Development of Circularly Polarized Microstrip ... - CEReS - 千葉大学

Development of Circularly Polarized Microstrip
Antennas for CP-SAR System Installed on
Unmanned Aerial Vehicle
July 2012
YOHANDRI
Graduate School of Advanced Integration Science
CHIBA UNIVERSITY
i

( 千 葉 大 学 学 位 申 請 論 文 )
Development of Circularly Polarized Microstrip
Antennas for CP-SAR System Installed on
Unmanned Aerial Vehicle
2012 年 7 月
千 葉 大 学 大 学 院 融 合 科 学 研 究 科
情 報 科 学 専 攻 知 能 情 報 コース
ヨハンドリ
i

Abstract
The discussion about development of circularly polarized microstrip antennas for circularly
polarized synthetic aperture radar (CP-SAR) system will be presented in this dissertation. The
antennas are designed to operate in L-band and will be installed on an Unmanned Aerial Vehicle
(UAV). The single element and array configuration of the antenna are designed, numerically
analyzed, fabricated and measured experimentally. The single element antenna is designed
using circular microstrip shape, and the circular polarization radiation is produced with a
triple-fed which the phase shift between the three feeds is adjusted about of 120 to obtain
the good 3-dB axial ratio performance. In other hands, an array antenna (2 × 6 elements) is
developed with the proximity feed and a circular-sector stub as a power divider is adopted in
feeding network. In addition, a design of broadband circularly polarized microstrip antenna
and a low side lobe level array antenna also will be presented in this dissertation. There
are several considerations in developing the antenna, the specification of the CP-SAR system,
payload of the UAV and the space for the antenna onboard UAV. In general, numerical analyses
of the proposed antennas using the method of moments can lead to a good agreement with
experimental results. The slight differences of antenna performance between the simulation
and measurement are probably due to imperfection during the fabrication processes. These
antennas with its good performance will be useful for several L-band applications, especially
for circularly polarized synthetic aperture radar system. Based on the whole research works,
new analysis has been made on the characteristics of proposed antennas with new feed method
both single element and array configuration of the antenna.
ii

Abstract
無 人 航 空 機 搭 載 用 CP-SAR システムのための 円 偏 波 マイクロス
トリップアンテナの 開 発
本 研 究 では、 円 偏 波 合 成 開 口 レーダ(CP-SAR) 搭 載 小 型 衛 星 の 実 現 に 向 けて、CP-SAR 搭 載
無 人 航 空 機 (UAV)を 開 発 した。 本 論 文 のテーマは、その 中 でもとくに CP-SAR UAV システ
ム 用 の 円 偏 波 マイクロストリップアンテナの 開 発 とその 結 果 について 議 論 する。 L バンド
(1.27 GHz) 帯 で 使 用 する CP-SAR システム 用 のアンテナを、 小 型 、 薄 型 、 軽 量 という 条 件
のもとで 単 素 子 およびアレー 構 造 で 設 計 した。その 際 、モーメント 法 (MoM)を 利 用 して、 単
素 子 およびアレー 型 アンテナを 数 値 的 に 解 析 して 最 適 な 構 造 を 見 出 し、 試 作 を 行 った。また、
この 解 析 による 予 測 結 果 と、 電 波 無 響 室 内 における 実 験 結 果 とを 比 較 検 討 した。 単 素 子 アン
テナの 設 計 では、 円 形 マイクロストリップ 素 子 を 使 用 した。トリプル 分 岐 の 給 電 型 マイクロ
ストリップラインを 約 120 o の 角 度 で 調 整 した 結 果 、3 つの 給 電 位 相 シフターで 最 適 な 3dB の
軸 比 を 得 ることができた。 一 方 、アレー 型 アンテナは 切 り 掛 け 部 分 が 付 いた 正 方 形 のマイク
ロストリップで 構 成 した。このアンテナの 給 電 系 は、 電 磁 結 合 型 給 電 系 と 円 形 セクタースタ
ブの 電 力 分 配 器 により 構 成 した。UAV 上 のアンテナ 収 納 スペースによってアレー 素 子 数 が 制
限 されるので、アンテナ 素 子 数 (2 x 6 素 子 )に 合 わせて、 円 形 セクタースタブによる 電 力 給
電 の 分 配 を 調 整 した。 本 研 究 の 結 果 、MoM による 数 値 解 析 の 予 測 は 実 験 結 果 とほぼ 一 致 する
ことが 明 らかになった。ただし、 製 造 プロセス 中 の 誤 差 によって 一 部 の 特 性 に 誤 差 が 生 じて
おり、 今 後 、 製 造 法 の 改 良 が 必 要 であることがわかった。 本 論 文 で 開 発 されたアンテナは、
今 後 UAV に 搭 載 する CP-SAR センサとして 実 地 に 活 用 される 予 定 である。
iii

Acknowledgements
First and foremost, my greatest debts are to my supervisor Professor Hiroaki Kuze and Professor
Josaphat Tetuko Sri Sumantyo, who takes so much effort and patience in mentoring me to
become a qualified researcher. From leading me at the beginning into the Maxwell domain and
then step by step to its actual application in Antennas, to revise my poorly written papers and
edited presentations during my study in Chiba University. I was taught everything I need to be
a good researcher including being creative, thinking deeply, and the skills for presenting ideas
and writing papers.
I would also like to thank the Directorate General of Higher Education (DGHE) of Indonesia
for their financial support during my doctor studies at Chiba University. Also our gratitude to
the Japan Society for the Promotion of Science (JSPS) for Grant-in-Aid for Scientific Research -
Young Scientist (A); National Institute of Information and Communication Technology (NICT)
for International Research Collaboration Research Grant.
I always feel lucky to be with so many excellent researchers in Microwave Remote Sensing
Laboratory (MRSL), Center for Environmental Remote Sensing (CEReS), Chiba Univeristy. I
sincerely thank to my labmates and colleagues: Victor Wissan, Prilando RA, Luhur Bayuaji,
Bambang Setiadi, Ilham Alimuddin, Iman Firmansyah, Takafumi Kawai, and many others that
I cannot mention everyone in here for being a good partner and veluable suggestions and sharing
the knowledge.
I am very grateful to my mother Nurjasmi and my father Azwir who encourage and support
me during their life time. This dissertation is dedicated to my family, thank you for letting me
be far away from you all, to pursue my doctor degree in Japan. To my sisters Gusnayetti, Rina
Mazriani, Rahmi Junita, Nita Yusmaniarti and my late brother Yohanes thank you very much
for all of your supplications and support.
Finally, I would like to give a special thanks to my wife Fivit Andriani, my son Dzaky Afandi
Tsaqif and my daughter Queensha Mizuki Azkadina for their love, patience and encouragement.
iv

Contents
Abstract
Abstract (Japanese)
Acknowledgements
Contents
List of Figures
List of Tables
ii
iii
iv
v
viii
xii
1 Introduction 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Motivation and objectives of research . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Dissertation focus and organization . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Circularly Polarized Synthetic Aperture Radar 8
2.1 The principles of radar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2 Synthetic Aperture Radar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Polarization of electromagnetic waves . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.4 Circularly Polarized Synthetic Aperture Radar . . . . . . . . . . . . . . . . . . . 17
2.5 Design of CP-SAR onboard UAV . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.5.1 Design of CP-SAR parameters . . . . . . . . . . . . . . . . . . . . . . . . 19
2.5.2 Design of CP-SAR system . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.5.2.1 UAV system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.5.2.2 CP-SAR sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
v

CONTENTS
3 Circularly Polarized Microstrip Antenna 29
3.1 Microstrip antenna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.1.1 Radiating patch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.1.2 Dielectric substrate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.1.3 The ground . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.1.4 Feeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.2 Antenna basic parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.2.1 Scattering parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.2.2 Antenna efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2.3 Antenna Gain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2.4 Polarization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.2.5 Bandwidth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2.6 Radiation pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.2.7 Current density and distribution . . . . . . . . . . . . . . . . . . . . . . . 36
3.3 Circularly polarized microstrip antenna . . . . . . . . . . . . . . . . . . . . . . . 36
3.3.1 Dual feed circularly polarized microstrip antenna . . . . . . . . . . . . . . 37
3.3.2 Single feed circularly polarized microstrip antenna . . . . . . . . . . . . . 38
3.3.3 Triple feed circularly polarized microstrip antenna . . . . . . . . . . . . . 40
3.4 Microstrip Array Antenna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.4.1 Linear array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.4.2 Planar array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.4.3 Dolph-Chebyshev array . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4 Methodology 55
4.1 Design methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.2 Design procedure and electromagnetic modeling . . . . . . . . . . . . . . . . . . . 55
4.2.1 Design and analysis using Method of Moment (MoM) . . . . . . . . . . . 55
4.2.2 Design and analysis using Finite Element Method (FEM) . . . . . . . . . 57
4.3 Fabrication of proposed antenna . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.4 Antenna measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.4.1 Instrumentation system . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.4.2 Anechoic chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.4.3 Input characteristics measurement . . . . . . . . . . . . . . . . . . . . . 61
4.4.4 Radiation characteristics measurement . . . . . . . . . . . . . . . . . . . 62
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5 Results and Discussion 64
5.1 Triple Proximity-fed circularly polarized microstrip antenna . . . . . . . . . . . . 64
5.1.1 Design of proposed antenna . . . . . . . . . . . . . . . . . . . . . . . . . 65
vi

CONTENTS
5.1.2 Parameter study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.1.3 Input characteristic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.1.4 Radiation characteristic . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.2 LHCP array microstrip antenna . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.2.1 Array antenna design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.2.2 Input characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.2.3 Radiation characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.3 RHCP array microstrip antenna . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.3.1 Array antenna configuration . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.3.2 Input characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.3.3 Radiation characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.4 Broadband circularly polarized microstrip antenna . . . . . . . . . . . . . . . . . 88
5.4.1 Geometry design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.4.2 Antenna performances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.5 Low sidelobe level array Antenna . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.5.1 Dolph-Chebyshev synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.5.2 Geometry design of the antenna . . . . . . . . . . . . . . . . . . . . . . . 93
5.5.3 Simulated and measured results . . . . . . . . . . . . . . . . . . . . . . . . 95
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6 Conclusions 103
6.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.2 Future works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
List of Publications 106
Curriculum Vitae 108
Appendix A IE3D Electromagnetic Simulation 109
Appendix B HFSS Electromagnetic Simulation 113
Appendix C Antenna fabrication 119
vii

List of Figures
1.1 Electric field components for dihedral corner (Harold, 1986) . . . . . . . . . . . . 2
1.2 Design of CP-SAR experiment onboard an UAV. . . . . . . . . . . . . . . . . . . 4
1.3 Information data on LP-SAR: (a) magnitude and (b) phase. . . . . . . . . . . . . 5
2.1 Basic principle of radar (Merrill, 2001). . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Block diagram of fundamental radar system. . . . . . . . . . . . . . . . . . . . . 9
2.3 Basic principles of aperture synthesis. . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4 Synthetic aperture radar geometry. . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.5 Azimuth ambiguities: Doppler frequency histories of targets A and B are equal
due to aliasing about the sampling frequency of PRF (Li L.K. et al., 1983). . . . 13
2.6 Range ambiguities: Portions of the radar return from a previous pulse overlap
the return from the present pulse (Li F.K. et al., 1983). . . . . . . . . . . . . . . 14
2.7 The LHCP wave shown at a fixed instant of time (Warren L., 1993). . . . . . . . 15
2.8 Polarization of the waves: (a) linearly polarized, (b) Left-Hand Ellipse Polarized
(LHEP) and (c) Right-Hand Ellipse Polarized (RHEP). . . . . . . . . . . . . . . 17
2.9 Basic principle of aperture synthesis. . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.10 Description of 3-dB axial ratio beamwidth for CP-SAR. . . . . . . . . . . . . . . 19
2.11 CP-SAR geometry: (a) slant range view and (b) azimuth plane view. . . . . . . . 20
2.12 The required (a) θ ECP and (b) θ ACP for 1 m image resolution. . . . . . . . . . . 22
2.13 Design of the UAV (unit in mm). . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.14 Unmanned aerial vehicles: (a) profile of UAV and (b) photograph. . . . . . . . . 25
2.15 Design of CP-SAR sensor onboard UAV. . . . . . . . . . . . . . . . . . . . . . . . 26
3.1 Basic geometry of microstrip antenna. . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2 Various shapes of microstrip antennas (Girish and Ray, 2003). . . . . . . . . . . . 31
3.3 The signal flow graph representation of a two-ports network network. (a) Defenition
of incident and reflected wave. (b) Signal flow graph (Pozar, 2005). . . . . 32
3.4 Parameters in electromagnetic wave polarization. . . . . . . . . . . . . . . . . . . 34
3.5 Dual-feed (a) SMSA and (b) CMSA (Girish and Ray, 2003). . . . . . . . . . . . . 37
viii

