (80) 25210268

+91 (80) 41255492

Mobile Phone Number: +91 (98) 45410417

http://www.smcel.com

S M Creative Electronics Ltd

#10, Electronic City, Sector-18

Gurgaon 122 015, Haryana

Tel: +91 124-4909850

Fax: +91 124-2455 212

smcel@smcel.com

Supplier 2

SIMAL

# 60 & 60/1, 18th Cross, 4th Main

Malleswaram, Bangalore, 560 055

India

agencies@simal.com.sg

+91 (80) 41532079

+91 (80) 23444410

Mobile Phone Number: +91 (99721) 24165

http://www.simal.com.sg

46. Cordierite (Mg, Al, 4.5 Trans-Tech

silicate) Skyworks Solutions, Inc.

(D-4) 5520 Adamstown Road

47. Forsterite (Mg, Si, 6.3 Adamstown, MD 21710

oxide) Supplier

(D-6) SM Electronic Technologies Pvt. Ltd.

#1790, 5th Main, 9th Cross, RPC Layout

48. MgTi 15.0

Vijayanagar 2nd Stage

(D-15)

Bangalore 560 040

49. MgTi 16.0 India

(D-16)

(continued)

Annexure-1 293

50. Ba, Zn, Ta-oxide 29.030.7 Mr. Manjunath

(D-29) +91-80-23301030

51. Ba, Zn, Ta-oxide 29.531.0 smgroup@vsnl.com

(perovskite)

(D-87)

52. BaZnCoNb 35.036.5

(D-83)

53. Zirconium titanate based 44.746.2

(D-43)

54. E-11 11.0 T-CERAM, RF & Microwave

55. E-20 20.0 Okrun 1144

500 03 Hradec Krlov

56. E-37 37.0

Czech Republic, EU

sales@t-ceram.com

www.t-ceram.com

+420 774 406 438

CZ 42196078

57. TE-21 21.0 Token Electronics Industry Co., Ltd.

58. TE-30 30.0 No. 137, Sec. 1, Chung Shin Rd., Wu Ku

Hsiang, Taipei Hsien, Taiwan, R.O.C

59. TE-36 36.0

TEL: 886-2-2981 0109; FAX: 886-2-2988

60. TE-45 45.0 7487

61. TE-80 80.0 http://www.token.com.tw rfq@token.com.

62. TE-90 90.0 tw

63. MgCaTi 20.0 MCV Microwave

(MDR20) 6640 Lusk Blvd, Suite A102 San Diego, CA

64. Ta with other ingredients 24.0 92121

(MDR24) Tel: 858-450-0468

Fax: 858-869-8404

65. Ta with other ingredients 30.0

www.mcv-microwave.com

(MDR30)

66. ZnSnTi 38.0

(MDR38)

67. LaBaTi 45.0

(MDR45)

68. DR-30 30.0 TCI Ceramics, Inc.

69. DR-36 36.0 18450 Showalter Rd., Hagerstown, MD

21742

70. DR-45 45.0

Ph: 301-766-0560 Fax: 301-766-0566

71. DR-80 80.0 E-mail: sales@tciceramics.com

www.tciceramics.com

72 RT6010, RT-6002 10.2

73 MCT-25 25 TRANS TECH

SMAT 27

BaTiO3 14

74 ECCOSTOCKSHIK 10,20,30,40

Annexure-2

in Cavity Resonator

Probe inserted dl

Characteristic equation of RDRA is given below:

2

A2:1

r2? h2 Hz or Ez 0; Helmholtz equation

X

Ez Cmn sinmpx=a sinnpy=b ecmnz ejxt

In the above equation, Cmn are the amplitude coefcients and wave is propagating

in z-direction

q q

) cmn Propagation constant h2mn k2 h2mn x2 l

Hence, computation of eld Ez when all the four sides of resonator are trans-

parent and magnetic walls (PMC walls) and top and bottom walls are PEC

(Electrical walls). We are well versed that Hz 0 at magnetic walls and Ez 0 at

electric walls.

The feed probe is inserted into rectangular DRA at point (a/2, b/2) in z-direction.

I(t) Current can be expressed in terms of magnetic vector potential Az.

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3

296 Annexure-2

ground plane

RDRA

^z lIdl jkr

Az e ; r is far field point:

4pr

jk^r ^z lIdl ejkr

div A

4pr

jk cos h lIdl ejkr

div A ;

4pr

jx6 9

2

c

Annexure-2 297

Let,

p kc lIdl

69 a 2 b2 ejkr

x 4pr

cos hlIdl jkr

e

4pr

Now,

1

Er 0 O 2

r

Eh jxAh jxAz; at z 0

E jxA 0 at z 0

Hence,

Eh Ez jxAz

jxlIdl jkr 2 mpx npy

e p sin sin

4pr ab a b

p

where r x2 y2

dm; m0 ; dn; n0 hunm ; um0 n0 i; where z = 0; (property of orthogonality as product

of basis function becomes zero)

X

Ez Cmn umn x; yecmnz

mn

hmn x

At z = 0

X jxlIdl p

p ejk x y ;

2 2

Ez ) Cmn umn x; y

mn 4p x y

2 2

Za Zb r

jxlIdl umnx;y 2 2

Cmn pejk x a=2 x b=2 dxdy;

4p x2 y2

0 0

298 Annexure-2

Za Zb r

jxlIdl umnx;y jk

2 2

Cmn r

2 e x a=2 x b=2 dxdy

4p 2

0 0 x a=2 y b=2

A2:2

m 2 n 2

Hence p

2

x2

a b

if a [ b and m 1; 2; 3; . . . n 1; 2; 3; . . .

2 1 1 2 2 1

p x \p

2

a2 b2 a2 b2

r r

1 1 x 2 1

\

a2 b2 p a2 b2

jpp

cmn

d

k2 c2mn h2mn

p2 p2

hence, k2 h2mn

d2

umn depends on input excitation;

hmn resonant mode (cut off frequency); and

k-propagation constant.

Generation of modes or characteristics frequencies xmnp e.m. of electro-

magnetic elds oscillations inside the cavity resonator has been discussed. The

basic Maxwells theory can be applied with boundary conditions to express res-

onator elds as superposition of these characteristics frequencies.

The elds

X Z

Ez x; y; z; t Re Cmnp ejxmnpt umnp x; y; z

mnp

or

X

Cmnp umnp x; y; z cosxmnp Umnp;

mnp

Annexure-2 299

npy

where umn x; y p2

ab

sin mpx

a sin b ; xmnp is the characteristic frequency and

xmnp is the phase of current applied. The rectangular cavity resonator is excited

at the centre with an antenna probe carrying current I(t) of some known frequency

xmnp. This generates the eld Ez inside the cavity of the form given below:

Z p2

jxlIdlx2 y2 jxtxc x2 y2 d

Ez x; y; d; t Gx; y 3=2 e I xejxt dx

4p x2 y2 d2

Equating resonator eld with the antenna current elds at z d plane;

Antenna or resonator radiation current or elds

r

X 2 ppd

Cmnp sin cosxmnpt /mnp

p

d d

Z

jxlIdlx2 y2

Gx; y 3=2 I x

4p x2 y2 d2

p2

e dx e jxt c x y d wmnp umn x; ydxdy;

x 2 2

jxt

Multiply both sides by ejxmnpt and then taking time averaging (KAM) gives us

the following

r ZT

2 ppd jmnp 1 jxlIdlx2 y2 jxmnpt

Cmnp sin e lim Gx; y 3=2 e

d d T!/ 2p

4p x2 y2 d2

T

p2

I xejxt dx e jxt c x y d wmnp umn x; ydt

x 2 2

It is clear that for these two expressions to be equal, the probe current can be

dened as

1X h i

Ix jI mnpj dx xmnpejmnp ejmnp dfx xmnpg

2 mnp

The antenna probe current must contain only the resonator characteristics

frequencies xmnp, then

300 Annexure-2

r

X 2 ppd

Cmnp sin cosxmnpt /mnp

p

d d

Z

jxlIdlx2 y2 A2:3

Gx; y 3=2 I x

4p x2 y2 d2

p2

ejxt dxe jxt c x y d wmnp umn x; ydxdy

x 2 2

Antenna probe current = Resonator radiated current or magnetic elds, as per the

law of conservation of energy. The modes diagrams are given below (Figs. A2.4,

A2.5, A2.6, A2.7, A2.8, A2.9, A2.10, A2.11, A2.12, A2.13, A2.14 and A2.15):

Annexure-2 301

302 Annexure-2

Mode sketch

Rectangular design

2 1

xmn pp

m n2 2

le a2 b2 , in two-dimensional case

kx2 x 0; ky2 y 0; kz2 z 0

@2x @2y @2z

hkx x is the harmonic function and can be written as follows: sinkx x or coskx x:

These are solution of wave function and if boundary conditions are applied, then

eigenvalues can be dened as follows:

2pf0 q

k0 ; ky tanky d=2 er 1k02 ky2

c

c q

kx2 ky2 kz2 er k02 ; Resonant frequency f0 p kx2 ky2 kz2

2p er

kz d q

where kx m pa ; ky n pb ; and kz tag 2 er 1k02 kz2 :

Annexure-2 303

The resonance frequency of this antenna can be estimated using the approximate

analytical expressions for the resonance frequency of TE111 mode in the a rectan-

gular resonator (three dimensional) given by

r

c p2 p 2 p2

f111 p ;

2p er a 2b d

Three-dimensional case

1

a0 X 2np 2np

f x an cos x bn sin x

2 n1

a a

Za

2 2np

an f x cos x dx

a a

0

Za

2 2np

bn f x sin x dx

a a

0

Half-wave Fourier analysis will have odd or even terms, i.e., sinesine or

cosinecosine.

If f(x) = f( x), then even harmonics will take place and only cosine terms will

occur, i.e.,

X

1 pnx

f x Cn cos

n1

a

Z np

2 a

where Cn f x cos x dx

a 0 a

Similarly for odd terms, f x 6 f x;

X

1 pnx

f x Bn sin

n1

a

Z a np

2

where Bn f x sin x dx:

a 0 a

304 Annexure-2

Every wave can be subjected to the process of spectral resolution, i.e., can be

represented as a superposition of monochromatic waves of various frequencies. The

character of this expansion varies according to the character of the time dependence

of the elds.