LIST OF FIGURES
3.6 Rectangular patch arrangements for circular polarization: (a) Through a power
divider and (b) Through a 90 o hybrid (Balanis, 2005). . . . . . . . . . . . . . . . 38
3.7 Single-feed arrangements for circular polarization microstrip antennas: (a) Square,
(b) Circular and (c) Elliptical. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.8 Single feed circularly polarized microstrip antennas; (a) Nearly square patch and
(b) Amplitude and phase of the two modes (Lee, S.K., et al., 2005). . . . . . . . 39
3.9 Geometry design of triple feed circular microstrip antenna. . . . . . . . . . . . . 40
3.10 Linearly polarized wave on triple feed circular microstrip antenna. . . . . . . . . 41
3.11 Linear array geometry (Balanis, 2005). . . . . . . . . . . . . . . . . . . . . . . . . 44
3.12 Geometry of a planar array. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.13 Polar coordinates showing incremental solid angle dA = r 2 dΩ on the surface of
a sphere of radius r where dΩ = solid angle subtended by the area dA. . . . . . . 47
3.14 Non-uniform amplitude arrays of even and odd number of elements (Balanis, 2005). 49
4.1 Design methodology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.2 Flow chart of the antenna fabrication. . . . . . . . . . . . . . . . . . . . . . . . . 58
4.3 Diagram of required antenna measurement equipment. . . . . . . . . . . . . . . . 60
4.4 Schematic of the antenna measurement system. . . . . . . . . . . . . . . . . . . . 60
4.5 Photograph of antenna measurement in anechoic chamber at MRSL, Chiba University:
(a) Array antenna and (b) Triple-fed antenna. . . . . . . . . . . . . . . . 61
4.6 Antenna gain measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.1 The 3-way circular-sector-shaped power divider. . . . . . . . . . . . . . . . . . . . 65
5.2 Geometry design of proposed antenna; (a) top view and (b) side view. . . . . . . 66
5.3 Photograph of fabricated CMA: (a) triple proximity-fed and (b) circular radiator. 66
5.4 Simulation results showing the frequency dependence of the axial ratio (AR) of
the CMA for various values of phase shift applied to the feeds. . . . . . . . . . . 67
5.5 Vector current distribution of the designed CMA for various phase values of the
source, (a) φ s = 0 o , (b) φ s = 45 o , (c) φ s = 90 o , and (d) φ s = 135 o . . . . . . . . 68
5.6 The efficiency of the antenna configuration as a function of frequency. . . . . . . 68
5.7 Simulated and measured reflection coefficient vs. frequency. . . . . . . . . . . . 69
5.8 Simulated and measured input impedance (Z in ) plotted as a function of frequency.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
5.9 Simulated and measured axial ratio (AR) vs. frequency at θ = 0 o . . . . . . . . . 70
5.10 Simulated and measured gain (G) vs. frequency at θ = 0 o . . . . . . . . . . . . . 71
5.11 Measured and simulated radiation pattern of proposed antenna at f = 1.28
GHz: (a) in the y − z plane (Az = 90 o and 270 o ) and (b) in the x − z plane
(Az = 0 o and 180 o ). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
5.12 Configuration of the antenna array consisting of three blocks, each block having
2×2 element patches: (a) top view, and (b) side view. . . . . . . . . . . . . . . . 75
ix

LIST OF FIGURES
5.13 Photograph of the fabricated antenna: (a) feed network and (b) 2 × 6 radiation
patch. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.14 Geometry layout of a single piece of antenna array. . . . . . . . . . . . . . . . . . 77
5.15 Simulated and measured reflection coefficient plotted as a function of frequency. . 78
5.16 Simulated and measured real input impedance (Z in ) plotted as a function of
frequency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.17 Simulated and measured axial ratio (AR) plotted as a function of frequency. . . . 79
5.18 Relationship between antenna gain and frequency at θ angle = 0 o . . . . . . . . . 80
5.19 Array antenna characteristics in the theta plane (negative theta for Az = 180 o
and positive for Az = 0 o ) (x − z plane) at f = 1.27 GHz: (a) gain versus theta
angle, and (b) axial ratio versus theta angle. . . . . . . . . . . . . . . . . . . . . . 81
5.20 Array antenna characteristics in the theta plane (negative theta for Az = 270 o
and positive for Az = 90 o ) (y − z plane) at f = 1.27 GHz: (a) gain versus theta
angle, and (b) axial ratio versus theta angle. . . . . . . . . . . . . . . . . . . . . . 82
5.21 Cross polarization (E-left and E-right) of array antenna at f = 1.27 GHz. . . . . 83
5.22 RHCP array antennas: (a) configuration design and (b) photograph. . . . . . . . 84
5.23 Simulated and measured reflection coefficient plotted as a function of frequency. . 85
5.24 Simulated and measured axial ratio (AR) plotted as a function of frequency. . . . 86
5.25 Relationship between antenna gain and frequency at θ angle = 0 o . . . . . . . . . 86
5.26 Array antenna characteristics in the theta plane (negative theta for Az = 180 o
and positive for Az = 0 o ) (x − z plane) at f = 1.27 GHz: (a) gain versus theta
angle, and (b) axial ratio versus theta angle. . . . . . . . . . . . . . . . . . . . . . 87
5.27 Geometry of the proposed antenna: (a) top and side view and (b) 3D view. . . . 89
5.28 Reflection coefficient plotted as a function of frequency. . . . . . . . . . . . . . . 89
5.29 Axial ratio (AR) plotted as a function of frequency. . . . . . . . . . . . . . . . . . 90
5.30 Relationship between antenna gain and frequency at θ angle = 0 o . . . . . . . . . 90
5.31 Radiation pattern of the antenna at f = 1.27 GHz. . . . . . . . . . . . . . . . . . 91
5.32 Axial ratio plotted as a function of theta angle. . . . . . . . . . . . . . . . . . . . 91
5.33 A 3D beam pattern of the antenna at f = 1.27 GHz. . . . . . . . . . . . . . . . . 92
5.34 Dolph-Chebyshev array factor for five elements and λ 0 /2 spaced. . . . . . . . . . 94
5.35 Geometry design of the Dolph-Chebyshev array antenna. . . . . . . . . . . . . . . 94
5.36 Photograph of fabricated feed network and radiation patch. . . . . . . . . . . . . 95
5.37 Simulated and measured reflection coefficient plotted as a function of frequency. . 96
5.38 Simulated and measured VSWR plotted as a function of frequency. . . . . . . . . 97
5.39 Simulated and measured axial ratio plotted as a function of frequency. . . . . . . 97
5.40 Relationship between antenna gain and frequency at θ angle = 0 o . . . . . . . . . 98
5.41 Array antenna characteristics in the theta plane (negative theta forAz = 180 o
and positive for Az = 0 o ) (x − z plane) at f = 1.272 GHz, (a) normalized gain
versus theta angle, and (b) axial ratio versus theta angle. . . . . . . . . . . . . . 99
x

LIST OF FIGURES
5.42 Array antenna characteristics in the theta plane (negative theta forAz = 270 o
and positive for Az = 90 o ) (y − z plane) at f = 1.272 GHz, (a) normalized gain
versus theta angle, and (b) axial ratio versus theta angle. . . . . . . . . . . . . . 100
A.1 The flow chart of a basic IE3D EM simulation. . . . . . . . . . . . . . . . . . . . 109
B.1 Patch antenna layout. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
B.2 Lamped port. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
B.3 Patch antenna layout showing airbox and waveport. . . . . . . . . . . . . . . . . 115
B.4 Solution setup for the simulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
B.5 Frequency sweep in simulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
B.6 Validation check of the proposed antenna. . . . . . . . . . . . . . . . . . . . . . . 116
B.7 Return loss and axial ratio of antenna from 0.9 GHz to 1.6 GHz. . . . . . . . . . 117
B.8 Three-dimensional far-field patterns. . . . . . . . . . . . . . . . . . . . . . . . . . 118
C.1 (a) Seven Mini PCB Prototyping Machine (b) Tools used for microwave artwork
of the antenna. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
C.2 UV exposure machine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
C.3 Etching tank and chemical powder. . . . . . . . . . . . . . . . . . . . . . . . . . . 122
C.4 Bonding between SMA connector and microstrip feed lines for the equilateral
triangular antenna. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
C.5 Step by step fabrication the microstrip antenna with prototyping machine. . . . . 123
C.6 Step by step fabrication the microstrip antenna with dry film photoresist. . . . . 123
xi

List of Tables
2.1 Parameters of CP-SAR Onboard UAV. . . . . . . . . . . . . . . . . . . . . . . . . 23
2.2 Basic Specification of UAV (JX-1). . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5.1 Triple proximity-fed parameters (in units of mm). . . . . . . . . . . . . . . . . . . 65
5.2 Geometry parameters (mm) of circularly polarized array antenna. . . . . . . . . . 75
5.3 Optimum parameters of the proposed antenna. . . . . . . . . . . . . . . . . . . . 95
xii

Chapter 1
Introduction
1.1 Background
The principle of synthetic aperture radar (SAR) has been discovered in the early 50th. Nowadays,
progress in technology and digital signal processing has made SAR play an important
role in a variety of applications in military, topographical, and environmental situations. SAR
is a coherent radar system whose defining characteristic is its use of relative motion between an
antenna and its target region to simulate an extremely large antenna or aperture electronically.
Signal processing uses magnitude and phase of the received signals over successive pulses from
elements of a synthetic aperture to generate high-resolution remote sensing imagery.
As compared with optical sensors, a SAR sensor has many advantages. The SAR sensor
can be operated in all-weather condition and enables penetration through clouds and even
forest canopy, since it works in the microwave frequency. In addition, SAR is an active remotesensing
system which means radar self-illuminates an area by transmitting pulses of microwave
energy. These pulses of radar energy are reflected from the illuminated area and recorded by
the receiver. Thereby, its sensor can be operated both day and night time.
Until now, a number of airborne SAR sensors have employed the linearly polarized (LP)
antenna systems(Nemoto et al., 1991 and Raney et al., 1991) with limited information retrieval.
The characteristics of the conventional SAR sensor are bulky, high power consumption, with
sensitivity to Faradays rotation effect, etc.(Sumantyo et al., 2009). Also, radiating microwave
1

in linear polarization has some flaws due to propagation phenomena such as the variation of
geometric distance between the radar system and the earth surface, the occurrence of a phase
shift when microwave strikes smooth and reflective surfaces, leading to unwanted modulation
of backscatter signals. For the feature that has a dihedral corner segment, the polarization of
a transmitted linear polarized signal is rotated by twice (Harold, 1986) as ilustrated in Figure
1.1. It will result in reduced average power in the received signal due to polarization mismatch
loss in the receiver.
Figure 1.1: Electric field components for dihedral corner (Harold, 1986)
In addition, electromagnetic waves propagating through the ionosphere interact with the
electrons and magnetic fields causes Faraday rotation disturbances (Vine et al., 2007). The
disturbances will change the vector of the electric field of the electromagnetic wave. The vector
rotation of the electric field is called as Faraday rotation with an angle ψ. Estimation of this
angle can be formulated (Rignot, 2000) as.
ψ = 2.6 × 10 −13 T ECBλ 2 cos(θ) (1.1)
Where ψ is in radians, T EC is the total electron content of the slab of the ionosphere below
which the radar measurements are collected in electrons/meter. B is the Earths magnetic field
2

in Tesla, λ and θ is the radar wavelengths in meters and angle between the magnetic field and
the radar illumination in radians, respectively.
It is clear from (1.1), the Faradays rotation interference will impact to the radar system,
especially for longer wavelength. The ionosphere condition has significant contribution to the
state of the microwave polarization plane.
In some bands of the SAR sensor, the Faraday
rotation effect can reach 321 o for the P-band, 40 o for L-band and 2.5 o for C-band (Freeman
and Saatchi, 2004). For example, the Faraday rotation effect greater than 25 o is found in the
PALSAR sensor installed onboard Advance Land Observation Satellite (ALOS) that working
in L-band range (Meyer and Nicoll, 2008).
One possible solution to reduce or eliminate the destructive effects of LP-SAR is radiating
the microwave in circular polarization (CP) (Maini and Agrawal, 2007).
The use of CP in
SAR sensor is recommended, since CP signal that arrives on the Earths surface will stay the
same as the transmitted one (Dubois F. et al., 2008). In addition, the full characterization
of SAR signals backscattered from a random object can only be possible through the use of
circular polarization (Raney, 2007).
In a study of analyzing quadrature-polarized synthetic
aperture radar data from sloping terrain, the synthesis results of circularly polarized data are
far better compared to those constructed from the conventional linearly polarized data (Lee et
al., 2000). Furthermore, in study surface roughness using polarimetric SAR data, more sensitive
measurements can possibly be obtained using circular polarization (Mattia et al., 1997).
In order to realize the CP-SAR sensor, currently, synthetic aperture radar with circularly
polarized antenna is developed in the Microwave Remote Sensing Laboratory (MRSL), CEReS,
Chiba University. This sensor will be installed onboard an unmanned aerial vehicle (UAV).
The working frequency of this sensor is selected on L-band region with the center frequency at
1.27 GHz (λ = 23.6 cm). This band has superior features which relatively longer wavelength
such as yielding useful information to distinguish characteristics of the earth surface and better
penetration through vegetation canopies.
In other hand, for the larger objects and surface
features, this band provides strong backscattering signal (Birk et al., 1995). The capability of
an L-band SAR data also can be found in estimation the arctic glacier motion (Strozzi et al.,
2006).
In experiment design, CP-SAR sensor onboard UAV is composed of right-handed circular
3

polarization (RHCP) and left-handed circular polarization (LHCP) array antennas. The radar
signal from UAV platform is transmitted by either LHCP or RHCP antenna. The backscattering
signal from the target is captured by both LHCP and RHCP array antennas to generate the
axial ratio image (see Figure 1.2). As it can be seen, the microstrip antenna is the important
part in the system. Therefore, in this dissertation, the developments of circularly polarized
microstrip antennas are proposed. The numerical calculation and the experimental results of
the antennas are shown and discussed.
Figure 1.2: Design of CP-SAR experiment onboard an UAV.
1.2 Motivation and objectives of research
There are also the most important issues regarding the CP-SAR system. In LP-SAR system,
the information is limited, which consists of magnitude and phase (Figure 1.3). As compared
with LP-SAR sensor, a greater amount of information about scenes and targets being imaged
would be provided with a CP-SAR sensor, such as axial ratio, ellipticity and tilt angle. These
parameters are expected can to be implemented in various remote-sensing experiments.
To realize the CP-SAR system, an antenna for the L-band CP-SAR sensor is required.
4

Figure 1.3: Information data on LP-SAR: (a) magnitude and (b) phase.
Hence, as objective of the research is to describe the development of circularly polarized microstrip
antenna for CP-SAR system installed on UAV.
1.3 Dissertation focus and organization
In this dissertation, we intend to realize the CP-SAR system installed on UAV with focus on
design and development circular polarized microstrip antenna. The new technique in generating
single element CP antenna and array antenna configuration will be discussed.
This dissertation is divided into 6 chapters. In Chapter 1, a general introduction, motivation
and objective of the research and dissertation focus is made. Chapter 2 cover the theory
circularly polarized synthetic aperture radar, the basic principle of radar, SAR parameters, discussion
on L-band airborne SAR and design of CP-SAR system will be presented. The theory
of microstrip antennas, proximity coupled feed method, single feed and dual feed CP microstrip
antennas, as well as the microstrip array will be explained in chapter 3. On chapter 4, the
methodology used to design and develop the CP microstrip antennas is described. The numerical
simulation and measurement results are presented in Chapter 5 along with explanations of
the results. Finally, the conclusion is given in Chapter 6, where the summary and future works
for this dissertation are also provided.
5

BIBLIOGRAPHY
Bibliography
[1] Birk M., Camus W., Valenti E. and McCandless W.J.,“ Synthetic aperture radar imaging
systems,” IEEE Aerospace and Electronic Systems Magazine, Vol. 10, No. 11, 15–23, 1995.
[2] Freeman A. and Saatchi S., “On the detection of faraday rotation in linearly polarized,
L-band SAR backscatter signatures,” IEEE Trans. On Geoscience And Remote Sensing,
Vol. 42, No. 8, 1607–1616, 2004.
[3] Harold M., Polarization in antennas and radar , John Wiley & Sons Inc., New York, 1986.
[4] Lee J., Schuler D.L. and Ainsworth T.L., “ Polarimetric SAR data compensation for terrain
azimuth slope variation,” IEEE Trans. On Geoscience And Remote Sensing, Vol. 38, No. 5,
2153–2163, 2000.
[5] Le Vine D.M., Jacob S.D., Dinnat E.P., de Matthaeis P. and Abraham S.,“ The in uence
of antenna pattern on faraday rotation in remote sensing at L-band,” IEEE Trans. On
Geoscience And Remote Sensing, Vol. 45, No. 9, 2737–2746, 2007.
[6] Maini A.K. and Agrawal V., Satellite technology: principles and applications , John Wiley
& Sons Inc., England, 2007.
[7] Mattia F., Toan T.L., Souyris J., Carolis G.D., Floury N., Posa F. and Pasquariello G.,
“The effect of surface roughness on multifrequency polarimetric SAR data,” IEEE Trans.
On Geoscience And Remote Sensing, Vol. 35, No. 4, 954–966, 1997.
[8] Meyer F.J. and Nicoll J.B., “ Prediction, detection, and correction of faraday rotation in
full- polarimetric L-band SAR data,” IEEE Trans. On Geoscience And Remote Sensing,
Vol. 46, No. 10, 3076–3086, 2008.
[9] Nemoto Y., Hideo N., Makoto O., Hitoshi M., Katsuhiko N. and Kaorui T., “Japanese
earth resources satellite-1 synthetic aperture radar,” Proceedings of the IEEE, Vol. 79,
No. 6, 800–809, 1991.
[10] Raney R.K.,“Hybrid-polarity SAR architecture,” IEEE Trans. On Geoscience And Remote
Sensing, Vol. 45, No. 11, 3397–3404, 2007.
6

BIBLIOGRAPHY
[11] Raney R.K., Luscombe A.P., Langham E.J., and Ahmed S.,“ Radarsat [SAR imaging],”
Proceedings of the IEEE, Vol. 76, No. 6, 839–849, 1991.
[12] Rignot J.M.,“ Effect of faraday rotation on L-band interferometric and polarimetric
synthetic- aperture radar data,” IEEE Trans. On Geoscience And Remote Sensing, Vol. 38,
No. 1, 383–390, 2000.
[13] Sri Sumantyo J.T., Wakabayashi H., Iwasaki A., Takahashi F., Ohmae H., Watanabe H.,
and et al“Development of circularly polarized synthetic aperture radar onboard microsatellite,”
Progress in Electromagnetic Research Symposium (PIERS), Page 382–385, Beijing,
China, March 2009.
[14] Strozzi T., Wiesmann A., Sharov A., Kouraev A., Wegmuller U. and Werner C., “Capabilities
of L-band SAR data for arctic glacier motion estimation” IEEE International
Geoscience and Remote Sensing Symposium, IGARSS, 3816–3819, Denver, Colorado, August
2006.
7

Chapter 2
Circularly Polarized Synthetic
Aperture Radar
2.1 The principles of radar
Radar is an acronym derived from the phrase RAdio Detection And Ranging and applies to
electronic equipment designed for detecting and tracking objects (targets) at considerable distances.
The basic concept of radar is relatively simple even though in many instances its
practical implementation is not. Radar operates by radiating electromagnetic energy and detecting
the echo returned from reflecting objects, which provides information about the target
(Figure 2.1). Radar signals are returned well by materials of considerable electrical conductivity,
especially by most metals, by seawater, and by wetlands.
Any target or any obstacle located within the illuminated sector will return part of the
energy it has received and the radar will detect this echo. The time (t R ) it takes a wave to
travel to, and from the target at a light speed (c) is a measure of the distance separating the
antenna from the target:
D = ct R
2
(2.1)
8

Figure 2.1: Basic principle of radar (Merrill, 2001).
Based on equation (2.1), the radar system also possible to determine the positions and movement
of the objects more precisely.
Initially, radar was developed to satisfy the needs of the military for surveillance and weapon
control. However, radar has seen significant employed in civil applications for the safe travel of
a transportation platform, the remote-sensing for environmental, weather and many other applications.
The block diagram of a fundamental radar system is shown in Figure 2.2. Modulator
Figure 2.2: Block diagram of fundamental radar system.
9

generates all the necessary timing pulses (triggers) for use in the radar and associated systems.
Its function is to ensure that all subsystems make up the radar system operates in a definite
time relationship with each other and that the intervals between pulses, as well as the pulses
themselves, are of the proper length. The transmitter generates powerful pulses of electromagnetic
energy at precise intervals. A duplexer is essentially an electronic switch that permits
a radar system to use a single antenna to both transmit and receive.
The receiver accepts
the weak r f echoes from the antenna system and routes them to the indicator. Output of the
receiver will be displayed on indicator subsystem.
2.2 Synthetic Aperture Radar
To improve radar image resolution, the antenna has to be lengthened. Since this cannot be
physically done, it takes a virtual solution to attain this goal. In June of 1951, Carl Wiley
described the use of Doppler frequency analysis to improve radar image resolution by using
platform movement and signal coherence to reconstruct a large antenna by calculation. As the
radar moves between two pulse transmissions, it is indeed possible to combine in phases all the
echoes and synthesize a very sizable antenna array. The technique today is known as Synthetic
Aperture Radar (SAR) (John C. and Robert N., 1991). Hence, SAR is a major advance in radar
remote sensing to improve azimuth resolution by synthesizing a long antenna. The different
geometric positions of the antenna elements are result of the moving platform now. The SAR
processor stores all the radar returned signals from a point P, as amplitudes and phases, for
the time period T from position A to C (Figure 2.3).
In general, the performance of a SAR sensor can be determined by its sensitivity, spatial
resolution in azimuth directions, image quality, ambiguities, and swath coverage (Pokuls et al.,
1998). Here, the antenna is the main key in realizing a fine SAR system, since an antenna
is a structure which serves as a transition between wave propagating in free space, and the
fluctuating voltages in the circuit to which it is connected. An antenna either receives energy
from an electromagnetic field or radiates electromagnetic waves produced by a high frequency
generator. Therefore, the antenna should be satisfy the SAR system requirement in term of
the size, radiation characteristic, beam width, pattern characteristic, and other electrical and
10

Figure 2.3: Basic principles of aperture synthesis.
mechanical. Figure 2.4 illustrated the simple geometry of the SAR system.
Figure 2.4:
Synthetic aperture radar geometry.
The key parameters of the SAR system are shown in Figure 2.4. This figure illustrates a SAR
sensor with the length of antenna (L a ) and high of the antenna (W a ), travelling along straight
11

line platform flight trajectory is known as the azimuth at speed v. The distance from the radar
antenna to its target area is called the slant range R m (m).
The radiation of the antenna
is pointed in a side-looking direction perpendicular to the platform tract. It will illuminate
the swath on the ground of width (W gmax ) that can be determined by the beam width of the
antenna in the elevation plane (Freeman et al., 2000), as follows
W gmax = θ elR m
cos θ i
= λR m
W a cos θ i
(2.2)
Where θ el is antenna elevation beam width (radian) and θ i is an incident angle (radian).
Other parameter should be considered in SAR sensor is the limiting resolution in the azimuth,
given by.
δ az ≥ L a
2
(2.3)
As we can see in equation (2.3), The azimuth resolution depends on antenna size and radar
wavelength.
Fine azimuth resolution is enhanced by taking advantage of the radar motion
in order to synthesize a larger antenna aperture. This is essentially because a tinier antenna
has a wider lobe, and a target will be seen for a longer time, which amounts to synthesize a
larger antenna, hence with a finer resolution. This simply states that the best possible azimuth
resolution can be achieved for side-looking SAR with antenna of length La, is half that antenna
length.
In synthetic aperture radar, azimuth ambiguities are caused by the aliasing of the Doppler
phase history of each target, which is sampled according to the sensor azimuth sampling frequency,
which equals the Pulse Repetition Frequency (PRF). PRF is the frequency at which
the radar transmits pulses is known as the pulse repetition frequency. The signal components
outside this frequency interval will fold back into the main part of the spectrum because the
azimuth spectrum repeats at PRF intervals and the desired signal band will be contaminated
by the ambiguous signals from adjacent spectra (Li F.K. et al., 1983) as illustrated in Figure
2.5. Theoretically, the azimuth ambiguity over signal ratio (AAR) < −20 can be formulated
(John C. and Robert N., 1991) as
12

Figure 2.5: Azimuth ambiguities: Doppler frequency histories of targets A and B are equal due
to aliasing about the sampling frequency of PRF (Li L.K. et al., 1983).
AAR =
∑ ∝
m=−∝
∫ Bp/2
−B p/2 G2 (f + m.P RF )df
∫ Bp/2
−B p/2 G2 (f)df
(2.4)
Where B p is azimuth processing bandwidth (Hz), G is antenna power gain, f is frequency (Hz),
respectively. The azimuth ambiguities can be avoided as long the minimum PRF requirement
is set to satisfy the following equation (Freeman, 2006).
P RF > 2V p
L a
(2.5)
Other type of ambiguity in the SAR sensor is range ambiguity. Range ambiguity in the
response is due to the radar return from two successive pulses overlaps at the receiver (see
Figure 2.6). This ambiguity will occur when the pulse repetition frequency (PRF) is set too
high.
13

Figure 2.6: Range ambiguities: Portions of the radar return from a previous pulse overlap the
return from the present pulse (Li F.K. et al., 1983).
2.3 Polarization of electromagnetic waves
The electromagnetic wave is composed of electric and magnetic lines of force. The electric field
determines the direction of polarization of the wave as shown in Figure 2.7.
The instantaneous
electric field associated with a general plane wave travelling in the +z direction can be
decomposed into x and y components.
E x (t, z) = E 1 cos(ωt − βz) (2.6)
E y (t, z) = E 2 cos(ωt − βz + δ) (2.7)
Where, E 1 and E 2 are amplitudes of the instantaneous electric field, ω is radian frequency, β
is phase constant, and δ is phase by which the y component leads the x component of electric
field. The resultant electric field is the vector combination of two components at each instant
of time.
⃗E(t, z) = ˆxE x (t, z) + E y (t, z) (2.8)
14

Figure 2.7: The LHCP wave shown at a fixed instant of time (Warren L., 1993).
By entering equation (2.6) and (2.7) to (2.8) for z = 0, the resultant vector is
⃗E(t, z) = ˆxE 1 cos ωt + ŷE 2 cos(ωt + δ) (2.9)
When the phase angle δ = 0 the equation (2.9) become
⃗E(t, z) = (ˆxE 1 + ŷE 2 ) cos(ωt + δ) (2.10)
By using the trigonometry identity the eq. (2.9) can be written as
⃗E y (t) = E 2 (cos ωt cos δ − sin ωt sin δ) (2.11)
The x component from equation (2.9) can be written as
cos ωt = E x
E 1
(2.12)
sin ωt = √ 1 − cos 2 ωt = √ 1 − (E x /E 1 ) 2 (2.13)
15

Substituting (2.12) and (2.13) into (2.11) leads to the following result that eliminates time
variation
(
Ex
⃗E y (t) = E 2 cos δ − √ )
1 − (E x (/E 1 )
E 2 sin δ
1
(2.14)
E y
E 2
− E x
E 1
cos δ = √ 1 − (E x (/E 1 ) 2 sin δ (2.15)
By squaring the both side we find
( ) 2 ( ) ( ) ( ) (
2 ( ) ) 2 Ey Ex Ey
Ex
− 2
cos δ + cos 2 Ex
δ = 1 − sin 2 δ (2.16)
E 2 E 1 E 2 E 1 E 1
and substituting the trigonometry identity
cos 2 δ = 1 − sin 2 δ (2.17)
( ) 2 ( ) ( ) ( ( ) ) 2 ( ) 2 Ey Ex Ey
Ex
− 2
cos δ = 1 − sin 2 Ex
δ − (1 − sin 2 δ) (2.18)
E 2 E 1 E 2 E 1 E 1
( ) 2 ( ) ( ) ( ) 2 Ey Ex Ey
Ex
− 2
cos δ = − + sin 2 δ (2.19)
E 2 E 1 E 2 E 1
The general equations for ellipse can be formulated as shown in (2.20)
( ) 2 ( ) ( ) ( ) 2 Ex Ex Ey
Ey
− 2
cos δ + = sin 2 δ (2.20)
E 1 E 1 E 2 E 2
The polarization ellipse can have any shape depend on the axial ratio parameter and orientation
(tilt). Based on (2.20) can be concluded for some case of δ as illustrated in Figure 2.8.
In case the amplitude of E 1 and E 2 is equal and components are in-phase quadrature
(δ = ±90 o ) the polarization will be circular. When δ = 90 o the wave is left-hand circularly
16