One category consists of those cases where the expansion contains frequencies

forming a discrete sequence of values. The simplest case of this type arises in the

resolution of a purely periodic eld. This is the usual expansion in Fourier series. It

contains the frequencies which are integral multiples of the fundamental fre-

quency x0 2p=T; where T is the period of the eld. We therefore write it in the

form as follows:

X

1

f fnejx0 nt

n1

where f is any of the quantities describing the eld. The quantities fn are dened in

terms of the function f by the integrals

ZT=2

1

fn f tejnx0 tdt :

T

T=2

fn fn :

in more complicated cases, the expansion may contain integral multiples of several

different incommensurable fundamental frequencies. When the sum is squared and

averaged over the time, the product of terms with different frequencies is given zero

because they contain oscillating factors.

Only terms of the form fn fn jfn j2 remain. Thus, the average of the square of

the eld, i.e., the average intensity of the wave, is the sum of the intensities of its

monochromatic components.

P P

f 2 n 11 jfn j2 2 1 2

n1 jfn j , where it is assumed that the average of

the function f over a period is zero. Another category consists of elds which are

expandable in a Fourier integral containing a continuous distribution of different

frequencies. For this to be possible, the function f(t) must satisfy certain denite

conditions; usually we consider functions which vanish for t ! 1:

Similarly, fx fx ; let us express the total intensity of the wave, i.e., the

integrals of f 2 over all time, in terms of the intensity of the Fourier components.

Now, we have

Annexure-2 305

8 9 8 9

Z1 Z1 < Z1 = Z1 < Z1 = dx

dx

f 2 dt f fx ejxt dt fx fejxt dt

: 2p; : ; 2p

1 1 1 1 1

Z1

dx

fx fx ;

2p

1

or

Z1 Z1 Z1

2 dx dx

f dt

2

jfx j 2 jfx j2 :

2p 2p

1 1 0

R

1 1 jxt

f t 2p 1 fx e dx, where the

R 1Fourier components are given in terms of the

function f t by the integrals, fx 1 f tejxt dt:

expressed as follows:

y q sin / r sin h sin / z r cos h:

p

q x2 y2 r sin h:

p

1 x2 y2 q

h tan tan1

z z

coordinate systems are given by

Az Aq cos / A/ sin /

Ar sin h cos / Ah cos h cos / A/ sin /

Ay Aq sin / A/ cos /

Ar sin h sin / Ah cos h sin / A/ cos /

Az Ar cos h Ah sin h

Aq Az cos / Ay sin / Ar sin h Ah cos h

A/ Ax sin / Ay cos /

306 Annexure-2

Aq sin h Az cos h

Aq cos h Az sin h

and ur ; uh ; u/

uq qd/dz u/ dqdz uz qdqd/

ur r 2 sin hdhd/ uh r sin hdrd/ u/ rdrdh

Length L ux dx uy dy uz dz

uq dq u/ qd/ uz dz

A B A1 B1 A2 B2 A3 B3

r rv r2 v

r r A r r A r 2 A

Rerej r cosxt h

Imrej r sinxt h

Kronecker Tensor

na2

f 2pr

1

2 e

2r2where a is the mean and is the variance and vector multipli-

cation can be dened as:

u1 u 2 u 3

A B A1 A2 A3

B1 B2 B3

Annexure-2 307

Divergence r A;

curl r A

@ @ @

r ux uy u z

@x @y @z

@x @x @x

r x ux uy uz

@x @y @z

@Ax @Ay @Az

rA

@x @x @x

ux uy u z

@ @ @

r A @x @y @z

A A A

z y z

r2x

@x2 @y2 @z2

@x 1 @x @x

rx uq u/ uz

@q q @/ @z

1 @ 1 @A/ @Az

rA qAq

q @q q @/ @z

r A ur u/ uz qA/

q @/ @z @z @q q @q q @/

1 @ @x 1@ 2

@ 2

r2x q 2 x2 x2

q @q @q q @/ @z

308 Annexure-2

@x 1 @x 1 @x

r x ur u/ u/

@r r @/ r sin h @/

1 @ 2 1 @ 1 @A/

rA 2 r Ar Ah sin h

r @r r sin h @h r sin h @/

1 @ @Ah

r A ur Ah sin h

r sin h @h @/

1 1 @Ar d 1 @ @Ar

uh rA/ u rAh

r sin h @/ dr r @r @h

1 @ @x 1 @ @x 1 @2

r2x 2 r2 2 sin h 2 2 x2

r @r @r r sin h @h @h r sin h @/

R ux x uy y uz z

r 0 ux x 0 uy y 0 uz z 0

q

0

jr r j x x0 2 y y0 2 z z0 2

0

Ilejkjrr j

A

4pjr r 0 j

To emphasize that A is evaluated at the eld point (x, y, z) and Il is situated at the

source point x0 ; y0 ; z0 (Table A2.1),

0

Ilr 0 ejkjrr j

Ar A

4pjr r 0 j

Frequency Symbol Frequency in Hz

Tera T 1012

Giga G 109

Mega M 106

Kilo K 103

Hecto H 102

Deca Da 101

Deci d 101

Centi cm 102

Milli mm 103

Micro 106

Nano n 109

Pico p 1012

Femto f 1015

Annexure-3

Steps !

1. Export the model from HFSS and save in G drive or any le (without Path).

2. Now right click the ADS icon and click run as administration.

3. Then, click the yes button.

4. Then, click the cancel and go create new project.

5. Now on schematic will open, then go to layout button, then go to create update

layout.

6. Then go to le button and then go to import button, the layout model is

complete.

7. Now click on line which is connected to the patch on layout model then delete

it.

8. Then go to the view button then go to layer view then go to by name. Then go

to conductor 2 button, now then drag the feed or patch and date it.

9. Now go the each capacitor then click double and give it value according to the

formula.

Cv 26 f at v 0 Ci 1:298lf

Cf 0 f and add each capacitor by line by clicking on line icon.

10. Now go to S-parameter then click on termination which also given in Fig. A3.1.

11. Now go to the S.P (S-parameter) button and put on schematic window then

click the S-parameter which is on the schematic window and put frequency 1 to

3 by stepping 1 MHz frequency then ok.

12. Now go to simulate button and simulate it then after completing the button.

13. Now then go to EDS model, then go to substrate and create update then go to

open button put substrate (RT Duroid-5880) then put the thickness of the

substrate (1.524 mm) loss tangent (0.001) then go to apply and then go to ok.

14. Now again go to EDS model then go to component. Now go to create update

then put start frequency and stop frequency 3 GHz.

15. Now put the port on the patch by single clicking on the patch from port Ze on.

16. Now minimize it.

Springer India 2016 309

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3

310 Annexure-3

button and connected to the

termination

17. Now go to schematic window and then go to library le and click on anywhere

on schematic window.

18. Now go to lumped element and select on capacitor and put three by pressing

control button.

PCB manufacturing from HFSS model

1. Save HFSS model bottom as view .dxf le after going to modeler and

exporting it

2. open .dfx in AutoCAD to generate .pdf or image as .jpg format.

3. use butter paper to place this design on to PCB

4. now connect SMA connectors and it is ready for testing antenna parameters.

II. HFSS design steps:

APPLY MAGNETIC AND ELECTRIC BIAS TO MHD ANTENNA

MAGNETIC BIASING STEPS WITH HFSS:

1. MAKE THREE SLOTS

2. SLOTS SHOULD BE ENCLOSING MICRO-STRIP FEED LINE

3. THE UPPER EDGE OF ALL THE SLOTS SHOULD TOUCH EACH

OTHER

4. THE SUBS AND SLOTS SHOULD NOT INTERSECT

5. UNITE ALL THE SLOTS

6. SELECT MATERIAL

7. (A) FERRITE

8. (B) MAGNETIC SATURATION EG 500 TESLA

9. GO TO BOX ASSIGN EXCITATION MAGNETIC BIAS

10. NEXT

11. PERMEABILITY

12. X, Y, Z VALUE-DESIRED

13. FINISH

14. CHECK FOR VALIDATION

Annexure-3 311

16. SIMULATE

1. INSERT TWO BOXES OF COPPER INSIDE THE DRA OVER THE

SLOT

312 Annexure-3

ENTER VOLTAGE AND E FIELD DIRECTION

ELECTRODE

Annexure-4

Helmholtz equation

r2 w k2 w lj medium with source

Maxwells equations

@H

r E l

@t

@E

rH j

@t

i j k

@ @Ez @Ey @Ez @Ex @Ey @Ex

r E @x @

@y

@

@z i j k

Ex Ey Ez @y @z @x @z @x @y

i j k

@ @Hz @Hy @Hz @Hx @Hy @Hx

r H @x @

@y

@

@z i j k

Hx Hy Hz @y @z @x @z @x @y

Comparing with RHS in both equations and getting value of Hx, Hy, Hz and Ex,

Ey, Ez, we get

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3

314 Annexure-4

in rectangular resonators

1 @Ez @Ey

Hx A4:1

jxl @y @z

1 @Ez @Ex

Hy A4:2

jxl @x @z

1 @Ey @Ex

Hz A4:3

jxl @x @y

1 @Hz @Hy

Ex A4:4

jxe @y @z

1 @Hz @Hx

Ey A4:6

jxe @x @z

1 @Hy @Hx

Ez A4:7

jxe @x @y

@

Substituting: @z c;

jxe @E @Hz

@y c @x

z

jxl @H @Ez

@y c @x

z

Hx Ex

c2 x2 le c2 x2 le

jx @E @Hz

@x c @x

z jxl @H @Ez

@x c @y

z

Hy Ey

c2 x2 le c2 x2 le

d2 H z d2 H z

c2 x2 le Hz 2

dx2 dy

d2 Ez d2 Ez

c2 x2 le Ez 2

dx2 dy

Now, rewriting Helmholtz equation for source-free medium

Annexure-4 315

r2 w k 2 w 0

W X xY yZ z

2

1 dX 1 d2 Y 1 d2 Z

k2 0

X dx2 Y dy2 Z dz2

1 d2 X

kx2

X dx2

1 d2 Y

ky2

Y dy2

1 d2 Z

kz2

Z dz2

W A sin kx x B cos kx x C sin ky y D cos ky y ejkz z

mode

Xn mpx npyo

Hz Cmn cos cos ejkz z ; Cmn Fourier Coefficients A4:8

mn

a b

TM mode

Xn mpx npyo

Ez Dmn sin sin ejkz z ; Dmn Fourier Coefficients A4:9

mn

a b

constant at any instant.