Figure 2.8: Polarization of the waves: (a) linearly polarized, (b) Left-Hand Ellipse Polarized
(LHEP) and (c) Right-Hand Ellipse Polarized (RHEP).
polarized (LHCP) and when δ = −90 o , the wave is right-hand circularly polarized (RHCP).
2.4 Circularly Polarized Synthetic Aperture Radar
Circularly polarized synthetic aperture radar (CP-SAR) is a SAR system, which transmits and
receives the microwave signals circularly. Basically, the concept of circularly polarized SAR is
similar to the linear polarization SAR. A linearly polarized antenna radiates wholly in one plane
containing the direction of propagation. In a circularly polarized SAR, the plane of polarization
rotates in a circle making one complete revolution during one period of the wave. The circular
polarization can be RHCP or LHCP depends on the vector is turning clockwise (CW) or counter
clockwise (CCW) in relation to the time.
In the current SAR system (LP-SAR), the basic parameter of the SAR sensor is determined
based on the half power beamdwith. The half power beamwidth is used in calculation for both
elevation and azimuth direction. In elevation direction, the half power beamwidth (θ ELP ) is
employed to calculate the covered area that can illuminate by sensor on the ground (swath
width). In azimuth direction, the half power beamwidth (θ ALP ) is needed to determined the
synthetic aperture length (L SA ) as shown in following formula (Samuel, et al., 2004)
L SA = Rθ ELP (2.21)
Here, R is the nearest target range in azimuth plane view as illustrated in Figure 2.9 (Samuel,
et al., 2004). The half power beamwidth in linear polarization (θ ELP and θ ALP ) is determined
17

Figure 2.9: Basic principle of aperture synthesis.
by the physical size of the antenna width, W and length, L (Mahafza BR., 2000) as formulated
θ ELP = λ W and θ ALP = λ L
(2.22)
In contrary with LP-SAR system, in CP-SAR system the SAR parameters are determined
by axial ratio characteristic of the antenna rather than only the physical size of the antenna.
To guarantee the SAR sensor operate in circular polarization, the axial ratio of the antenna
should have lower than 3 dB at the working frequency. In ideal condition, the perfect circular
polarization can be obtained when the axial ratio 0 dB. However, in many reports (Baharuddin
et al., 2011 and Yohandri et al., 2012), the axial ratio of the CP antenna very challenging to
be realized at 0 dB. Therefore, the 3-dB AR beamwidth is used to define the key parameters
of CP-SAR system as following (Rizki Akbar et al., 2010)
θ ECP ≤ 3 dB AR and θ ELP (2.23)
θ ACP ≤ 3 dB AR and θ ALP (2.24)
Where θ ECP is 3-dB AR beamwidth in elevation direction and θ ACP is 3-dB in azimuth direc-
18

tion. The definition and limitation of the 3-dB axial ratio is gived in Figure 2.10 (Rizki Akbar
et al., 2010).
Figure 2.10: Description of 3-dB axial ratio beamwidth for CP-SAR.
2.5 Design of CP-SAR onboard UAV
As already mentioned earlier, in order to design CP-SAR sensor onboard UAV and obtain high
quality and efficiency in these applications, the CP-SAR parameters, size and weight should be
thoroughly considered. In this section, the design of CP-SAR parameters and CP-SAR system
will be presented.
2.5.1 Design of CP-SAR parameters
The parameters CP-SAR onboard UAV can be derived based on the geometry system both
in slant range and azimuth view as illustrated in Figure 2.11.
In this figure, the essential
parameters of the CP-SAR are the airborne altitude, H, off nadir angle θ o , incidence angle
θ i , and the desired swath width and the maximum swath width W g and W gmax , respectively.
Parameter R m is the CP-SAR middle slant range, R a (R n ) is CP-SAR near slant range, R b (R f )
is CP-SAR far slant range for the desired swath width (maximum swath width). Based on the
19

CP-SAR geometry, the slant range , R n is determined by the value of θ ECP as following
R n =
H
cos (θ o − (θ ECP /2))
(2.25)
R m =
H
cos θ o
(2.26)
R f =
H
cos (θ o + (θ ECP /2))
(2.27)
Based on the figure 2.11a, the maximum swath width can be calculated as
Figure 2.11: CP-SAR geometry: (a) slant range view and (b) azimuth plane view.
W g =
√
R 2 f − H2 − √ R 2 n − H 2 (2.28)
Substitute equation (2.27) to (2.28)
W g =
√ ( )
√
2 ( ) 2
1
1
− 1 −
− 1 − W g
cos (θ o + (θ ECP )/2)
cos (θ o − (θ ECP )/2) H = 0 (2.29)
20

The resolution of the CP-SAR in azimuth direction δ ACP can be determined as
δ ACP =
Rλ
2L SACP
(2.30)
Where, the length of synthetic aperture of the CP-SAR L SACP = Rθ ACP , therefore
δ ACP =
λ
2θ ACP
(2.31)
As we can see, the azimuth resolution of the CP-SAR sensor is not depend on the physical
length of the antenna. In other hand, the ground resolution δ rgCP
as function of chirp pulse
bandwidth (B), light velocity (c) and incidence angle (θ i ) can be formulated as
δ rgCP =
c
2B sin θ i
(2.32)
Where the incidence angle θ i can be calculated
(
)
θ i = sin −1 R e + H
sin θ o
R e
(2.33)
For the airborne system, the high of the platform, H is relative small compare to the earth
radius (R e ≈ 6371 km), (H

Figure 2.12: The required (a) θ ECP and (b) θ ACP for 1 m image resolution.
In our project, the CP-SAR sensor resolution is targeted around 1 m. The resolution of the
SAR images can be determined from the swath width W g as
δ rgCP = W g
1000
(2.35)
δ ACP = δ rgCP (2.36)
Based on equation (2.35), the desired W g should be around 1 km. The θ ECP can be obtained
using equation (2.23) and (2.28). Since, the θ ACP is also set to 1 m (equation (2.36)), then the
required θ ACP
can also be estimated using equation (2.31). Figure 2.12 (Rizki Akbar et al.,
2010) illustrate the required θ ECP and θ ACP for θ o ranging from 40 o up to 60 o . By substituting
the antenna size (1.5m × 0.4 m) to equation (2.22), the maximum value of θ ECP
and θ ACP
equal to 29.78 o and 7.10 o . As shown in Figure 2.12, the smaller θ ECP
is obtained at higher
altitude, while the θ ACP almost constant (see Figure 2.12b). Due to the half power beam width
limitation in equation (2.22), for θ o equals to 40 o and 41 o , the maximum W g that could be
obtained are around 0.95 and 0.99 km. Therefore, the targeted resolution becomes 0.95 and
22

0.99 m (equation 2.36).
In azimuth direction, the processed Doppler bandwidth (B p ) will ensure only data that
capture by θ ACP is handled. This situation can be seen applying θ ACP parameters in the
following equations (John C. and Robert N., 1991 and George W., 1998).
B P = B D (2.37)
B D = 2vθ ACP
λ
(2.38)
Based on the parameters calculation, the 3-dB axial ratio beamwidth is the key parameter
in development of CP-SAR sensor.
This result also can give the opportunity to realize the
smaller SAR sensor. The main advantage of the CP-SAR sensor is the SAR parameter not
be determined by the physical size of the antenna, instead is determined by 3-dB axial ratio
beamwidth. The key parameters of the CP-SAR sensor onboard UAV are listed in table 2.1.
Table 2.1: Parameters of CP-SAR Onboard UAV.
Parameters
Value
Altitude
1-4 km
Frequency
1.27 GHz (L-Band)
Polarization
Tx : RHCP + LHCP
Rx : RHCP + LHCP
Image Size 50 km 2
Pulse Length 3.9 up to 23.87 µs
Pulse Bandwidth 61.14 up to 244.69 MHz
Off Nadir
40 o up to 60 o
Resolution
≈1m
Swadth width
1 km
PRF
1000 Hz
Peak Power
8.65 W up to 94.38 W
SNR
15 dB
Data Take Duration ≈31.70 minutes
The bandwidth requirement must be compatible with a low axial ratio (AR) (below 3 dB)
for ensuring transmitting/receiving circularly polarized waves. To satisfy the matching of input
impedance, the return loss must be smaller than 10 dB in this bandwidth range.
23

2.5.2 Design of CP-SAR system
2.5.2.1 UAV system
The UAV is an acronym for Unmanned Aerial Vehicle, commonly known as a drone, is an
aircraft with no pilot on board. UAVs can be a remote-controlled aircraft (e.g. flown by a
pilot at a ground control station) or can fly autonomously based on pre-programmed flight
plans or more complex dynamic automation systems.
With the arrival of new technologies
such as digital cameras, satellite navigation, and computer microprocessors, the capabilities of
the robotic aircraft have increased enormously. Their variety has increased as well, from mini-
UAVs the size of a model airplane to endurance UAVs with the wingspan of a modern jumbo jet.
Indeed, the capabilities of UAVs have developed to such a point that many air forces today are
wondering if the robotic aircraft will replace piloted aircraft in the next generation of warplanes
(Steven, 2008). In our mission, the UAV will be operated for remote-sensing application to
carry the CP-SAR sensor. The technical specification of the UAV is listed in Table 2.2.
Table 2.2: Basic Specification of UAV (JX-1).
Parameters
Value
Payload
Endurance
Altitude
Speed
Wing span
Sensor
25 kg
4-5 hours
1-4 km
100-120 kph
6 m
CP-SAR, LP-SAR, GPS, GPS-SAR, Cameras
General Configuration for UAV in RPV (Remotely Piloted Vehicles) mode consists of airframe,
engine servo and its electrical equipment, control modem, and R/C. The physical size
and profile and photograph of UAV are shown in Figure 2.13 and 2.14, respectively.
2.5.2.2 CP-SAR sensor
Basically, the hardware of UAV CP-SAR sensor can be divided into three parts: transmitter
subsystem (Tx), receiver subsystem (Rx) and antennas. A general block diagram of hardware
of CP-SAR sensor onboard UAV is shown in Figure 2.15.
In general, the circuit system part consists of two subsystems, which are transmitter and
24

Figure 2.13: Design of the UAV (unit in mm).
Figure 2.14: Unmanned aerial vehicles: (a) profile of UAV and (b) photograph.
25

BIBLIOGRAPHY
Figure 2.15: Design of CP-SAR sensor onboard UAV.
receiver subsystem. Filter components, such as low-pass filter (LPF) and band pass filter (BPF)
are implemented in both subsystems to ansure only the wanted radar signal is processed. In
the transmitter, the chirp signal is converted to the RF frequency (1.27 GHz as the center
frequency) by using up-converter mixer and then amplified by high-power amplifier (HPA)
to obtain adequate power transmit in the system. In the receiver part, the reflected signal is
amplified by Low Noise Amplifier (LNA) and later shifted to the baseband frequency by using a
down-converter mixer hence the received signal is ready to be sampled and digitized by analog
to digital converter (ADC) unit. The clock unit controls and manages the timing for many
components in the sensor such as the chirp generator, the signal processor component and also
the frequency generator. This timing function is correlated to transmitting and receiving timing
of the chirp pulse.
Bibliography
[1] Baharuddin M., Wissan V., Sri Sumantyo J.T. and Kuze H.,“Elliptical microstrip antenna
for circularly polarized synthetic aperture radar,” International Journal of Electronics and
Communications (IJEC), Vol. 65, No. 1, 62–67, 2011.
[2] Christensen E.L. and Dich M.,“SAR antenna design for ambiguity and multipath suppression,”
International Geoscience and Remote Sensing Symposium, IGARSS, Page 784–787,
Tokyo, Japan, August 1993.
26

BIBLIOGRAPHY
[3] Curlander J.C. and Robert M.N., Synthetic Aperture Radar : Systems and Signal Processing,
Wiley, New York, 1991.
[4] Dozier J., “Active microwave remote sensing,” URL http://fiesta.bren.ucsb.edu/
dozier/Class/ESM266/Slides/08-ActiveMicrowave.ppt., 2008.
[5] Freeman A., “On ambiguities in SAR design,” URL trs-new.jpl.nasa.gov/dspace/
bitstream/2014/39753/1/06-1066.pdf., 2006.
[6] Freeman A., Johnson W.T.K., Huneycutt B., Jordan R., Hensley S., Siquiera P. and Curlander
J., “The myth of the minimum SAR antenna area constraint,” IEEE Trans. On
Geoscience And Remote Sensing, Vol. 38, No. 1, 320–324, 2000.
[7] Keydel W., “From radar antenna to antenna radar perspectives for future antenna development
for airborne and spaceborne SAR,” In Joint RTO IST and SET Symposium on
Smart Antennas,, Page 119, Chester, United Kingdom, April 2003.
[8] Li F.K., and Johnson W.T.K., “Ambiguities in spaceborne synthetic sperture Radar systems,”
IEEE Trans. On Aerospace and Electronic System, Vol.AES 19, No. 3, 389–397,
1983.
[9] Mahafza B.R., Radar systems and analysis and design using Matlab, Chapman &
Hall/CRC., New York, 2000.
[10] Merrill I.S., Introduction To Radar Systems, Mcgraw-Hill., New York, 2001.
[11] Pokuls R., Uher J. and Pozar D.M., “Dual-frequency and dual polarization microstrip
antennas for SAR applications,” IEEE Trans. On Antenna Propagation, Vol. 46, No. 9,
1289–1296, 1998.
[12] Rizki Akbar P., Sri Sumantyo J.T. and Kuze H.,“CP-SAR UAV development,” International
Archives of The Photogrammetry, Remote Sensing And Spatial Information Science,
Page 203–208, Kyoto, Japan, 2010.
[13] Samuel W., McCandless Jr., and Jackson CR.,Synthetic aperture radar marine’s user manual,
Capter: Principles of synthetic aperture radar, Department of Commerce USA, Washington
DC., 2004.
27

BIBLIOGRAPHY
[14] Steven Z., Unmanned Aerialvehicles, Robotic Air Warfare 1917-2007 , Osprey, New York,
2008.
[15] Stimson G.W., Introduction to Airborne Radar, Second Edition, SciTech Publishing Inc.,
USA, 1998.
[16] Stutzman W.L., Polarization in Electromagnetic System, Artech House, USA, 1993.
[17] Wen Qin W., Jingye C. and Qicong P.,“Conceptual design of near-space synthetic aperture
radar for high-resolution and wide-swath imaging,” Elsevier Aerospace Science and
Technology, Vol. 13, No. 6, 340–347, 2009.
[18] Yohandri, Sri Sumantyo J.T. and Kuze H.,“A new triple proximity-fed circularly polarized
microstrip antenna,” International Journal of Electronics and Communications (IJEC),
Vol. 66, No. 5, 395–400, 2012.
28

Chapter 3
Circularly Polarized Microstrip
Antenna
3.1 Microstrip antenna
Nowadays, the microstrip antennas continuously develop to become one of the most attractive
antenna options in a wide range of modern microwave systems. This fast growth in microstrip
antenna applications and uses derived a continuous research effort for developing and improving
its characteristics (Pozar, 1992 and Garg et al., 2001). In general, a microstrip antenna can
attain a narrow frequency bandwidth at the expense of a low gain. As compared with conventional
microwave antennas, a microstrip antenna has additional advantages such as a compact
size, light weight, conformability to surfaces of substrates, low cost, and easier integration with
other circuits and versatility (Garg et al., 2001). From a designer point of view, a microstrip
antenna presents a wide range of options. The designer can vary the choice of the substrate
type, the antenna structure, type of perturbation and the feeding technique to achieve the
antenna design objective (Balanis, 2005 and Lo and Lee, 1993).
Basically, Microstrip antenna consists of four parts (Figure 3.1), radiating patch, dielectric
substrate, ground plane and feeding line. Where L is the length of the patch, W is the width
of the patch, h is the dielectric substrate height and ε r substrate relative permittivity. Many
29

developments of novel microstrip antenna configurations, basic properties, analytical models,
and design techniques for microstrip antennas and development of accurate analytical models
for the understanding the limitations of microstrip antennas have been produced from the
research work (Pozar, 1992, Mailloux et al., 1981 and Carver and Mink, 1981).
Figure 3.1: Basic geometry of microstrip antenna.
3.1.1 Radiating patch
A radiating patch consists of a very thin metallic sheet mounted on a dielectric substrate. The
shape of radiating patch can be designed in any shape such as square, rectangular, circular,
triangular, elliptical, semicircular, and annular ring shapes or any combination of these shapes
(Figure 3.2). Every shape has its own characteristics and is chosen to meet certain requirements.
The square, rectangular, and circular are the most popular shapes because they are the easiest
in analysis and fabrication.
3.1.2 Dielectric substrate
Substrate is the dielectric layer between the patch and the ground. There are a lot of substrate
material and specifications to choose from according to the antenna requirement. The substrate
has dual functions: electrically, and mechanically. Electrically is an integral part of the
transmission lines, circuits, and antennas. Mechanically is a supporter of the structure. The
most three factors specifying dielectric substrate is substrate height (0.003 4λ o ≤ h ≤ 0.05λ o ),
30

Figure 3.2: Various shapes of microstrip antennas (Girish and Ray, 2003).
dielectric constant (2.2 ≤ ε r ≤12) and dissipation factor (loss tan δ). As ε r gets higher in value
the antenna size gets smaller. Substrates that are thick with low dielectric constant are preferable
for enhancing efficiency, bandwidth and radiation in space. For a good overall efficiency
and a high circuit performance, the loss tan δ must be as low as possible. (typically loss tan
δ

to feed or excite a microstrip patch antenna include coaxial feeding, microstrip feeding, proximity
feeding and aperture feeding.
Feeding technique influences the input impedance and
characteristics of the antenna and also the size of the antenna.
3.2 Antenna basic parameter
Definitions of various parameters of antenna are necessary in order to describe the performance
of the antenna.
There is some basic antenna parameter should be considered in designing
the microstrip antennas, such as scattering parameters, antenna efficiency, gain, polarization,
bandwidth, radiation pattern, etc..
3.2.1 Scattering parameters
Scattering parameters (S-parameters) are a parameter set that relates to the traveling waves
that scattered or reflected when an n-port network is inserted into a transmission line. To define
the S-parameter, A two-port network schemetic is presented as shown in Figure 3.3.
Figure 3.3: The signal flow graph representation of a two-ports network network. (a) Defenition
of incident and reflected wave. (b) Signal flow graph (Pozar, 2005).
Here, (a 1 , a 2 ) are incident waves and (b 1 , b 2 ) are reflected waves. The S 11 is defined as the
input reflection coefficient with the output port terminated by a matched load (Z L = Z 0 sets
32

a 2 =0) and can be defined as
S 11 = Z 1 − Z 0
Z 1 + Z 0
(3.1)
Where Z 1 is input impedance at port 1 (Ω) and Z 0 is impedance characteristic (Ω).
3.2.2 Antenna efficiency
The efficiency of an antenna relates the power delivered to the antenna and the power radiated
or dissipated within the antenna. Radiation efficiency is defined as the ratio of the total power
radiated by an antenna to the net power accepted by the antenna from the connected transmitter.
A high efficiency antenna has most of the power present at the antenna’s input radiated
away. A low efficiency antenna has most of the power absorbed as losses within the antenna, or
reflected away due to impedance mismatch. For example, if a transmitter delivers 100 W into
an antenna having an efficiency of 80%, then the antenna will radiate 80 W as radio waves and
produce 20 W of heat. In order to radiate 100 W of power, one would need to use a transmitter
capable of supplying 125 W to the antenna.
3.2.3 Antenna Gain
The term antenna gain describes how much power is transmitted in the direction of peak
radiation when connected to a power source (Balanis, 2005). Gain is not a quantity which can
be defined in terms of a physical quantity such as the Watt or the Ohm, but it is a dimensionless
ratio. The antenna gain signifies the ratio of radiated power in a given direction relative to that
of an isotropic radiator which is radiating the total amount of electrical power received by the
antenna. In terms of U the antenna gain G in a specified direction can be calculated
G =
U(θ, φ)
P in /4π
(3.2)
where U(θ, φ) is radiation intensity (Watt/unit solid angle), and P in signifies the electrical
power received by the antenna from the transmitter(Watt).
However, in measurement, the
antenna gain is obtained from indirect measurement. The measured gain is relative gain that
33

defines as the ratio of the power gain in a given direction to the power gain of a reference
antenna in its reference direction.
3.2.4 Polarization
Polarization is defined as the orientation of the electric field of an electromagnetic wave. Polarization
is in general described by an ellipse. Two often used special cases of elliptical polarization
are linear polarization and circular polarization. The polarization of an electromagnetic wave
can be categorized by using axial ratio (AR) parameter. Axial ratio is the ratio of the major
axis to the minor axis of the polarization ellipse which commonly stated in dB unit. Figure 3.4
shows the elliptical polarization which the elliptical curve is formed by the tip of the instantaneous
electric field vector. Here, ε and τ are the ellipticity angle and the tilt angle of the
Figure 3.4: Parameters in electromagnetic wave polarization.
microwave signal, respectively. The range value of ε is from -45 o up to 45 o and τ angle is from
0 o up to 180 o . Based on Figure 3.4, the ellipticity as a function of AR is defined as
ε = cot −1 (−R) (3.3)
The polarization of a microwave signal is categorized based on their axial ratio value (R).
34

The absolute value of the axial ratio can be formulated as (Warren L., 1993)
|R| =
major axis length
minor axis length = OA
OB ≥ 1 (3.4)
As can be seen from Figure 3.4, the value of OA and OB are indicating the maximum and
minimum value of the electric field amplitude of the microwave signal. The R is equal to 1
for perfect circular polarization and infinite for linear polarization. In the between of 1 and
infinite, electromagnetic wave is classified as elliptical polarization. Practically, R is represented
in decibel (dB) unit as
R(dB) = 20 log |R| (3.5)
Here, factor 20 is used since |R| is the ratio of field quantities. The sign of R describes the
sense of the microwave signals. Here, sense is the rotation direction of electric field vector over
time while it propagates. In terms of polarization sense, the sign of R is positive for RHCP
and negative for LHCP.
3.2.5 Bandwidth
The bandwidth of an antenna refers to the range of frequencies over which the antenna can
properly radiate or receive energy. The bandwidth can also be described in terms of percent
bandwidth, because the percent bandwidth is constant relative to frequency. In linear antennas,
the bandwidth of an antenna is measured from the reflection coefficient characteristic.
For
circularly polarized antennas, the antenna bandwidth is determined by both the reflection
coefficient and the axial ratio. The bandwidth of an antenna can be defined as
Bandwidth(%) = f 2 − f 1
f c
× 100 (3.6)
where f 2 is the upper frequency, f 1 is the lower frequency, and f c is the center frequency.
35

3.2.6 Radiation pattern
In the field of antenna, the antenna radiation pattern is a measure of its power or radiation
distribution with respect to a particular type of coordinates. This power variation as a function
of the arrival angle is observed in the far field. It is typically represented by a three dimensional
graph, or polar plots of the horizontal and vertical cross sections.
The pattern of an ideal
isotropic antenna, which radiates equally in all directions, would look like a sphere.
Many
nondirectional antennas, such as monopoles and dipoles, emit equal power in all horizontal
directions, with the power dropping off at higher and lower angles (omnidirectional pattern).
Based on the radiation pattern, the beamwidth (usually in 3dB) and side lobe level of the
antenna can be obtained.
3.2.7 Current density and distribution
The performance designed antenna can be evaluated using average current density on the surface
of the antenna. The degradation color in the antenna shows the distribution intensity and a
good antenna has a high intensity around the center of the antenna. The vector electric current
distribution on the surface of the antenna also aid in design by displaying the animation and
different size of vectors. In array configuration, the current distribution can be used to observe
the power distribution to each elements of the antenna.
3.3 Circularly polarized microstrip antenna
Nowadays, circular polarization is very important in the antenna design industry. It eliminates
the importance of antenna orientation in the plane perpendicular to the propagation direction.
It gives much more flexibility to the angle between transmitting and receiving antennas, also it
enhances weather penetration and mobility (Hayt and Buck, 2001 and Rahmani et al., 2009).
The quality of the circularly polarized wave is generally specified by an axial ratio value (Hall
and Dahele, 1997). It is used in a bunch of commercial and militarily applications. However, it
is difficult to build good circularly polarized antenna (Milligan, 2005). For circular polarization
to be generated in microstrip antenna two modes equal in magnitude and 90 out of phase are
36

equired (Tamakuma and Iwasak, 2003 and James and Hall, 1989). Microstrip antenna on its
own doesnt generate circular polarization; subsequently, some changes should be done to the
patch antenna to be able to generate the circular polarization (Suzuki et al., 1987). The most
commonly used feeding techniques in circular polarization generation are dual feed and single
feed. However, with some phase adjustment, the multiple feed also can be employed to generate
the circular polarization antenna (Yohandri et al., 2012).
3.3.1 Dual feed circularly polarized microstrip antenna
A circularly polarized microstrip antenna can be realized by exciting two orthogonal modes
with equal magnitudes, which are in phase quadrature. The simplest way to obtain CP is using
two feeds at orthogonal as the 90 o phase shift between the fields in the microstrip antenna.
The two feed points are chosen perpendicular to each other. Figure 3.5 show the dual feed
configuration for square microstrip antenna (SMSA) and circular microstrip antenna (CMSA).
Figure 3.5: Dual-feed (a) SMSA and (b) CMSA (Girish and Ray, 2003).
With the help of external polarizer, the microstrip patch antenna is fed by equal in magnitude
and orthogonal feed. Several transmission lines can be used to realize the power divider
network. The quadrature phase difference can be obtained by feeding the element with a 90 o
power divider or 90 o hybrid (Balanis, 2005) as shown in Figure 3.6.
37

Figure 3.6: Rectangular patch arrangements for circular polarization: (a) Through a power
divider and (b) Through a 90 o hybrid (Balanis, 2005).
3.3.2 Single feed circularly polarized microstrip antenna
Single feeding techniques are very common with microstrip antennas as they are simple, easy to
manufacture, low in cost and compact in structure as shown in Figure 3.7. It eliminates the use
of complex hybrid polarizer, which is very complicated to be used in an antenna array (Haneishi
et al., 1982). In order to achieve circular polarization using only single feed two degenerate into
modes should be exited with equal amplitude, and in-phase quadrature have to be induced.
Since basic shapes microstrip antenna produce linear polarization, there must be some changes
in the patch design to produce circular polarization. Perturbation segments are used to split
the field into two orthogonal modes with equal magnitude and 90 o phase shift.
To realize
Figure 3.7: Single-feed arrangements for circular polarization microstrip antennas: (a) Square,
(b) Circular and (c) Elliptical.
the circular polarization using single feed, the two orthogonal mode in patch surface must be
set properly to get same amplitude but 90 o out of phase with respect to other mode. Figure
3.8 shows a nearly square patch antenna having physical dimensions a and b with b > a. The
circular polarization either RH or LH in this path can be generated by fed along one of the two
38