Let c jkz and m, n are integers and a, b are dimensions;

mp2 np2

kc mn ; cut off frequency

a b

mp 2 np 2

kz2 x2 le

a b

316 Annexure-4

mp 2 np 2

x le

2

[0

a b

s

1 mp2 np2

xc p

le a b

propagate.

Now, rewriting Hz and Ez

Xn mpx npyo

Hz Cmn cos cos ejkz z

mn

a b

Xn mpx npyo

Ez Dmn sin sin ejkz z

mn

a b

Eixx;y Incident EM wave in x-direction;

Eiyx;y Incident EM wave in y-direction;

" #

X jxlDmn np cmn Cmn mp mpx npy

Eixx;y b a

cos sin expcmn z;

h2m;n a b

Similarly,

" #

X jxlDmn mp cmn Cmn np mpx npy

Eiyx;y a b

cos sin expcmn z;

h2m;n a b

On simplication

np mp

jxlDmn cmn Cmn

Eixm;n b a

h2m;n

Annexure-4 317

Similarly

mp np

jxlDmn cmn Cmn

Eiym;n a b

h2m;n

" mp cm;n np jxl

#

c np jxl mp ;

Eiym;n m;n

b h2m;n a h2m;n

Dmn

we can now get the value of Cmn , Dmn after substitution of Eixm;n ; Eiym;n values.

2 2

Where h2m;n ma nb and

q

cmn h2m;n x2 le

modes.

Hence, we get solution of possible amplitudes and phase of wave propagating

through rectangular waveguide called as modes of propagation.

Half-wave Fourier expansion in waveguide is given as follows:

Za mpx mpx Zb

npy npy

fmn cos cos dx sin sin dy;

a a b b

0 0

even or odd terms, i.e., f(x) = f(x) for even term (all cosine terms) or even modes.

Where m, m and n, n 1

Za Zb mpx npy

2

Eixm;n Eixx;y cos cos dxdy

ab a b

0 0

Za Zb mpx npy

2

Eiym;n Eiyx;y sin sin dxdy

ab a b

0 0

Half-wave Fourier analysis will have odd or even terms, i.e., sinesine or

cosinecosine.

If f(x) = f(x), even harmonics will take place and only cosine terms will occur,

i.e.,

X

1 pnx

f x Cn cos

n1

a

318 Annexure-4

where

Za np

2

Cn f x cos x dx

a a

0

X

1 pnx

f x Bn sin

n1

a

where

Za np

2

Bn f x sin x dx

a a

0

then H elds. Now shape and size of resonator is given, wave equation shall give

solution of characteristic frequencies xmnp called eigenvalues or eigenfrequen-

cies of e-m oscillations of cavity resonator.

Lowest eigenfrequency x1 is cl ; where l is the dimension of resonator.

2

Higher frequency x cl ; then x is 2pvx2 c3 :

Hence, it depends on volume and net on shape of resonator.

X pmx pny prz

For resonator: fmnr sin sin sin f x; y; z

mnr

a b d

@ 2

@ @ 2 2

wx; y; z k 2 wx; y; z f x; y; z;

@x2 @y2 @z2

Helmholtz equation

X pmx pny prz

wx; y; z Cmnr sin sin sin

mnr

a b d

m2 h2 r 2

k 2 p2 C mnr fmnr

a2 b2 d 2

fmnr

Amplitude coefficient; C mnr 2 2 2

.

k2

m

a2 hb2 dr 2

xmnr

Annexure-4 319

p

k x l

Hence,

fmnr

C mnr

xmnp d2 xmnr2

fmnr

dxmnp xmnr

xt Aejxt

x20 x2 A B

Hence,

B

A

x20 x2

jxt

xt xBe

2 x2 , if x0 x; then x(t) will be 1

0

Now, x x0 d when d is small deviation

Bejxt

x0 xx0 x

Bejxt

d2x0

320 Annexure-4

@2 1 @2

ux; t 0; at boundaries

@x2 c2 dt2

@ 2 x2

u^x; x 0

@x2 c2

xx xx

A sin B cos 0

c c

^u0; x 0

^uL; x 0

c 0; Hence k L np; sine values to be zero.

sin xL

x kc npc

L , when n = 1, 2, 3 where k = /c;

when 2L, it is fundamental frequency x1

when L, the frequency is 2x1

when 2L/3, the frequency 3x1 ;

which can be generalized as:

X npx

C nsin

n

L

resonator

y=b

x=a

Annexure-4 321

2 0

@x2 @y2 c @t2

w0; y; t wa; y; t 0

wx; 0; t wx; b; t 0

@2w @ @w @ @w

rdxdy 2 Tdy dx Tdx dy

@t @x @x @y @y

Y 00 X 00

kY2 ; kX2 ;

Y X

@2w

c2 r 2 w 0

@t2

wx; y; t X xY yT t

00

T 00 t X x Y 00 y

x2 c2 A4:10

T t X x Y y

Let

X x sinkx x

Y y sin ky y

322 Annexure-4

x2

kx2 ky2

c2

np mp

kx ; ky

b a

q

2 2

Equation (A4.1) can be written as xmn cp ma nb :

From Fourier series analysis

X

1 mpx npyh i

xm; n sin sin Cm; nejxm;nt Dm; nejxm;nt

mn1

a b

A4:12

At t 0; wx; y; 0 w0 x; y

On differentiating equation w0 x; y, we get w1 x; y; 0 w1 x; y:

When t 6 0;

X

1 mpx npy

w0 x; y Cm; n Dm; n sin sin A4:13

mn1

a b

X

1 mpx npy

w1 x; y jxm; nC m; n Dm; n sin sin A4:14

mn1

a b

Za Zb mpx npy

2

p w0 x; y sin sin dxdy C m; n Dm; n

A4:15

ab a b

0 0

Similarly,

Z mpx npy

2

jxm; n p w1 x; y sin sin dxdy C m; n Dm; n

ab a b

A4:16

Hence, obtain the value of Cm; n; Dm; n from Eqs. (A4.3) and (A4.4)

Annexure-4 323

ZZ mpx npy

1

Cm; n; Dm; n

p w0 x; y sin sin dxdy

ab a b

Z mpx npy

A4:17

1

w; x; y sin sin dxdy

jxm; n a b

m px n py

0 0

w0 x; y A sin sin

a b

m px n py

0 0

w1 x; y B sin sin

a b

Solving equation (A4.17)

Cm; n; Dm; n dm m0

dn n0

A a b 1 B a b

p p

ab 2 2 ab jxm0 ; n0 2 2

p

p A j B ab

C m; n; Dm; n ab A jBdm m0

dn n0

4 4 4

p m px n py

ab

ejxm0 ;n0 t

0 0

wx; y; t ReA jB sin sin

2 a b

p m px n py

ab 0 0

wx; y; t A cosxm0 n0 t B sinxm0 n0 t sin sin

2 a b

A4:18

to excitation T(x, y)

324 Annexure-4

Alternate method

m 2b sin h;

n 2a cos h;

Dividing both sides of above equations by 2a and 2b and adding them gives us

1 n2 m2

2 2; where k2 kx2 ky2

k2 4a 4b

Half-wave Fourier analysis:

1

a0 X 2np 2np

f x an cos x bn sin x

2 n1

a a

Z a

2 2np

an f x cos x dx

a a

0

Za

2 2np

bn f x sin x dx

a a

0

Half-wave Fourier analysis will have odd or even terms, i.e., sinesine or

cosinecosine.

If f(x) = f(x), even harmonics will take place and only cosine terms will occur,

i.e.,

X

1 pnx

f x Cn cos

n1

a

where

Za np

2

Cn f x cos x dx

a a

0

X

1 pnx

f x Bn sin

n1

a

Annexure-4 325

where

Za np

2

Bn f x sin x dx

a a

0

Every wave can be subjected to the process of spectral resolution, i.e., can be

represented as a superposition of monochromatic waves of various frequencies. The

character of this expansion varies according to the character of the time dependence

of the elds.

One category consists of those cases, where the expansion contains frequencies

forming a discrete sequence of values. The simplest case of this type arises in the

resolution of a purely periodic eld. This is the usual expansion in Fourier series. It

contains the frequencies which are integral multiples of the fundamental fre-

quency x0 2p=T , where T is the period of the eld. We therefore write it in the

form as follows:

X

1

f fnejx0 nt

n1

(where f is any of the quantities describing the eld). The quantities fn are dened in

terms of the function f by the integrals

ZT=2

1

fn f tejnx0 tdt:

T

T=2

fn fn :

in more complicated cases, the expansion may contain integral multiples of several

different incommensurable fundamental frequencies. When the sum is squared and

averaged over the time, the product of terms with different frequencies is given zero

because they contain oscillating factors.

Only terms of the form fn fn jfn j2 remain. Thus, the average of the square of

the eld, i.e., the average intensity of the wave, is the sum of the intensities of its

P P1

monochromatic components. f 2 1 2

n1 jfn j 2

2

n1 jfn j ; where it is

assumed that the average of the function f over a period is zero. Another category

consists of elds which are expandable in a Fourier integral containing a continuous

distribution of different frequencies. For this to be possible, the function f(t) must

satisfy certain denite conditions; usually we consider functions which vanish for

t ! 1.

326 Annexure-4

Similarly, fx fx ; Let us express the total intensity of the wave, i.e., the

integrals of f 2 over all time, in terms of the intensity of the Fourier components.

Now, we have

8 9 8 9

Z1 Z1 < Z1 = Z1 < Z1 = dx

dx

f 2 dt f fx ejxt dt fx fe jxt dt

: 2p; : ; 2p

1 1 1 1 1

Z1

dx

fx fx ;

2p

1

or

Z1 Z1 Z1

2 dx dx

f dt

2

jfx j 2 jfx j2 :

2p 2p

1 1 0

R

1 1 jxt

f t 2p 1 fx e dx; where the Fourier components are given in terms of the

function R1

f(t) by the integrals, fx 1 f tejxt dt:

Let x(t) is the input signal, i.e., voltage signal. As per Parsevals power theorem,

energy associated with this signal be

Z1

E jxtj2 dt; in time domain

1

Z1

1

jXxj2 dx; in frequency domain

2p

1

Antenna having radiation resistance Rr shall be

Z1 Z1

1 2 1

E jxtj dt jXxj2 dx

Rr 2pRr

1 1

Z1 Z1

2 R

E Rr jxtj dt jX xj2 dx

2p

1 1

Annexure-4 327

ESD energy spectral density; energy spread per unit volume across 1 resister

ESD jX xj2

Discrete Fourier transform (DFT) in time domain into frequency domain spectral

analysis

X

N 1

j2pnk

x k xne ; k 0; 1; . . .; N 1:

n0

N

1XN 1

j2pkn

X n X ke ; n 0; 1; 2; . . .; N 1:

N k0 N

DFT has nite length N, period N

wh; / k !rn r0 n 1kd sin h

X

N

E sin h ejn1kd sin h

n1

Annexure-5

In this annexure, resonant modes TE and TM have been generated inside RDRA

whose dimensions are a, b, and d. Two parallel plates are attached along with

dielectric slab in between these plates to RDRA. This slab forms non-resonant part

and RDRA is main resonant. This is shown in Fig. A5.1a, b. The resonant modes

dominant and higher-order modes are being generated by maintaining appropriate

aspect ratio of RDRA. Then, the non-resonant slab inductance and capacitance is

introduced into main RDRA. This lumped value of inductance and capacitance is

seen in the resonant frequency.

(a) The increase in the length of internal strip introduce shift in the higher

resonant modes frequency, as they shift toward lower side and vice versa.

Hence, resonant frequency is reduced.

(b) On the other side, increase in the length of external strip introduces shift in

the lower resonant modes frequency shifts toward higher side and vice

versa. Hence, frequency is increased with strip length.

(c) Increase in spacing between parallel plates introduces the combined effect

of internal as well as external strip length variation, i.e., higher- and

lower-order resonant modes shift toward the centre frequency which can be

seen as mode-merging effect.

(d) Finally, the effect of placing a lumped varactor diode between parallel plates is

seen. The increase in the capacitance value of lumped varactor diode causes

shift in the higher resonant frequency toward lower resonant frequency

side.

These results have been investigated using HFSS and they shown with S11

results along with each RDRA model. By varying length, a, width b, and

height d of RDRA modes are generated. The internal strip, external strip, and

dielectric slab and dielectric constant provided several degrees of freedom in the

RDRA design. This has extended the control on the amount of coupling, hence

resonant frequency. This shall have large impact on resonant modes, compactness

of antenna, radiation pattern, and polarization.

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3

330 Annexure-5

parallel standing strips having

rectangular non-resonant slab

in between. b RDRA with

lumped varactor diode

between strips

on Resonant Modes

Inside the RDRA

Annexure-5 331

(a)

-2.50 14.9500 -17.6485

m1 Curve Info

m2 17.1000 -17.6193 dB(S(waveport,waveport))

m3 18.0500 -19.9365 Setup6 : Sweep

dB(S(waveport,waveport))

-5.00

-7.50

-10.00

-12.50

-15.00

m1 m2

-17.50

m3

-20.00

10.00 12.00 14.00 16.00 18.00 20.00

Freq [GHz]

(b)

0.00 16.0500 -12.9889

m1 Curve Info

m2 19.1500 -17.4607 dB(S(waveport,waveport))

dB(S(waveport,waveport))

Setup6 : Sweep

-2.50

-5.00

-7.50

-10.00

m1

-12.50

-15.00

m2

-17.50

10.00 12.00 14.00 16.00 18.00 20.00

Freq [GHz]

Fig. A5.2 a Higher-order modes generated in RDRA with square base. b Higher-order modes

generated in the rectangular base RDRA

332 Annexure-5

Structure x (mm) y (mm) z (mm) er

DRA 4.6 9 10.8 9.8

substrate 20 30 0.8 2.2

Micro-strip 2.4 (width)

Rect. SLAB 1 9 10.8 1

External strip 2.4 10.5

Internal strip 2.4 3

Modes Inside RDRA

The effect of the internal strip length is seen on resonance frequency and resonant

modes of RDRA.

The reflection coefcient plot can be seen for the possible changes as given in

Fig. A5.4a.

The effective electrically length of RDRA is changed by introducing change in

length of internal strip as given below.

Changing the effective dimension of the dielectric resonator changes the reso-

nant frequency.

Annexure-5 333

Fig. A5.4 a External strip (xed) = 10.5 mm and variation in internal strip from (2 mm).

b External strip (xed) = 10.5 mm and variation in internal strip from (2.5 mm). c External strip

(xed) = 10.5 mm and variation in internal strip from (3 mm)

334 Annexure-5

Modes Inside RDRA

The effect of the external strip length on resonance frequency and resonant modes is

shown in Fig. A5.5b. Internal strip (xed) = 3 mm and variation in external strip

from 10.5, 7, 0 mm is investigated. Contrary to the previous case, the third reso-

nance stays mainly xed at the same frequency, while the rst and second resonant

frequencies are considerably decreased with increasing external strip length.

Standing Strips and er

The effect of the spacing between parallel plates and permittivity of the rectangular

slab between parallel plates is seen on resonance frequency and modes (Figs. A5.6

and A5.7).

The separation width variation ranges as 0.5, 1.5, and 2.5 mm [er 1, external

strip = 10.5 mm and inner strip = 3.5 mm (xed)].

we will change the variable separation Width (0.5, 1.5, 2.5) for er 2 keeping

external strip = 10.5 mm, inner strip = 3.5 mm constant (Fig. A5.8).

Annexure-5 335

Fig. A5.5 a Internal strip (xed) = 3 mm and variation in external strip from (0 mm). b Internal

strip (xed) = 3 mm and variation in external strip from (7 mm). c Internal strip (xed) = 3 mm and

variation in external strip from (10.5 mm)

336 Annexure-5

rated plates

Annexure-5 337

m1 10.2000 -17.1071 Curve Info

m2 14.6000 -15.6812 dB(S(waveport, waveport))

m3 16.8500 -16.0264 Setup6 : Sweep

m4 17.6500 -16.4270

m5 19.5000 -16.2692

dB(S(waveport,waveport))

-7.50

-10.00

-12.50

-15.00 m2

m3 m5

m4

m1

-17.50

10.00 12.00 14.00 16.00 18.00 20.00

Freq [GHz]

m1 10.4500 -17.9665 Curve Info

m2 14.3500 -16.3577 dB(S(w aveport,w aveport))

m3 16.8000 -16.7685 Setup6 : Sw eep

m4 17.5000 -16.4319

m5 19.5000 -21.9771

dB(S(waveport,waveport))

-7.50

-10.00

-12.50

-15.00

m2 m4

m3

m1

-17.50

-20.00

m5

-22.50

10.00 12.00 14.00 16.00 18.00 20.00

Freq [GHz]

m1 14.6000 -16.6991 Curve Info

m2 16.9500 -16.3861 dB(S(waveport, waveport))

m3 17.9000 -20.2745 Setup6 : Sweep

-5.00

dB(S(waveport,waveport))

-7.50

-10.00

-12.50

-15.00 m2

m1

-17.50

m3

-20.00

-22.50

10.00 12.00 14.00 16.00 18.00 20.00

Freq [GHz]

Fig. A5.7 a Separation width (0.5 mm) for er 1, external strip = 10.5 mm, inner strip = 3.5 mm.

b Separation width (1.5 mm) for er 1, external strip = 10.5 mm, inner strip = 3.5 mm.

c Separation width (2.5 mm) for er 1, external strip = 10.5 mm, inner strip = 3.5 mm

338 Annexure-5

(a) Name X Y

5.00 14.0000 -13.9564

m1

XY Plot 4 HFSSDesign1

ANSOFT

Curve Info

m2 16.9000 -22.9742 dB (S(waveport,waveport))

m3 19.3500 -20.7517 Setup6 : Sweep

-0.00

dB (S(waveport,waveport))

-5.00

-10.00

m1

-15.00

-20.00 m3

m2

-25.00

10.00 12.00 14.00 16.00 18.00 20.00

Freq [GHz]

(b) Name X Y

-4.00 14.3500 -15.3507

m1

XY Plot 4 HFSSDesign1 ANSOFT

Curve Info

m2 17.7000 -15.9854 dB (S(waveport,waveport))

m3 19.5000 -16.1834 Setup6 : Sweep

-6.00

dB (S(waveport,waveport))

-8.00

-10.00

-12.00

-14.00

m1

m2 m3

-16.00

-18.00

10.00 12.00 14.00 16.00 18.00 20.00

Freq [GHz]

(c) Name X Y

-2.00 14.4500 -13.4533

XY Plot 4 HFSSDesign1 ANSOFT

m1 Curve Info

m2 16.7500 -12.0819 dB (S(waveport,waveport))

m3 Setup6 : Sweep

-4.00 17.1500 -13.6281

m4 17.6500 -14.9946

m5 19.5500 -16.0278

-6.00

dB (S(waveport,waveport))

-8.00

-10.00

m2

-12.00

m1 m3

-14.00

m4

m5

-16.00

-18.00

10.00 12.00 14.00 16.00 18.00 20.00

Freq [GHz]

Fig. A5.8 a Separation width (0.5) for er 2, external strip = 10.5 mm, inner strip = 3.5 mm.

b Separation width (1.5) for er 2, external strip = 10.5 mm, inner strip = 3.5 mm. c Separation

width (2.5) for er 2, external strip = 10.5 mm, inner strip = 3.5 mm

Annexure-5 339

in Between the Plate

The effect of the varactor diode capacitance placed in between the parallel standing

strips is seen. The resonant modes get shifted lower side (Fig. A5.9).

The separation width = 1.0, er 1, external strip = 10.6 mm, inner

strip = 3.0 mm, varactor diode (variation from 1 to 5 F with step of 1 F) at

position (z = 2.3) in vertical direction. The resulting effect is shown in Fig. A5.10.

HFSS steps_Project1

1. Open HFSS.

2. Create le name project1.

3. Dene in the Cartesian co-ordinate system origin as (x = 0, y = 0, z = 0).

4. Choose 3-D rectangular box for substrate by dening the desired substrate

material and its dimensions such as (RT Duroid and x = 20 mm, y = 30 mm,

z = 0.8 mm).

5. Create DRA structure with desired material and dimensions on the substrate top

surface (e.g., If substrate dimension from origin was 0.8 mm in z-direction.

Then choose DRA #d dimension keeping substrate dimension as reference).