Figure 3.8: Single feed circularly polarized microstrip antennas; (a) Nearly square patch and
(b) Amplitude and phase of the two modes (Lee, S.K., et al., 2005).
39

diagonals axis. The area of the perturbation segment (∆S) must be such that the two modes
satisfy the conditions for circular polarization as shown. Illustration of detuning degenerates
into the modes of a symmetrical patch by perturbation segments is illustrated in Figure 3.8b.
Several techniques were used to achieve circular polarization in single fed microstrip antenna.
Among these techniques: fractal boundary (Rao et al., 2009), square patch with shaped slots
(Nasimuddin et al., 2008), embedding cross slot in metallic patch or the ground plane (Iwasaki,
1996), staking antennas (Wang et al., 2008), and truncated edges patches (Yohandri et al.,
2011).
3.3.3 Triple feed circularly polarized microstrip antenna
The configuration of triple feed circular microstrip antenna is depicted in Figure 3.9. A symmetrical
configuration should be considered to achieve the good circular polarization (Chen
et al., 2010).
The circular patch with radius r is excited through three feeding points that
distributed averagely on the circumference with the same distance d from the patch center.
Figure 3.9: Geometry design of triple feed circular microstrip antenna.
Unlike dual-fed CP antennas, the feeding positions of this antenna do not satisfy the orthogonal
conditions in structure and result in three non-orthogonal linear polarizations. In this
situation, the electric field of a wave traveling in z−direction consists of three linearly polarized
components in directions 1, 2, and 3, respectively, as illustrated in Figure 3.10.
40

Figure 3.10: Linearly polarized wave on triple feed circular microstrip antenna.
In case the amplitudes of the three linearly polarized wave is same (E o ), these electric field
components for each direction as a function of time and position are given by
E 1 = E 0 sin(wt − βz) (3.7)
E 2 = E 0 sin(wt − βz − ϕ) (3.8)
E 3 = E 0 sin(wt − βz + ϕ) (3.9)
Here, ϕ is a phase difference between the components. Combining (3.7-3.9) gives the instantaneous
total vector, which can be expressed in terms of two orthogonal linearly polarized
components, one in x−direction and one in y−direction.
E = ⃗ E 1 + ⃗ E 2 + ⃗ E 3 (3.10)
E = ˆx(E 3 cos θ − E 2 cos θ) + ŷ(E 1 − E 3 sin θ − E 2 sin θ) (3.11)
41

By substitute eq. (3.7-3.9) to (3.11), the amplitude of linearly polarized components in x and
y directions at z = 0 can be expressed as
E x = E 0 [sin(wt + ϕ) cos θ − sin(wt − ϕ) cos θ] (3.12)
E y = E 0 [sin(wt) − sin(wt + ϕ) sin θ − sin(wt − ϕ) sin θ] (3.13)
Expanding (3.12) and (3.13) yields
E x = E 0 [2 cos(wt) sin ϕ cos θ] (3.14)
E y = E 0 [sin(wt) − 2 sin(wt) cos ϕ sin θ] (3.15)
From (3.14) and (3.15) we have
cos(wt) 2 =
E 2 x
E 2 0 (2 sin ϕ cos θ)2 (3.16)
sin(wt) 2 =
E 2 y
E 2 0 (1 − 2 cos ϕ sin θ)2 (3.17)
Combining (3.16) and (3.17) eliminates wt, we obtain
E 2 0
E 2 x
(2 sin ϕ cos θ)2
+
E
2 y
E0 2 = 1 (3.18)
(1 − 2 cos ϕ sin θ)2
In order to achieve circular polarization, two denominators in (3.18) should be equal yielding
(2 sin ϕ cos θ) 2 = (1 − 2 cos ϕ sin θ) 2 (3.19)
Based on geometry design of the antenna, the θ = 30 o is introduced into eq. (3.19). The phase
differences for the three feed are ϕ = ± 120, corresponding, left and right circularly polarized
42

waves respectively.
3.4 Microstrip Array Antenna
In single element mode, the radiation pattern of a microstrip antenna is relatively wide, and each
element provides low values of directivity (gain). To increase the performance of the microstrip
antenna, the several single antennas is connected and arranged in a regular structure to form
a single antenna. Antenna arrays are able to produce radiation patterns that combined, have
characteristics that a single antenna would not. For example, in array configuration, the gain
of the antenna will be higher and the directivity increases, therefore the beam-width becomes
narrower. The total field of the array is determined by the vector addition of the fields radiated
by the individual elements. In an array antenna, the fields from the individual elements add
constructively in some directions and destructively (cancel) in others. The major advantage of
antenna arrays over a single antenna element is their electronic scanning capability; that is, the
major lobe can be steered toward any direction by changing the phase of the excitation current
at each array element (phased array antennas) Ideally this can be accomplished, but practically
it is only approached. The performance of an array antenna is determined by some parameters
such as geometry, distance between element, amplitude, phase, and radiation pattern of each
individual element (Balanis, 2005).
The array antenna can be arranged in a linear, planar, or volume array form. Linear arrays
consist of equally spaced elemental radiators laid out in a straight line, while two-dimensional
planar arrays consist of radiators oriented on a geometric grid in a plane. Rectangular arrays
may be thought as a set of linear arrays placed next to each other, equally spaced, forming
the two-dimensional array.
A linear array may also be wrapped around a curved surface,
usually a circle or a cylinder.
Linear, planar, and conformal arrays can be designed with
either a fixed main beam, or a scanned beam which is rapidly positioned in space by means
of electromechanical or electronically actuated devices connected in the feed lines behind the
array radiators.
43

3.4.1 Linear array
Among the different geometries of antenna arrays, the one-dimensional linear array is the simplest
and one of the most practical arrays. In a uniform array, all the elements have identical
magnitudes and each succeeding element has a β progressive phase lead current excitation relative
to the preceding one (β represents the phase by which the current in each element leads
the current of the preceding element). Figure 3.11 shows an N-element uniform linear array of
isotropic sources positioned along the z-axis. From Figure 3.11, the array factor (AF) can be
Figure 3.11: Linear array geometry (Balanis, 2005).
obtained by considering the elements to be point sources which is given by (Balanis, 2005),
N∑
AF = e j(n−1)ψ where ψ = kd cos θ + β (3.20)
n=1
The array factor of (2.20) can also be expressed in an alternate, compact and closed form
whose functions and their distributions are more recognizable. This is accomplished as follows.
Multiplying both sides of (2.20) by e jψ , it can be written as
(AF )e jψ = e jψ + e j2ψ + e j3ψ + ... + e j(N−1)ψ + e jNψ (3.21)
44

Subtracting (2.24) from (2.25) reduces to
AF (e jψ − 1) = (−1 + e jNψ ) (3.22)
This can also be written as
AF =
[ e jNψ ] [
− 1
e
e jψ = e j[(N−1)/2]ψ j(N/2)ψ − e −j(N/2)ψ ]
− 1
e j(1/2)ψ − e −j(1/2)ψ
(3.23)
[ (
sin
N
AF = e j[(N−1)/2]ψ 2 ψ)
]
sin ( 1
2 ψ)
(3.24)
If the reference point is the physical center of the array, the array factor of (3.28) reduces to
AF =
[
sin
( N
2 ψ)
sin ( 1
2 ψ) ]
(3.25)
The maximum value of (3.25) is equal to N. To normalize the array factors so that the maximum
value of each is equal to unity, (3.24) can be written in normalized form as
(AF ) n = 1 N
[
sin
( N
2 ψ)
sin ( 1
2 ψ) ]
(3.26)
For small values of ψ, the above expression can be approximated by
(AF ) n ≃
[
sin
( N
2 ψ)
]
N
2 ψ
(3.27)
With the array factor known, the total field can be formed by multiplying the array factor and
the field of a single element. This is known as pattern multiplication rule and it applies only
for arrays of identical elements.
3.4.2 Planar array
In addition to placing elements along a line (to form a linear array), individual radiators can
be positioned along a rectangular grid to form a rectangular or planar array. Planar arrays
45

provide additional variables which can be used to control and shape the pattern of the array.
Planar arrays are more versatile and can provide more symmetrical patterns with lower side
lobes. In addition, they can be used to scan the main beam of the antenna toward any point
in space (Balanis, 2005).
Figure 3.12: Geometry of a planar array.
The radiation pattern of a two dimensional planar array can be written as the product of
radiation pattern in the two planes which contain the principal axes of the antenna. Based on
the Figure 3.12, the field point is located at a range of and the point has angles θ to the z axis
and φ to the x axis. In the coordinate system, the field point is located at
(x f , y f ) = (r sin θ cos φ, r sin θ sin φ) (3.28)
A planar array with equal element spacing of d x and d y in principal planes is referred to
as having a rectangular grid or grid is called as square (d x = d y ). The position vector of filed
point (r mn ) relative to (mn th ) can be determined as
r mn =
√
(x f − md x ) 2 + (x y − nd y ) 2 ≃ r + md x sin θ cos φ + nd y sin θ sin φ (3.29)
46

The array factor of the planar antenna can be written as (Stutzman WL, et al., 1998)
N∑ M∑
AF (θ, φ) = I mn e j(βˆr.rmn+αmn) (3.30)
n−1 m−1
The double summation is useful in geometries that employ row and columns. The phase
term alpha mn is that portion of excitation current phase used to scan the main beam. If all
row and column have identical current distribution, the current I mn = I xm I yn and array factor
can be formulated as
N∑
M∑
AF (θ, φ) = I yn e j(βyn sin θ sin φ) . I xm e j(βxm sin θ cos φ) (3.31)
n−1
m−1
The Radiation pattern of an antenna can be defined as the variation in field intensity as
a function of position or angle.
In planar array, the beam of the radiation pattern can be
steered in two dimensions. In polar two dimensional coordinates an incremental area dA on the
surface of a sphere is the product of the length r dθ direction (latitude) and r sin θ dφ direction
(longitudinal), as shown in Figure 3.13.
Figure 3.13: Polar coordinates showing incremental solid angle dA = r 2 dΩ on the surface of a
sphere of radius r where dΩ = solid angle subtended by the area dA.
47

The beam area or solid angle or Ω A of an antenna is given by the integral of the normalized
power pattern over a sphere (4πsr).
Ω A =
∫ φ=2π ∫ θ=π
φ=0 θ=0
P n (θ, φ) sin θ dθ dφ (3.32)
and
∫ ∫
Ω A =
P n (θ, φ)dΩ (sr) Beam area (3.33)
Where dΩ = sin θ dθ dφ, (sr) and 1 sr = 1rad 2 . The beam are Ω A is the solid angle through
which all of the power radiated by antenna would stream if P (θ, φ) maintained its maximum
value over Ω A and was zero elsewhere. Therefore the power radiated = P (θ, φ)Ω A watts.
3.4.3 Dolph-Chebyshev array
In the linear array, there is also the uniform spacing but nonuniform amplitude distribution
arrays. In these non-uniform linear arrays, the amplitude as well as the phase can be used to
control the formation and distribution of the total array factor. The non-uniform linear arrays
can be realized the using many methods, one of the methods is Dolph-Chebyshev.
Chebyshev method is primarily a compromise between uniform and binomial arrays.
Dolph-
For a
given side lobe level, Dolph Chebyshev arrays produce the smallest beamwidth between the
first nulls. Conversely, for a given beamwidth between the first nulls, Chebyshev design leads
to the smallest possible side lobe level. Because of this useful property, Chebyshev is able to
optimize the relationship between beamwidth and the side lobe level, producing both a narrow
main beam as well as low side lobes.
The general approach for Chebyshev method is to equate the array polynomial to the Chebyshev
polynomial. These relations are valid only in the region −1 ≤ z ≤ 1 and all polynomials,
of any order, pass through the point (1,1). Within the range −1 ≤ z ≤ 1, the polynomials have
values within -1 to +1 with maxima of +1 and minima of -1. This property of equal ripple
or side lobe is a distinctive characteristic of Chebyshev method. For |z| > 1, the Chebyshev
polynomials are related to the hyperbolic cosine functions, defining the main beam (Balanis,
48

Figure 3.14: Non-uniform amplitude arrays of even and odd number of elements (Balanis, 2005).
49

2005).
An array of even number of isotropic elements 2M (where M is an integer) is positioned
symmetrically along the z-axis, shown in Figure 3.14a. The separation between the elements
is d, and M elements are placed on each side of the origin. Assuming the amplitude excitation
is symmetrical about the origin, the array factor for non-uniform amplitude broadside array is
(Balanis, 2005),
(AF ) 2M = 2
M∑
n−1
[ ]
(2n − 1)
a n cos kd cos θ
2
(3.34)
In normalized form,
(AF ) 2M =
M∑
n−1
[ ]
(2n − 1)
a n cos kd cos θ
2
(3.35)
where a n ’s are the excitation coefficients of the array elements.
If the total number of isotropic elements of the array is odd 2M+1, as shown in Figure
3.14b, the array factor can be written as,
which inn ormalized form reduces to
M+1
∑
(AF ) 2M+1 = 2 a n cos[(n − 1)kd cos θ] (3.36)
n−1
(AF ) 2M+1 =
M+1
∑
n−1
The amplitude excitation of the center element is 2a 1 .
a n cos[(n − 1)kd cos θ] (3.37)
Equations (3.35) and (3.37) can be written in normalized form as
M∑
(AF ) 2M (even) = a n cos[(2n − 1)u] (3.38)
n−1
(AF ) 2M+1 (odd) =
M+1
∑
n−1
a n cos[2(n − 1)u] (3.39)
50

where
u = πd
λ
cos θ (3.40)
Referring to (3.38) and (3.39), the array factor of an array of even or odd number of elements
with symmetric amplitude excitation is nothing more than a summation of M or M + 1 cosine
terms. The largest harmonic of the cosine terms is one less than the total number of elements of
the array. Each cosine term, whose argument is an integer times a fundamental frequency, can
be rewritten as a series of cosine functions with the fundamental frequency as the argument.
That is,
m = 0 cos(mu) = 1
m = 1 cos(mu) = cos u
m = 2 cos(mu) = cos(2u) = 2 cos 2 u − 1
m = 3 cos(mu) = cos(3u) = 4 cos 3 u − 3 cos u
m = 4 cos(mu) = cos(4u) = 8 cos 4 u − 8 cos 2 u + 1
m = 5 cos(mu) = cos(5u) = 16 cos 5 u − 20 cos 3 u + 5 cos u
m = 6 cos(mu) = cos(6u) = 32 cos 6 u − 48 cos 4 u + 18 cos 2 u − 1
m = 7 cos(mu) = cos(7u) = 64 cos 7 u − 112 cos 5 u + 56 cos 3 u − 7 cos u
m = 8 cos(mu) = cos(8u) = 128 cos 8 u − 256 cos 6 u + 160 cos 4 u − 32 cos 2 u + 1
m = 9 cos(mu) = cos(9u) = 256 cos 9 u − 576 cos 7 u + 432 cos 5 u − 120 cos 3 u + 9 cos u
For z = cos u the series cosine function can be written as
m = 0 cos(mu) = 1 = T 0 (z)
m = 1 cos(mu) = z = T 1 (z)
m = 2 cos(mu) = 2z 2 − 1 = T 2 (z)
m = 3 cos(mu) = 4z 3 − 3z = T 3 (z)
m = 4 cos(mu) = 8z 4 − 8z 2 + 1 = T 4 (z)
m = 5 cos(mu) = 16z 5 − 20z 3 + 5z = T 5 (z)
m = 6 cos(mu) = 32z 6 − 48z 4 + 18z 2 − 1 = T 6 (z)
m = 7 cos(mu) = 64z 7 − 112z 5 + 56z 3 − 7z = T 7 (z)
51