340 Annexure-5

0.00 10.9619 -26.4758

m1 Curve Info

m2 14.7295 -21.6984 dB(S(waveport,waveport))

m3 17.4549 -26.3304 Setup1 : Sweep

c='0pF'

-5.00 dB(S(waveport,waveport))

Setup1 : Sweep

dB (S(waveport,waveport))

c='1pF'

dB(S(waveport,waveport))

-10.00 Setup1 : Sweep

c='2pF'

dB(S(waveport,waveport))

Setup1 : Sweep

-15.00 c='3pF'

dB(S(waveport,waveport))

Setup1 : Sweep

c='4pF'

-20.00 dB(S(waveport,waveport))

m2 Setup1 : Sweep

c='5pF'

-25.00 m1 m3

-30.00

10.00 12.00 14.00 16.00 18.00 20.00

Freq [GHz]

6. Create two parallel strips adjacent to DRA above the substrate surface with

rectangular slab in between them keeping substrate dimension as reference.

7. Apply micro-strip feeding to the DRA structure by dening the micro-strip port

with appropriate length and width for impedance matching (e.g., wave port)

assigning in the desired direction of input excitation.

8. Variation in height of external strip keeping the internal strip height xed and

vice versa.

9. Effect of the permittivity of rectangular slab can be seen by varying the material

property and thickness of the slab in between two xed parallel plates.

10. Placing a lumped capacitor between two parallel standing strips with desired

value (e.g., 2 F) and perform parametric analysis for variable capacitance

value of lumped element.

11. Performing the simulation for the steps 8, 9, 10 mentioned above separately and

for mode analysis of DRA which give modal frequency response and effect of

the variation of radiation parameters associated with DRA and non-resonant

slab with parallel standing strip geometry.

12. Analysis of the simulated structure can be performed by taking various

response quantities such as S11, radiation pattern, gain, and eld distribution.

13. The above mechanism can also be validated in RDRA by VNA with anechoic

chamber on prototype model after structure is simulated.

Annexure-6

There are three different coordinate systems, i.e., Cartesian, cylindrical, and

spherical systems. Cartesian are (x, y, z), cylindrical are q; /; z, and spherical are

r; h; / representation (Figs. A6.1 and A6.2).

p

X q cos / q x2 y2

Y q sin y

tan1

Zz x

Zz

(c) Spherical to cartesian (d) Cartesian to cylindrical

p

X r sin h cos / r x2 y2 z2

Y r sin h sin / p!

x2 y2

Z r cos h h tan1

z

y

/ tan1

x

(e) Cylindrical to spherical (f) Spherical to cylindrical

p

r q2 z2 q r sin h

q //

h tan1

z z r cos h

//

The Cartesian r (Del) is given as follows:

~ ~ @ @ @

r x ~ y ~z

@x @y @z

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3

342 Annexure-6

components

~ q @ ^ 1 @ ^z @

r ^ /

@q q @/ @z

@ @ @q @ @/ @ @z

^x 6 6 ^x

@x @q @x @/ @x @z @x

@q

@ p 1 2x x

x2 y2 p p cos /

@x @x z x2 y2 x y2

2

p

) q x2 y2

p

q x2 y2

Hence,

x

cos /

q

@/ @ h 1 y 1 x0 y1 x2 y

tan 2

@x @x x 2

1 2

y x 2 x2 y2 x

x

@/ 1 y 1

p p sin /

@x x2 y2 x2 y2 q

@/

0 ) z is the same z as in Cartesian system it doesn't depend on x

@x

Annexure-6 343

Thus, we have

@ @ @P @ @/ @ @z

^x ^x

@x @p @x @/ @x @z @x

A6:1

@ @ 1

^x cos / sin / 0

@q @/ p

@ @ @q @ @/ @ @z

^y ^y

@y @q @y @/ @y @z @y

@q @ p 2y y

x2 y2 p p sin /

@y @y 2 x y2 2 x y2

2

@/ @ 1 1 1 x 1

tan yx 2 cos /

@y @y y2 x

1 2 x y 2 q

x

@z

0

@y

Thus,

y @ @q @ @/ @ @z

^y ^y

@y @q @y @/ @y @z @y

A6:2

@ 1 @

^y sin / cos / 0

@q q @/

@ @

A6:3

@z @z

~ ^x @ ^y @ ^z @

r

@x @y @z

~ ^x cos / @ 1 sin / @ ^y sin / @ 1 cos / @ ^z @

r

@q q @/ @q q @/ @z

Cylindrical

r A6:4

@q q @/ @z

344 Annexure-6

^p ^x cos / ^y sin / ^p

^ ^x sin / ^y cos / /

/ ^

^z ^z ^z

~ q @ 1^ @ @

r ^ / ^z

@q q @/ @z

2. DEL r expression as spherical system (Figs. A6.3, A6.4 and A6.5):

p

X r sin h cos / r x2 y2 z2

Y r sin h sin / p!

1 x2 y2

Z r cos h h tan

z

y

/ tan1

x

~ @ ^x @ ^y @ ^z

r A6:5

@x @y @z

@ @ @r @ @h @ @/

^x ^x A6:6

@x @r @x @h @x @/ @x

Annexure-6 345

components

subcomponents

@ @ @r @ @h @ @/

^y y A6:7

@y @r @y @h @y @/ @y

@ @ @r @ @h @ @/

^z z A6:8

@z @r @z @h @z @/ @z

346 Annexure-6

@r @ p

x2 y2 z 2

@x @x

2x

p

2 x y2 z 2

2

x A6:9

p

x2 y2 z 2

r sin h cos /

r

sin h cos /

@r @ p

x2 y2 z 2

@y @y

2y

p

2 x2 y2 z 2

y

p A6:10

x y2 z 2

2

r sin h sin /

r

sin h sin /

@r @ p

x2 y2 z 2

@z @z

2z

p

2 x y2 z 2

2

z A6:11

p

x y2 z 2

2

r cos h

r

cos h

Annexure-6 347

p!

@h @ x2 y2

tan1

@x @x z

1 1 2x

x2 y2

p

1 z 2 x2 y2

z2

2

z x

p

x y z x y2

z2 2 2 2 A6:12

x

p

x2 y2 2

z x y2 z 2

r sin h cos /

r 2 tan h

cos h cos /

r

p!

@h @ x2 y2

tan1

@y @y z

1 1 2y

x2 y2

p

1 z 2 x2 y2

z2

2

z y

p

z 2 x2 y2 z x2 y2 A6:13

y

p

x y 2

2 2

z x y2 z 2

r sin h sin /

r 2 tan h

cos h sin /

r

348 Annexure-6

p!

@h @ x2 y2

tan1

@z @z z

p

1 x2 y2

1 x zy

2 2

z2

2

A6:14

p!

z x2 y2

2

x y z

2 2 z

sin h

r

@/ @ y

tan1

@x @x x

1 x0 y1

2

1 yx2 x2

x2 h y i A6:15

2 2

x y2 x

y

2

x y2

sin /

r sin h

@/ @ y

tan1

@y @y x

1 1

y2 x

1 x2 A6:16

x

2

x y2

cos /

r sin h

Annexure-6 349

@/ @ y

tan1 0 A6:17

@z @z x

Put Eqs. (A6.5), (A6.8), (A6.11) in Eq. (A6.2), Put Eqs. (A6.6), (A6.9), (A6.12)

in Eq. (A6.3) and Put Eqs. (A6.7), (A6.10), (A6.13) in Eq. (A6.4).

^x ^x sin h cos / A6:18

@x @r @h r r sin h @/

^y ^y sin h sin / A6:19

@y @r @h r r sin h @/

@ @ @ sin h

^z ^z cos h A6:20

@z @r @h r

And by using original denition to Spherical unit vector,

^h ^x cos h cos / ^y cos h sin / ^z sin h

^ ^x sin / ^y cos /

/

We get

~ ^r @ ^h 1 @ /

r ^ 1 @

@r r @h r sin h @/

350 Annexure-6

S. Permittivity Dimension (a(length) b Resonant Effective Multiple %

No. (width) d(depth)) (mm) frequency width (b) factor change

in width

1. 10.0 14.3 25.4 26.1 3.5 34.22 1.3474 34.7381

2. 10.0 14 8 8 5.5 14.13 1.7665 76.6535

3. 10.0 15.24 3.1 7.62 6.21 8.33 2.8872 168.7230

4. 20.0 10.2 10.2 7.89 4.635 15.31 1.5014 50.1419

5. 20.0 10.16 10.2 7.11 4.71 15.15 1.4858 48.5797

6. 35.0 18 18 6 2.532 24.12 1.34 33.9973

7. 35.0 18 18 9 2.45 25.64 1.4244 42.4423

8. 100.0 10 10 1 7.97 11.24 1.1242 12.4237

Hence r

3. E and H elds in RDRA

Fields converting into TE and TM modes inside rectangular DRA (Fig. A6.6).

4. Transcendental equation solution using MATLAB programs (simulated

rectangular DRA) (Fig. A6.7; Table A6.1).

Annexure-6 351

%%Dimensions of DRA

%%length

d=[14.3,14.0,15.24,10.2,10.16,18,18,10];

%%width

w=[25.4,8,3.1,10.2,10.2,18,18,10];

%%height

h=[26.1,8,7.62,7.89,7.11,6,9,1];

%%Mode

m=1;

n=1;

p=1;

c=3e8;

cons=[10.0,10.0,10,20,20,35,35,100];

syms y real

for i=drange(1:8)

kx(i)=pi/d(i);

kz(i)=pi/2/h(i);

ko=sqrt((kx(i).^2+y.^2+kz(i).^2)/cons(i));

f=real(y.*tan(y*w(i)/2)-sqrt((cons(i)-1)*ko.^2-y.^2));

ky(i)=fzero(inline(f),[0,(pi/w(i))-0.01]);

%%Resonant frequency

fre(i)=c/2/pi*sqrt((kx(i).^2+ky(i).^2+kz(i).^2)/cons(i))*1e3;

Effwidth(i)=pi/ky(i);

factor(i)=Effwidth(i)./w(i);

perchangwidth(i)=((Effwidth(i)-w(i))/w(i))*100;

end

352 Annexure-6

m=1;

n=1;

p=1;

E_r=10;

a=15.24e-03;

b=3.1e-03;

d=7.62e-03;

c=3e+08;

kx=m*pi/a;

ky=n*pi/b;

kz=p*(pi/d)/2;

ko=sqrt(kx^2+ky^2+kz^2)/sqrt(E_r);

fo=(c*ko/pi)/2;

foghz=fo/(1e+09);

Annexure-6 353

MATLAB programs taking parameters a,b,d same and comparing frequency using :

m=1

n=1

p=1

E_r=10

a=14.3e-03

b=25.4e-03

d=26.1e-03

c=3e+08

k_x=m*pi/a

k_y=n*pi/b

k_z=p*(pi/d)/2

k_o=sqrt(k_x^2+k_y^2+k_z^2)/sqrt(E_r)

f_o=(c*k_o/pi)/2

f_oGHz=f_o/1e+09

354 Annexure-6

m=1;

n=1;

p=1;

E_r=10;

a=14.3e-03;

b=25.4e-03;

d=26.1e-03;

c=3e+08;

syms y real

kx=pi/a;

kz=pi/d/2;

ko=sqrt(kx^2+y^2+kz^2)/sqrt(E_r);

f=real(y*tan(y*b/2)-sqrt((E_r-1)*ko^2-y^2));

ky=fzero(inline(f),[0,(pi/b)-0.01]);

fre=c/2/pi*sqrt((kx^2+ky^2+kz^2)/E_r)*1e3;

effwidth=pi/ky;

factor=effwidth/b;

perchangwidth=((effwidth-b)/b)*100;

Annexure-6 355

MATLAB programs taking parameters a,b,d same and comparing frequency using :

: Characteristic Equation

Where a=17mm

b=25mm

c=10mm

m=1;

n=1;

p=1;

E_r=10;

a=17e-03;

b=25e-03;

d=10e-03;

c=3e+08;

k_x=m*pi/a;

k_y=n*pi/b;

k_z=p*(pi/d)/2;

k_o=sqrt(k_x^2+k_y^2+k_z^2)/sqrt(E_r);

f_o=(c*k_o/pi)/2;

f_oGHz=f_o/1e+09;

356 Annexure-6

m=1;

n=1;

p=1;

E_r=10;

a=17e-03;

b=25e-03;

d=10e-03;

c=3e+08;

syms y real

kx=pi/a;

kz=pi/d/2;

ko=sqrt(kx^2+y^2+kz^2)/sqrt(E_r);

f=real(y*tan(y*b/2)-sqrt((E_r-1)*ko^2-y^2));

ky=fzero(inline(f),[0,(pi/b)-0.01]);

fre=c/2/pi*sqrt((kx^2+ky^2+kz^2)/E_r)*1e3;

effwidth=pi/ky;

factor=effwidth/b;

perchangwidth=((effwidth-b)/b)*100;

Annexure-6 357

MATLAB programs taking parameters a,b,d same and comparing frequency using :

Characteristic Equation

m=1

n=1

p=1

E_r=10

a=14.3e-03

b=25.4e-03

d=26.1e-03

c=3e+08

k_x=m*pi/a

k_y=n*pi/b

k_z=p*(pi/d)/2

k_o=sqrt(k_x^2+k_y^2+k_z^2)/sqrt(E_r)

f_o=(c*k_o/pi)/2

f_oGHz=f_o/1e+09

358 Annexure-6

m=1;

n=1;

p=1;

E_r=10;

a=14.3e-03;

b=25.4e-03;

d=26.1e-03;

c=3e+08;

syms y real

kx=pi/a;

kz=pi/d/2;

ko=sqrt(kx^2+y^2+kz^2)/sqrt(E_r);

f=real(y*tan(y*b/2)-sqrt((E_r-1)*ko^2-y^2));

ky=fzero(inline(f),[0,(pi/b)-0.01]);

fre=c/2/pi*sqrt((kx^2+ky^2+kz^2)/E_r)*1e3;

effwidth=pi/ky;

factor=effwidth/b;

perchangwidth=((effwidth-b)/b)*100;

Annexure-6 359

360 Annexure-6

Bibliography

1. Long SA, Mcallister MW, Shen LC (1983) The resonant cylindrical dielectric cavity antenna.

IEEE Trans Antennas Propag 31(3):406412

2. Kajfez D, Guillon P (eds) (1986) Dielectric resonators. Artech House, Dedham

3. Kranenburg RA, Long SA, Williams JT (1991) Coplanar wave guide excitation of resonator

antennas. IEEE Trans Antennas Propag 29(1):119122

4. Lee RQ, Simons RN (1994) Bandwidth of dielectric resonator antennas. In: IEEE antennas

propagate society of international symposium. Seattle, pp 15001503

5. Wu JY, Hung CY, Wong KL (1994) Low prole very high permittivity DRA excited by

coplanar waveguide. Microw Opt Technol Lett 22(2):9697

6. Yung EKN, Lee WWS, Luk KM (1994) A Dielectric resonator on a micro strip antennas. In:

IEEE antennas propagate society of international symposium serial. Washington, pp 1504

1507

7. Mongia RK, Luk KM, Yung EKN (1994) Low prole dielectric resonator antennas using a

very high permittivity material. Electron Lett 30(17):13621363

8. Oliver MB, Antar YMM, Mongia RK, Iittipiloon A (1995) Circularly polarized rectangular

dielectric resonator antenna. Electron Lett 31(6):418419

9. Pozar DM, Schubert DH (1995) Micro strip antennas: the analysis of micros trip antennas and

arrays. IEEE Press, New York

10. Keller MG, Roscone DJ, Oliver MB, Mongia RK, Antar YMM, Iittipiloon A (1995) Active

aperture coupled rectangular dielectric resonator antenna. IEEE Microw Guided Wave Lett 5

(11):376378

11. Esselle KP (1996) A low prole rectangular dielectric resonator antennas. Electron Lett 44

(9):12961297

12. Hwang Y, Zhang YP, Mulk K, Yung ENK (1997) Gain enhanced miniaturized rectangular

dielectric resonator antennas. Electron Lett 33(5):350352

13. Leung KW, To MW (1997) Slot coupled dielectric resonator antennas using a proximity feed

on a perpendicular substrate. Electron Lett 33(20):16651666

14. Hwang Y, Zhang YP, Luk KM, Yung EKN (1997) Gain-enhanced miniaturised rectangular

dielectric resonator antenna. IEEE Electron Lett 33(5):350352

15. Mongia RK, Ittipiboon A (1997) Theoretical and experimental investigation on rectangular

dielectric resonator antennas. IEEE Trans Antenna Propag 45:13481356

16. Mongia RK, Iittipiboon A (1997) Theoretical and experimental investigation on rectangular

dielectric resonator antennas. IEEE Trans Antennas Propag 45(9):13481356

17. Mongia RK, Ittipiboon A (1997) Theoretical and experimental investigations on rectangular

dielectric resonator antennas. IEEE Trans Antennas Propag 45(9):13481356

18. Junkar GP, Kishk AA, Glisson AW (1998) Multiport network description and radiation

characteristics of coupled dielectric resonator antennas. IEEE Trans Antennas Propag 46

(8):12421243

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3

362 Bibliography

19. Kranenburg RA, Long SA (1998) Micro strip transmission line excitation of dielectric

resonator antennas. Electron Lett 24(18):11561157

20. Wu JY, Hung CY, Wong KL (1999) Low prole very high permittivity DRA excited by a

coplanar wave guide. Microw Opt Technol Lett 22(2):9697

21. Hong CS (1999) Adjustable frequency dielectric resonator antennas. Proc Natl Sci Counc.

ROC (A) 23(6):736738

22. Henry B, Petosa A, Antar YMM, Morin GA (1999) Mutal coupling. Opt Technol Lett 21

(1):4448

23. Luk KM, Lee MT, Leung KW, Yung EVN (1999) Technique for improving coupling between

micro strip line and dielectric resonator antennas. Electron Lett 35(5):357358

24. Sangiovanni A, Garel PY, Dauvigenc JY, Pichor Ch (2000) Numerical analysis of dielectric

resonator antennas. Int J Numer Model Electr Netw Devices Fields 13:199215

25. Leung KW (2000) Conformal strip excitation of dielectric resonator antennas. IEEE Trans

Antennas Propag 48(6):961967

26. Ge Y, Esselle KP (2000) Computation of radiation pattern of a rectangular dielectric resonator

antenna using the method of moments. Microw Opt Technol Lett 27(6):382384

27. Hang CY, Kuo JS (2000) Frequency adjustable circularly polarized dielectric resonator

antennas. Microw Opt Technol Lett 34(3):211213

28. Sangiovanni A, Dauvignac JY, Pichod C (2001) Embedded dielectric resonator antenna for

bandwidth enhancement. Electron Lett 33(7):406408

29. Laisne A, Gillard R, Piton G (2001) Robust slot fed dielectric resonator antennas using an

intermediate. Electron Lett 37(25):14971498

30. Jerry Fiedziuszko S, Home S (2001) Dielectric resonatorrise your high SQ. IEEE Microw

Mag 5160

31. Deng SM, Chen TW, Kan HH (2001) A CPW fed rectangular dielectric resonator antennas. In:

Proceedings of IEEE APMC, Taipei, pp 954956

32. Al Salmeh MS, Antar YMM, Seguin G (2002) Coplanar wave guide fed coupled rectangular

dielectric resonator antennas. IEEE Trans Antennas Propag 50(10):114151419

33. Fiedziuszko SJ, Hunter IC, Itoh T, Kobyashi Y, Nishokawa T, Suitzer SN, Wakino K (2002)

Dielectric materials, devices and circuits. IEEE Trans Microw Theory Tech 50(3):706720

34. Petosa A, Littipiboon A, Thirkoune S (2002) Perforated dielectric resonator antennas. Electron

Lett 38(24):14931495

35. Kishk AA, Yan Y, Glisson AW (2002) Conical dielectric resonator antennas for wide band

applications. IEEE Trans Antenna Propag 50(4):469474

36. So KK, Leung KW (2002) Annular-slot excited dielectric resonator antenna with a backing

cavity. IEEE Antenna Propagation Society international symposium, San Antonio, Texas,

June 2002, pp 470473

37. Ferrando-Bataller M, Valero-Nogueira A, Cabedo-Fabrs M, Antonino Daviu E (2003)

Design of ultra-wideband antennas using characteristic modes. INICA/COST 284 workshop,