BIBLIOGRAPHY
m = 8 cos(mu) = 128z 8 − 256z 6 + 160z 4 − 32z 2 + 1 = T 8 (z)
m = 9 cos(mu) = 256z 9 − 576z 7 + 432z 5 − 120z 3 + 9z = T 9 (z)
The recursion formula for Tschebyscheff polynomials is
T m (z) = 2zT m−1 (z) − T m−2 (z) (3.41)
It can be used to find one Tschebyscheff polynomial if the polynomials of the previous two
orders are known.
Bibliography
[1] Balanis C.A., Antenna Theory Analysis and Design 3rd edition, John Wiley & Sons, USA,
2005.
[2] Carver K.R., Mink J.W., Siquiera P. and Curlander J., “Microstrip antenna technology,”
IEEE Trans. On Antenna Propagation, Vol. 29, No. 1, 2–24, 1981.
[3] Chen L., Fu S.Z., Yong C.J., Fan Z. and Xin X., Siquiera P. and Curlander J., “A threefed
microstrip antenna for wideband circular polarization,” IEEE Antennas and Wireless
Propagation Letter, Vol. 9, 359–362, 2010.
[4] Cheston TC. and Frank J., Phased Array RadarAntennas. Radar Handbook, 2nd Ed.,
McGraw Hill, 7.1-7.82.
[5] Garg R., Bhartia P., Bahl I. and Ittipiboon A., Microstrip Antenna Design Handbook,
Arteck House, Norwood, MA, 2001.
[6] Girish K. and Ray K.P., Broadband Microstrip Antennas, Arteck House, Norwood, MA,
2003.
[7] Hall P.S. and Dahele J.S., Dual and Circularly Polarized Microstrip Antennas, in Advances
in Microstrip and Printed Antennas, K.F. Lee and W. Chen (Eds.), John Wiley & Sons
Inc., New York, 1997.
52

BIBLIOGRAPHY
[8] Haneishi M., Nambara T. and Yoshida S., “Study on ellipticity properties of single-feedtype
circularly polarised microstrip antennas,” Electronics Letters, Vol. 18, No. 5, 191–193,
1982.
[9] Hayt W.H. and Buck J.J., Engineering Electromagnetic 6th edition, McGraw-Hill, New
York, 2001.
[10] Iwasaki H., “A circularly polarized small-size microstrip antenna with a cross slot,” IEEE
Trans. On Antenna Propagation, Vol. 44, No. 10, 1399–1401, 1996.
[11] James J.R. and Hall P.S., Handbook of Microstrip Antennas, Vol. I , Short Run Press Ltd.,
England, 1989.
[12] Kuo Y.L. and Wong K.L.,“A circularly polarized microstrip antenna with a photonic band
gap ground plane,” Proceedings of Microwave Conference, AMPC, Asia-Pacific, Vol. 2,
Page 647–650, Taipei, Taiwan, August 2002.
[13] Lee S.K., Sambell A., Korolkiewicz E. Loh S.F., Ooi S.F., and Qin Y., “A design procedure
for a circular polarized, nearly square patch antenna,” Microwave Journal, 2005.
[14] Li T.W., Lai C.L. and Sun J.S., “Study of dual-band circularly polarized microstrip antenna,”
Proceedings of the European Conference on Wireless Technology, Page 79–80, 2005.
[15] Lo Y.T. and Lee W., Antenna Handbook, Chapman & Hall, New York, 1993.
[16] Mailloux R.J., McIlvenna J.F., dan Kernweis N.P., Siquiera P. and Curlander J., “Microstrip
array technology,” IEEE Trans. On Antenna Propagation, Vol. 29, No. 1, 25–37,
1981.
[17] Nasimuddin, Chen Z.N. and Qing X.,“Single fed circularly polarized microstrip antenna
with c-slot,” Proceedings of the Microwave Conference, Page 1–4, Macau, December 2008.
[18] Pozar D.M., “Microstrip antennas,” Proceedings of the IEEE, Vol. 80, No. 1, 79–91, 1992.
[19] Pozar D.M., Microwave Engineering 2nd edition, John Wiley & Sons, USA, 1998.
[20] Pozar D.M. and Kaufman B., “Increasing the bandwidth of a microstrip antenna by proximity
coupling,” Electronics Letters, Vol. 23, No. 8, 368–369, 1987.
53

BIBLIOGRAPHY
[21] Rahmani M., Tavakoli ., Amindavar H.R., Reza A.M. and Dehkhoda P.,“Chalipa, a novel
wideband circularly polarized microstrip fractal antenna,” Proceedings of the 3rd European
Conference on Antennas and Propagation, EuCAP, Page 2389–2392, Berlin, June 2009.
[22] Simba A.Y., Yamamoto M., Nojima T. and Ito K., “Circularly polarised proximity-fed
microstrip antenna with polarisation switching ability,” Microwaves, Antennas and Propagation,IET
, Vol. 1, No. 3, 658–665, 2007.
[23] Stutzman W.L., Polarization in Electromagnetic System, Artech House, USA, 1993.
[24] Tamakuma T. and Iwasak H.,“A small size circularly polarized annular microstrip antenna,”
Proceedings of the International Symposium IEEE Antennas and Propagation Society,
Vol. 2, Page 716–719, August 2003.
[25] Wang Y., Feng J., Cui J. and Yang X.,“A dual-band circularly polarized stacked microstrip
antenna with single-fed for GPS applications,” Proceedings of the 8th International
Symposium on Antennas Propagation and EM Theory, ISAPE,, Page 108–110, Kunming,
November 2008.
[26] Waterhouse R.B., Microstrip Patch Antennas : A Designer’s Guide, Kluwer Academic
Publishers,, New York, 2003.
[27] Yohandri, Sri Sumantyo J.T. and Kuze H.,“A new triple proximity-fed circularly polarized
microstrip antenna,” International Journal of Electronics and Communications (IJEC),
Vol. 66, No. 5, 395–400, 2012.
[28] Yohandri, Wissan V., Firmansyah I., Rizki Akbar P., Sri Sumantyo J.T. and
Kuze H.,“Development of circularly polarized array antenna for synthetic aperture radar
sensor installed on UAV,” Progress in Electromagnetics Research C , Vol. 19, 119–133, 2011.
54

Chapter 4
Methodology
4.1 Design methodology
Generally, there are some steps should be done in developing the circularly polarized microstrip
antenna. The simple flow chart of the steps is shown in Figure 4.1.
4.2 Design procedure and electromagnetic modeling
The microstrip antenna will be designed to meet the specification requirement of the CP-SAR
system as discussed in chapter 3. In order to achieve the good characteristics of the antenna,
two simulation softwares are employed in electromagnetic modeling of a microstrip antenna. For
simple antenna geometry, the Method of Moment (MoM) is valid enough to do the numerical
calculation of the antenna.
However, for 3D antenna design, the complex geometry of the
antenna is suitable to be analyzed using Finite Element Method (FEM).
4.2.1 Design and analysis using Method of Moment (MoM)
The method of moments (MoM) was a very useful numerical technique for solving integral,
differential and Integra-differential equations. IE3D, from Zeland Software, Inc., is a full wave,
method of moments based electromagnetic simulator for analyzing and optimizing microwave
electronics component, including planar and 3D structures in a multilayer dielectric environment
55

Figure 4.1: Design methodology.
56

(Zeland, 2006). The software has a menu driven graphic interface for model generation with
automatic meshing and uses a field solver.
4.2.2 Design and analysis using Finite Element Method (FEM)
NSYS HFSS software is the industry-standard simulation tool for 3-D full-wave electromagnetic
field simulation and is essential for the design of high-frequency and high-speed component
design. HFSS offers multiple state-of the-art solver technologies based on finite-element method.
With HFSS, engineers can extract scattering matrix parameters (S, Y, Z parameters), visualize
3-D electromagnetic fields (near- and far-field) and generate ANSYS Full-Wave SPICE models
that link to circuit simulations. Each HFSS solver is based on a powerful, automated solution
process where users are only required to specify geometry, material properties and the desired
output. From there HFSS will automatically generate an appropriate, efficient and accurate
mesh for solving the problem using the selected solution technology (Ansoft, 2009).
4.3 Fabrication of proposed antenna
The new triple-fed circularly polarized microstrip antenna, LHCP and RHCP microstrip array
antenna have been fabricated to verify the simulated results. Careful and precise fabrication
process is required to guarantee the radiating behavior similar to the simulated model. The
stages for fabrication are as follows: (1) Microwave Artwork; (2) Etching; (3) Bonding. There
are two techniques implemented in the antenna fabrication. Firstly, proposed antenna can be
fabricated using a high-precision milling machine (Seven Mini) is used for the fabrication process
that involves several steps such as milling, drilling and edge cutting. Secondly, the antenna is
laminated using dry film and exposure under UV lamp. The etching process is implemented
for both these techniques to remove unnecessary copper part of the microstrip substrate. After
installing the plastic screws, then the antenna is ready for measurement. The description of
each stage can be found in Appendix C. The flow chart of the fabrication process of the antenna
can be seen in Figure 4.2.
57

Figure 4.2: Flow chart of the antenna fabrication.
58

4.4 Antenna measurement
The testing and evaluation of the antenna parameters is performed in antenna ranges. Typically,
there exist indoor and outdoor ranges with associated limitations for both. Outdoor ranges
are not protected from environmental conditions, while indoor ranges are limited by space
restrictions.
Indoor ranges make use of anechoic chambers, which are chambers lined with
radar absorbing material to eliminate reflections from the walls.
Various methods exist to
measure the antenna parameters: radiation pattern directivity, gain and polarization. Some of
the methods require the Far-Field criterion and uniform plane illumination and some can be
performed in the Near-Field of the Antenna Under Test (AUT).
4.4.1 Instrumentation system
The instrumentation required to accomplish a measuring task depends largely on the functional
requirements of the design. Antenna-range instrumentation must be designed to operate over a
wide range of frequencies, and it usually can be classified into five categories, which are source
antenna and transmitting system, receiving system, positioning system, recording system and
data-processing system (Figure 4.3). The Transmitting System should be capable of outputting
a stable known power. The output frequency should also be tunable (selectable), and reasonably
stable (stable means that the frequency you get from the transmitter is close to the frequency
you want). The Receiving System simply needs to determine how much power is received from
the test antenna. The receiving system can be more complex, with high quality amplifiers for
low power measurements and more accurate detection devices. The Positioning System controls
the orientation of the test antenna. Since we want to measure the radiation pattern of the test
antenna as a function of angle, we need to rotate the test antenna so that the source antenna
illuminates the test antenna from different angles.
4.4.2 Anechoic chamber
Anechoic chambers are indoor antenna ranges.
The walls, ceilings and floor are lined with
special electromagnetic wave absorbing material. Indoor ranges are desirable because the test
conditions can be much more tightly controlled than that of outdoor ranges.
The material
59

Figure 4.3: Diagram of required antenna measurement equipment.
is often jagged in shape as well, making these chambers quite interesting to see. The jagged
triangle shapes are designed so that what is reflected from them tends to spread in random
directions, and what is added together from all the random reflections tends to add incoherently
and is thus suppressed further.
The antenna gain, AR, and radiation patterns were measured inside the anechoic chamber
of our laboratory, having a dimension of 4 × 8.5 × 2.4 m 3 . Two conical log spirals (LHCP and
RHCP) and a dipole antenna were used as standard reference antennas. Extra caution was taken
to precisely align the antenna under test (AUT) and the reference antenna in order to obtain
accurate measurement results. The schematic of the measurement system and photograph of
AUT are shown in Figure 4.4 and 4.5.
Figure 4.4: Schematic of the antenna measurement system.
60