Berlin, Germany

38. Li B, So KK, Leung KW (2003) A circular polarized dielectric resonator antenna excited by

asymmetrical U-slot with a backing cavity. IEEE Antenna Wirel Propag Lett. 2:133135

39. Guo YX, Luk KM (2003) Dual polarized dielectric resonator antennas. IEEE Trans Antennas

Propag 51(5):11201123

40. Kishk AA (2003) Wideband truncated tetrahedron dielectric resonator antenna excited by a

coaxal probe. IEEE Trans Antenna Propag 51(10):29132917

41. Gentilli GB, Morni M, Selleri S (2003) Relevance of coupling effects on DRA array design.

IEEE Antenna Propaga 51(3):399404

42. Kishk AA, Zhang X, Glission AW, Kajfez D (2003) Numerical analysis of stacked dielectric

resonator excited by a coaxial probe for wideband application. IEEE Trans Antenna Propaga

51(8):19962006

43. Deng SM, Tsai CL, Chiu CW, Chnag SF (2003) CPW fed rectangular ceramic dielectric

resonator antennas 51(8):21422144

Bibliography 363

44. Petosa A, Ittipiboon A, Antaar Y (2003) Broadband dielectric resonator antennas. In:

Dielectric resonator antennas. Research Studies Press Ltd., UK

45. Ghosh B, Antar YMM, Petosa A, Littipiboon A (2004) Feed conguration of CPW fed DRA.

IEEE Antennas Propagation Society international symposium, Monterrey, CA, June 2004

46. Bit-Babik G, DiNallo C, Faraone A (2004) Multimode dielectric resonator antenna of very

high permittivity. In: Proceedings of IEEE antennas and propagation symposium, Monterrey,

CA, June 2004, pp 13831386

47. Bit Babik B, Di Nallo C, Fraone A (2004) Multimode dielectric resonator antennas of very

high permittivity. IEEE Antennas Propagation Society international symposium, Monterrey,

CA, June 2004

48. Sung Y, Ahn CS, Kim YS (2004) Micro strip fed dual frequency dielectric resonator antennas.

Microw Opt Technol 42(5):388390

49. Littipiboon A, Mongia RK, Antar YMM, Bhartia P, Cuhaci M (2004) An integrated

rectangular dielectric resonator antenna. IEEE Antennas Propagation Society international

symposium, Michigan, June 2004, pp 604607

50. Petosa A, Thirkoune S, Lttipiboon A (2004) Array of performed dielectric resonator antennas.

IEEE Antennas Propagation Society international symposium, Monterrey, CA, June 2004

51. Li B, Leung KW (2004) A wideband strip fed rectangular dielectric resonator antenna. In:

Proceedings of IEEE antennas and propagation symposium, DC, Washington, July 2004,

pp 13831386

52. Sheng XQ, Leung KW, Yung EKN (2004) Analysis of waveguide fed dielectric resonator

antenna using a hybrid innite element method moment method. IEEE Proc Microw Antenna

Propag 151:9195

53. Eshrah IA, Kishk AA, Yakovlev AB, Gilsson AW (2004) Modeling of a waveguide probe

excitation for dielectric resonator antenna. IEEE Antennas Propagation Society international

symposium, Monterrey, CA, June 2004

54. Nannini C, Ribero JM, Dauvignac JY, Pichot C (2004) Bifrequency behavior and bandwidth

enhancement of a dielectric resonator antenna. Microw Opt Technol Lett 42(5):432434

55. Nannini C, Ribero JM, Dauvignac JY, Pichot C (2004) Bifrequency and bandwidth

enhancement of a dielectric resonator antennas. Microw Opt Technol Lett 42(5):432434

56. Joliveau E, Smith J, Wolfe J (2004) Tuning of vocal tract resonances by sopranos. Nature

427:116

57. Li B, Leung KW (2005) Strip-fed rectangular dielectric resonator antennas with/without a

parasitic patch. IEEE Trans Antenna Propag 53(7):22002207

58. de Young CS, Long SA (2006) Wideband cylindrical and rectangular dielectric resonator

antenna. IEEE Antennas Wirel Propag Lett 5:426429

59. Petosa A, Thirakoune S, Littipiboon A (2009) Higher order modes in rectangular DRAs for

gain enhancement. Presented at the ANTEM 2009, Banff, Canada, Feb 2009

60. Chang T-H, Kiang J-F (2009) Bandwidth broadening of dielectric resonator antenna by

merging adjacent bands. IEEE Trans Antennas Propag 57(10):33163320

61. Sebastian ML (2009) Materials for wireless communication. Elsevier, Amsterdam

62. Kong LB (2010) Electrically tunable materials. Elsevier, Amsterdam

63. Huff GH, Rolando DL, Walters P, McDonald J (2010) A frequency recongurable dielectric

resonator antenna using colloidal dispersions. IEEE Antenna Wirel Propag Lett 9

64. Huff GH, Rolando DL, Walters P, McDonald J (2010) A frequency recongurable dielectric

resonator antenna using colloidal dispersions. IEEE Antennas Wirel Propag Lett 9

65. Petosa A, Thirakoune S (2011) Rectangular dielectric resonator antennas with enhanced gain.

IEEE Trans Antennas Propag 59(4):13851389

66. Yaduvanshi RS (2012) Hybrid MHD antenna. Intersci J IJUWBCS 2(4):201206

67. Wang YF, Denidni TA, Zeng QS, Wei G (2014) IEEE Lett

68. Khan QU Higher mode excitation for high gain micro strip antenna. AIU-Int J Electron

Commun

364 Bibliography

69. Guha D, Antaar YMM (2012) Higher order mode excitation for high-gain broadside radiation

from cylindrical dielectric resonator antennas. IEEE Antenna Propag

70. Popov A (1999) Multi-mode dielectric resonator antenna with controllable radiation pattern.

IEEE Trans Antenna Propag

71. Kreyszig E (2002) Advanced engineering mathematics. Wiley for Half wave Fourier analysis

72. Jackson JD (1962) Classical electrodynamics. Wiley, New York

73. Cheng X (2011) Hybrid mode radiation in patch antenna loaded with corrugated electric

arrays. Department of Electrical and Computer Engineering, University of Florida,

Gainesville, FL, USA

74. Chang K, Hsieh LH (2004) Microwave ring circuits and related structures. Wiley Interscience,

New York

75. Adam JJ, Bernhard JT (2013) Broadband equivalent circuit models for antenna impedances

and elds using characteristic modes. IEEE Trans Antenna Propag

76. Jaggi VP, Mathur AB (1985) Advanced engineering mathematics. Khanna publisher (Chapter

13 Series solutions and special functions I.e. Bessel function)

77. Neshati MH, Wu Z (2001) The determination of the resonant frequency of the TEy111 mode in a

rectangular dielectric resonator for antenna application. 11th international

78. Colline BS, Nahar V, Kingsley SP, Zhang SQ, Krupa S (2004) A dual band hybrid dielectric

antenna for laptop computer. IEEE Antenna Propagation Society international symposium

79. Jordan EC, Balmain KG (1968) Electromagnetic and radiating systems. Wiley, New York

80. Landau LD, Liftshiz EM (1984) Electrodynamics of continuous media. Elsevier, Amsterdam

81. Jacksion JD (1962) Classical electrodynamics. Wiley, New York

82. Balanis CA (2005) Antenna theory analysis and design. Wiley, New York

83. Harrington RF (1961) Time harmonic electromagnetic elds. IEEE Press series on

electromagnetic wave theory

84. Sadiku MNO (2009) Principles of electromagnetics. Oxford University Press, Oxford

85. Feynman RP (2013) Lecture series on physics volume II. Pearson Publishing

86. Guha D, Antar YMM (2011) Microstrip and printed antennas. Wiley, New York

Index

Antenna gain, 148, 149, 167, 229 Excitation angle, 126

Antenna recongurability, 230

Aperture coupled feed, 34, 193, 197 F

Field and pattern, 4, 6, 14, 21, 36, 119, 148,

B 165

Boundary conditions, 5, 6, 11, 14, 20, 22, Flux density, 144

2528, 3335, 46, 47, 50, 52, 57, 73, 107, Fourier transform, 9, 105, 118, 124, 125

111, 112, 116, 118, 121, 123, 128, 144, Frequency recongurability, 230

148, 181, 185, 212, 217, 218, 222, 225,

233, 239, 247 G

Gain of RDRA, 166

C

Cavity resonator, 107, 121, 123, 145 H

Characteristic equation, 12, 57, 59, 161 Helmholtz equation, 3, 1215, 20, 23, 25, 27,

Characteristic modes, 7, 12 52, 53, 107, 136, 137, 212, 213, 217, 222,

Charge conservation equations, 211 223, 225, 233

Cherenkov principle, 11 Higher modes, 14, 126, 148, 149, 154,

Circular polarization, 12, 126, 181, 213, 262, 166169, 213, 214, 229, 259

265, 269 Higher-order even modes, 126, 155, 166

Current density, 7, 12, 25, 27, 130, 135, 136, Higher-order odd modes, 126, 155

185, 191, 212, 213, 219, 243 Hybrid modes (HEM), 6, 14, 148, 155, 185,

214, 225, 230

D

Dielectric polarization, 11, 103 K

Dielectric resonator, 1, 2, 1113, 33, 165, 166, KAM (KalmogorovArnoldMoser) theory, 3,

213, 252 13, 24, 212, 214, 217, 225

Dipole moment, 8, 11, 14, 103, 123, 136, 149,

168, 185, 230 L

Directivity, 148, 168, 213 Longitudinal elds, 3, 33, 35, 123, 148

Discrete modes, 23, 53, 233 Lorentzs gauge condition, 118, 140

Dominant mode, 5, 59, 136, 148 Loss tangent, 4

E M

Eigen currents, 58, 148 Magnetic charge density, 106

Eigen values, 7, 166, 241, 242 Magnetic dipoles, 6, 126, 166, 167

Eigen vectors, 7, 12, 57, 148 Magnetic scalar potentials, 136

Electric charge density, 106 Magnetic vector potential, 12, 21, 118, 135,

Electric scalar potentials, 119, 136 138, 140143, 181, 186

R.S. Yaduvanshi and H. Parthasarathy, Rectangular Dielectric

Resonator Antennas, DOI 10.1007/978-81-322-2500-3

366 Index

Maxwells equation, 11, 12, 14, 103, 106, 136, Rectangular cavity resonator, 107, 121

148, 212214, 221 Rectangular dielectric resonator antenna

MHD antenna, 310 (RDRA), 18, 12, 13, 24, 28, 29, 3336,

Miniaturization, 168 45, 54, 57, 79, 82, 87, 102, 103, 106, 112,

118, 121, 123, 126, 135, 137, 141, 144,

O 146, 148, 149, 154156, 159, 165171,

Odd order modes, 126, 154, 168, 229 193, 198, 199, 205, 207, 211214, 217,

Orthogonal polarization, 231 219, 225, 227, 230, 232, 251, 262, 269,

Orthogonal radiation, 5, 12, 154 279, 282

Orthogonality property, 24, 54, 126, 202, 212 Relative permittivity, 11, 103

Orthonormality, 6, 34, 35, 104, 105, 122, 185, Resonant cavity, 11, 183

213, 216 Resonant frequency, 1, 68, 11, 13, 22, 2931,

Orthonormality principle, 6, 13, 216, 225 33, 35, 57, 59, 72, 78, 82, 85, 87, 103, 126,

136, 141, 144, 146, 148, 156, 159, 167,

P 168, 193, 198, 203, 225, 229, 282, 284, 286

Parsevals power theorem, 12, 136 Resonant modes, 47, 1113, 21, 2729, 31,

Patch antenna, 1, 165, 166 3436, 81, 103, 106, 120, 125, 130,

Pattern recongurability, 230 148150, 155, 165168, 172, 174, 185,

Perfect electric conductors (PEC), 107, 118, 186, 198, 199, 207, 218, 220, 229

212 Resonating modes of RDRA, 221

Perfect magnetic conductors (PMC), 107, 211 Resonator current, 105

Permittivity, 2, 4, 13, 30, 87, 112, 149, 156, Rigorous theoretical analysis, 11

169, 285

Perturbation, 4, 13, 112, 148, 166, 212 S

Perturbed modes, 13 Smith chart, 195, 260, 276

Polarization, 168, 213 Surface current density, 12, 185, 191, 213

Polarization recongurability, 229

Poynting theorem, 13 T

Poynting vector, 12, 137, 140, 143, 181 Taylors expansion, 199, 204

Probe current, 3, 12, 104, 122, 125, 127, 146 Transcendental equation, 33, 35, 57, 59, 65, 72,

Propagation constant, 6, 7, 12, 14, 33, 35, 52, 78, 102, 158, 161, 225

57, 59, 121, 129, 136, 159, 161 Transverse electric (TE), 14, 35, 148

Propagation parameter, 109 Transverse magnetic (TM), 14, 148

Q U

Quality factor, 3, 12, 34, 144146 Unperturbed modes, 242

R V

Radiating lobes, 79, 80 Variance function, 199

Radiation pattern, 7, 12, 14, 34, 35, 80, 125, Vector potential, 12, 21, 118, 119, 135, 136,

135, 136, 138, 139, 141143, 152, 166, 141, 142, 243

168, 181, 185, 186, 188, 192, 193, 198,

213, 229, 243, 246, 251, 256, 269 W

Random variables, 199 Wave number, 6, 21, 22, 28, 57

RDRA mathematical modeling, 22, 36, 135 Wave vector, 5, 12, 21, 135

RDRA modes, 4, 5, 11, 103, 121, 123, 168

RDRA resonant modes, 4, 5, 7, 11, 12, 28, 34,

79, 103

- 8-Yangui-et-al.pdfUploaded byMajdi Yangui
- Solution Manual for Time-Harmonic Electromagnetic Fields - HarringtonUploaded bysasinnn
- Cyclic Symmetry TopicsUploaded byMoisés Machado
- s511p10Uploaded byVenkatesh Punna
- wideband simple dieletric resonator antennaUploaded byRohit Sharma
- A Study on the Vibrational Effects of Adding an Auxiliary Chassis to a 6-Ton Truck.pdfUploaded byPaul
- Harrington, Roger F_[Time-harmonic electromagnetic fields].pdfUploaded bySaransh Malik
- Lecture 17Uploaded byGaurvi Arora
- Lec 1b IntroductionUploaded byMuhammadwaqasnaseem
- Antenas InteligentesUploaded byndlr81
- Ecc DerivationUploaded byJawed Aquil
- 3P7_0879Uploaded byfulanito795
- Stasiun Bumi LABUHA StabilloUploaded byRachmadHakimSutarto
- Nkwenda3cap_5 and 6 AcceptanceUploaded byAli Ali
- L12_LoopUploaded byRaj Hakani
- Andrew DBXLH-6565C-VTM.pdfUploaded bylucasgcba
- antenna designUploaded bydhruvaaaaa
- BPH-202 Waves & OpticsUploaded byVijay Kumar
- HFSS Antennas - Arrays.pdfUploaded byYohandri
- Sten_TAP_01Uploaded bySaukarsha Roy
- Antenna ReportUploaded byhasnizaaziz39
- Analysis of Rectangular Waveguides - An Intuitive ApproachUploaded byHammad Akhtar
- 1205 Ieee Antenna Webinar 2013 FinalUploaded byJeremy Retamozo Gozar
- Angle PrincipleUploaded byfredjmor
- Waveguides _ Transmission LinesUploaded byవేలుసామి లింగాసామి
- 08420471Uploaded byNguyễn Xuân Đỉnh
- Introduction to AntennaUploaded byAchriz
- INSTALLATION MANUAL DEH-P2600Uploaded byrichard_rodriguez
- AD-17-C-1512Uploaded bycuonglanchi
- Rideout_MME2012 (2)Uploaded bySahaya Grinspan

- 2012 04 Arbitrary Footprints from Arrays with Concentric Ring Geometry and Low Dynamic Range Ratio.pdfUploaded byDania Alashari
- Book - 1969 Antenna Theory Part 2Uploaded byDania Alashari
- Fundamentals of ElectromagneticsUploaded byPPPPMMMM
- Fundamentals of ElectromagneticsUploaded byPPPPMMMM
- 2103 05 Arrays of Concentric Rings of Elements Synthesis of Pencil Beams.pdfUploaded byDania Alashari
- 1991 04 Statistically Thinned Arrays with Quantized Element Weights.pdfUploaded byDania Alashari
- 1991 08 Sidelobe Performance in Quadratic Phase Conformal Arrays.pdfUploaded byDania Alashari
- Kraus Antennas 2nd.edition 1988Uploaded byDimitris Stathis
- Book - 2015 Microwave Circuit Design a Practical Approach Using ADSUploaded byDania Alashari
- 1992 04 Application od Uniform Linear Array Bearing Estimation Techniques to Uniform Circular Arrays.pdfUploaded byDania Alashari
- Book - 1998 Electromagnetic Field Theory Fundamentals by Bhag Singh GuruUploaded byDania Alashari
- Book - 1969 Antenna Theory Part 1Uploaded byDania Alashari
- antennaskrauss28195029Uploaded bybarthod
- 38576Uploaded byNaga Lakshmaiah
- Antenna Engineering Handbook.pdfUploaded byMatheus Freitas Kuhn
- LGW2eCh4SolutionsUploaded byDania Alashari
- 1987 02 Synthesis of Near-Field Patterns of Arrays.pdfUploaded byDania Alashari
- Lgw 2 Ech 9 SolutionsUploaded byDania Alashari
- 1990 03 Bearing Estimation of Coherent Sources by Circular Spatial Modulation Averaging CSMA.pdfUploaded byDania Alashari
- 1990 09 Large Circular Array with One Element Driven.pdfUploaded byDania Alashari
- LGW2ECh3SolutionsUploaded byDania Alashari
- Lgw 2 Ech 7 SolutionsUploaded byalwyne_edison
- Lgw 2 Ech 8 SolutionsUploaded byDania Alashari
- 1989 12 Feasibility Study of Stripline-Fed Slots Arranged as Planar Array with Circular Grid and Circular Boundary.pdfUploaded byDania Alashari
- LGW2ECh6solutionsUploaded byDania Alashari
- Lgw 2 Ech 10 SolutionsUploaded bykzhaider
- 1989 08 Circular Array and Nonsinusoidal Waves.pdfUploaded byDania Alashari
- LGW2ECh5solutionsUploaded byDania Alashari
- 1989 02 Supergain Antennas and the Yagi and Circular Arrays.pdfUploaded byDania Alashari

- 0109324v1Uploaded byMatteo Scandi
- Literature Review on Inverse KinamaticUploaded byFirdaus Azlan
- Ferroelectricity and SupersetsUploaded byDarani Priya
- Turbine BladeUploaded byrajakgec
- EnthalpyUploaded byManoj Kumar Chaurasia
- 9709_s12_qp_43Uploaded byAbrar Jahin
- Wave Interference - Short Questions With Answer _ Physics Class 12Uploaded byMohan Raj
- CHE3166 - Lecture - 11Uploaded byvenkiee
- 7.5 Design Report RaftUploaded byDe Silva Shmapk
- Strain Gages BridgesUploaded byK
- Kantian Caucality and Quantum quarks.pdfUploaded byKamil
- geodetic precession &frame draggingUploaded byOjeshKoul
- Ameri Paper 97Uploaded byanjananandhamm
- Hydraulic JumpUploaded bysafder_ali90
- 1.pdfUploaded byej victoria
- Strength of Material by u c JindalUploaded bykrishna
- CE6405-Soil Mechanics 2Uploaded bySekar Dinesh
- 612_aci Structural Journal by Wiryanto DewobrotoUploaded bymavane25
- Oscillations and WavesUploaded bybinaywatch
- SlidesM361-1-2012Uploaded byPreeshant Beezloll
- Structural Analysis Terex Demax Latice CraneUploaded bysolidcad
- Jurnal Hukum PoiseuilleUploaded byAnonymous lfvL0fA
- A2_StrainGages_2003Uploaded byFlexi Soufiane
- Tutorial 18 - Heat Transfer - TeapotUploaded by염규정
- HU_11P18Uploaded byvenkata
- The Magnetic Field and Electrical CurrentsUploaded bymidhun
- Hl Waves Reflect Refract AnswersUploaded bymandoolala
- Kinema TicsUploaded bylhadjkfh'sm
- Design of PSC Cross Girder as Per IRC 112Uploaded byrrajdeep54
- 145627072 Pile Foundation DesignUploaded byscist