Figure 4.5: Photograph of antenna measurement in anechoic chamber at MRSL, Chiba University:
(a) Array antenna and (b) Triple-fed antenna.
In antenna measurement, the far field region is the most important, as this determines the
antenna’s radiation pattern. The far-field is the region farthest away from the antenna where
the field distribution is essentially independent of the distance from the antenna (propagating
waves). In this region, the radiation pattern does not change shape with distance. Also, this
region is dominated by radiated fields, with the E- and H-fields orthogonal to each other and the
direction of propagation as with plane waves. If the maximum linear dimension of an antenna
is D and range between the AUT and the conical log-spiral is R, then the following conditions
must be satisfied to be in the far field region
R = 2D2
λ
(4.1)
4.4.3 Input characteristics measurement
Input characteristics of the antenna consist of reflection coefficient, input impedance and standing
wave ratio. These input characteristics are measured with the RF Vector Network Analyzer
(Agilent VNA E8364C). Before performing this measurement, a standard calibration process is
needed to minimize imperfections, which will cause the equipment to yield less than ideal measurements.
There are three calibrated reflection standards: a short circuit, an open circuit, and
a matched load. This one-port calibration makes it possible to derive the actual reflection S-
parameters of the Antenna-under-test (AUT). Fortunately, impedance measurements are pretty
61

easy if you have the right equipment. In this case, the right equipment is a Vector Network
Analyzer (VNA). This is a measuring tool that can be used to measure the input impedance
as a function of frequency. Alternatively, it can plot S 11 (return loss), and the VSWR, both of
which are frequency-dependent functions of the antenna impedance.
4.4.4 Radiation characteristics measurement
The radiation patterns (amplitude and phase), polarization, and gain of an antenna, which are
used to characterize its radiation capabilities, are measured on the surface of a constant radius
sphere. Since the radial distance is maintained fixed, only the two angular coordinates (θ, ϕ)
are needed for positional identification. A representation of the radiation characteristics of the
radiator as a function of θ and ϕ for a constant radial distance and frequency is defined as the
pattern of the antenna.
Gain measurement
The second experiment involves the gain characterization of various antennas using the gain
transfer method. The concept of gain, as the amount of received power in the optimum direction
relative to a nondirective (isotropic) antenna, is described in the experiment procedure. The
received signal power from the antenna under test (AUT) is measured relative to levels detected
by a standard gain dipole antenna. This allows the gain of the AUT to be calculated based on
the known gain of the standard.
Antenna Gain = 10 log(P t /P s ) dBic (4.2)
As a reference antenna is a dipole antenna of Anritsu MP651A, and the measuring equipment
having the same polarization as the AUT. The illustration of the gain measurement can be seen
in Figure 4.6. The antenna gain is derived from
Radiation pattern measurement
The patterns of antennas can be measured in transmit or receive mode. Some types of antennas
must be measured under both transmit and receive conditions. In general, the pattern of an
antenna is three-dimensional. Because it is not practical to measure a three-dimensional pattern,
a number of two-dimensional patterns are measured. A two-dimensional pattern is referred to
62

BIBLIOGRAPHY
Figure 4.6: Antenna gain measurement.
as a pattern cut. The antenna performance is often described in terms of its principal E-
and H-plane patterns. For a linearly polarized antenna, the E- and H-planes are defined as
the planes containing the direction of maximum radiation and the electric and magnetic field
vectors, respectively.
Axial ratio measurement
To perform the axial ratio measurement, the test antenna is used as the receiver whereas the
conical log-spiral antenna performs as the source. In measurement, the an AUT configured to
receive a right hand circularly polarized (RHCP) signal and a left hand circularly polarized
(LHCP) signal for the purpose of evaluating axial ratios. The powers from the both sources are
recorded as P R for LHCP and P L for the LHCP. The AR is derived using following equation
Axial Ratio (AR) = 20 log
∣
10 P R
20 + 10 P L
20
10 P R
20 − 10 P L
20
∣ ∣∣∣∣
dB (4.3)
Bibliography
[1] Ansoft, Ansoft HFSS - User-guide, Ansys Inc., USA, 2009.
[2] Zeland, IE3D User’s Manual Release 11.2 , Zeland Software Inc., USA, 2006.
63

Chapter 5
Results and Discussion
5.1 Triple Proximity-fed circularly polarized microstrip
antenna
In order to simplify the antenna design, an antenna patch is usually designed to be square
or circular, resulting in a square microstrip antenna (SMA) or a circular microstrip antenna
(CMA). In the dual feed design for both SMA and CMA, the CP radiation can be achieved
by providing currents with equal amplitude and mutual phase difference of 90 o (Girish and
Ray, 2003). Alternatively, a single feed approach has also been studied with configuring the
shape of the radiator (Raul et al., 2000 and Kim et al., 2008). A previous work demonstrated
that an equilateral triangle patch CP antenna can be realized with proximity-coupled dual
feeding (Baharuddin et al., 2009). The antenna performance of this CP antenna, however, was
not satisfactory in terms of both 3-dB axial ratio bandwidth and antenna gain.
Hence, we
introduce a new design of multiple fed microstrip antenna, namely a triple proximity-fed CMA.
The antenna is designed to operate in L-band and intended for various applications such as
circularly-polarized synthetic aperture radar (CP-SAR), global positioning system (GPS), etc.
64

5.1.1 Design of proposed antenna
In the present design, the CP radiation is produced with a triple-fed CMA. The phase shift
among the three feeds is adjusted in a range, including the conventional value of 120 o (Chen et
al., 2010), so as to give the best AR smaller than 3 dB. The network feeding is implemented
in proximity-coupled method (Pozar and Kaufman, 1987) and a circular-sector stub is adopted
as a power divider, which enables to share the current distribution equally for odd-number
feedings as well. The power from a 50 Ω feed line is divided symmetrically in the sector by
placing each feed at an angular distance of β = 22.5 o , with a width (w) of the feed of 4.3 mm
and radius (r f ) of the circular-sector of 20.5 mm. The sector angle, α, of the power divider is
selected to be 90 o to obtain the optimal performance (Abouzahra, 1988) as illustrated in Figure
5.1.
Figure 5.1: The 3-way circular-sector-shaped power divider.
The geometry design of the proposed antenna is shown in Figure 5.2.
The antenna is
fabricated on two layers of dielectric substrate, each having thickness t = 1.6 mm, conductor
thickness t c ≈ 35µm, dielectric constant ε r = 2.17 and loss tangent 0.0005 (See Figure 5.3).
The radius of the radiator (r) is 43.56 mm, while the ground plane size (L × W ) is 154 mm ×
150 mm. Other parameters of the CMA are listed in Table 5.1.
Table 5.1: Triple proximity-fed parameters (in units of mm).
Parameters w d l a l b l c r f l f w f
Size 4.3 20.1 28.6 93.0 144 20.5 18.0 5.7
65

Figure 5.2: Geometry design of proposed antenna; (a) top view and (b) side view.
Figure 5.3: Photograph of fabricated CMA: (a) triple proximity-fed and (b) circular radiator.
66

5.1.2 Parameter study
The sense of polarization (LHCP or RHCP) can be determined by turning the sequence of the
phase shift in the feeding network. Here, the CMA is designed to generate LHCP by adjusting
the 120 o phase shifted between l a , l b and l c , clockwise.
The method of moment (the IE3D
simulation software) with a finite ground plane model is employed to optimize the geometrical
design of the antenna.
During the optimization process of the CMA configuration, it was
observed that the choice of the phase shift among l a , l b and l c significantly affect the AR of the
emitted CP radiation. The simulated result of the AR is shown in Figure 5.4 as a function of
frequency for various values of the phase shift (∆ϕ). The best CP radiation is found for phase
shift around 120 o (λ/3).
Figure 5.4: Simulation results showing the frequency dependence of the axial ratio (AR) of the
CMA for various values of phase shift applied to the feeds.
Figure 5.5 shows the vector current distribution of the prototype antenna for various source
phases (φ s ). The directions and sizes of the vectors are affected by the phase of the sinusoidal
wave given to the main feed. In Figure 5.5, it can be seen that the rotation of vectors is in
the clockwise direction, indicating the sense of polarization (LHCP). The simulated efficiency
of the present antenna design is shown in Figure 5.6. This result indicates that more than 87%
of the power from the source can be radiated by the antenna into space. Moreover, the ratio
of the total power dissipated by the antenna to the net power accepted by the antenna at its
67

terminals during the radiation process (radiation efficiency) is around 88%. These high values
are in line with good efficiencies characterizing microstrip antennas (Faiz and Wahid, 1999).
Figure 5.5: Vector current distribution of the designed CMA for various phase values of the
source, (a) φ s = 0 o , (b) φ s = 45 o , (c) φ s = 90 o , and (d) φ s = 135 o .
Figure 5.6: The efficiency of the antenna configuration as a function of frequency.
5.1.3 Input characteristic
In Figure 5.7, the reflection coefficient (S 11 -parameter) is plotted as a function of frequency. At
the center of working frequency (1.28 GHz), the reflection coefficient of the measured result is
around -19.0 dB, which is consistent with the simulated value of -19.1 dB. A good agreement in
-10 dB impedance bandwidth of 34 MHz is obtained for both measured and simulated results,
though the measured result exhibits minimum impedance at around 1.272 GHz, 0.3% shifted
68

from the designed frequency. This frequency shift is presumably due to the effect of errors in
the antenna fabrication process (e.g., milling error) and the influence of the ground plane size.
Figure 5.7: Simulated and measured reflection coefficient vs. frequency.
Figure 5.8: Simulated and measured input impedance (Z in ) plotted as a function of frequency.
In Figure 5.8, the input impedance is plotted against the frequency. The measured value
at the working frequency of 1.28 GHz is 48.09 Ω and 6.44Ω for input resistance and input
reactance, respectively. The input resistance is about 2.6% smaller than the simulated value of
49.36Ω, probably resulting from the resistance of a connector and soldering. Nevertheless, both
the simulated and measured results provide good matching in impedance nearly equal to 50Ω.
69

5.1.4 Radiation characteristic
The relation between the AR and frequency is shown in Figure 5.9. The 3-dB AR bandwidth
achieved at the direction of θ = 0 o (i.e., the AUT is set perpendicular to the standard reference
antenna) is about 8.7 MHz, which corresponds to 0.68 % of the operation frequency of 1.28
GHz.
In the simulation, on the other hand, the value is 9.0 MHz, or around 0.70% of the
operation frequency. The minimum values of AR are obtained to be 0.08 dB and 1.13 dB for
simulation and measurement, respectively. A possible cause of this discrepancy is the deviation
of the phase shift among the three feeds due to slight inaccuracy in the milling process of the
feeding network, for instance. The difference in the ground plane size can also lead to such
a slight degradation of the 3-dB AR bandwidth.
When the ground plane size is increased,
the edge-diffracted fields cause tilting of the beam in the direction of low elevation angles and
reduce the maximum gain, ultimately affecting the characteristics of the AR (Sri Sumantyo et
al., 2005).
Figure 5.9: Simulated and measured axial ratio (AR) vs. frequency at θ = 0 o .
The antenna gain is simulated and measured as a function of frequency (Figure 5.10). From
this figure, it can be seen that the measured value of the maximum gain is about 7.2 dBic at
1.275 GHz, whereas the gain measured at 1.28 GHz is 7.11 dBic. This value is by 0.1 dBic
lower than the value expected from the simulation. Such a difference between the simulated
and measured results may possibly be ascribed to the loss of the substrate that supports the
70

Figure 5.10: Simulated and measured gain (G) vs. frequency at θ = 0 o .
proximity feeding.
Figure 5.11(a) and (b) show the radiation patterns in the x−z and y −z plane, respectively.
Both measured and simulated patterns at 1.28 GHz are plotted. In Figure 5.11(a), the measured
maximum gain at the azimuth angle of Az = 0 o is about 6.6 dBic, by approximately 0.6 dBic
lower than the simulated gain of 7.2 dBic. The 3-dB beamwidth of the fabricated antenna is
about 91 o , larger than the simulated beamwidth of 87 o . In y − z plane, similar patterns appear
as seen in Figure 5.11(b). The peak gain from the measurement is 6.8 dBic with a half power
(3-dB) beamwidth of around 87 o , while the simulated peak is about 7.2 dBic with the halfpower
beamwidth of 90 o . Radiation patterns in both the y − z and x − z planes exhibit typical
nulls on the dipole axis (x-axis), at the θ angle of 90 o . The results shown in Figure 5.11(a)
and (b) indicate that the agreement between the simulation and measurement is reasonable in
terms of the gain performance. The slight difference seen in gain patterns may be ascribable
to the imperfection on the measurement process, namely slight variations in antenna alignment
during rotation.
71

Figure 5.11: Measured and simulated radiation pattern of proposed antenna at f = 1.28 GHz:
(a) in the y − z plane (Az = 90 o and 270 o ) and (b) in the x − z plane (Az = 0 o and 180 o ).
72

5.2 LHCP array microstrip antenna
In previous works, several single element CP antennas have been fabricated in MRSL (Baharuddin
et al., 2010 and Baharuddin et al., 2011). For CP-SAR applications, an array configuration
must be adopted to satisfy the requirement of the system. Here we propose an L-band CP-
SAR antenna consisting of 12 elements of simple square-shaped, corner-truncated microstrip
antenna with a novel feeding network. A similar square-shaped CP array antenna has been
reported (Rahim et al., 2008) with a feed network using a T junction power divider. Obviously
the limitation of this previous antenna design is that the possible power split must be
2n(n = 1, 2, 3...). In the present work, we propose the use of a circular sector power divider,
which enables odd-number feeding as well. The proximity-coupled feeding method is adopted
for feeding each patch element. The main advantage of this feed technique is the ease in both
design and fabrication adjustment processes, resulting in good AR and impedance matching
(Pozar and Kaufman, 1987). While a good AR is generally attained by controlling the feed
point, the impedance matching can be achieved by controlling the length and width of the
microstrip line.
5.2.1 Array antenna design
The excellent performance of the overall CP-SAR system can only be attained through optimizing
the antenna design in terms of various antenna parameters such as the gain, directivity,
AR, reflection coefficient, etc. In this research, the operating frequency of array antenna is fixed
at 1.27 GHz (L-Band).
The square patch geometry is chosen since it can easily be arranged and fabricated to
produce CP radiation. For the sake of feeding, two opposite corners of each patch is truncated
for either LHCP or RHCP operation. The first step in the design is to specify the dimension of
a single element microstrip antenna. In practice, the patch dimension must be slightly less than
a half wavelength (23.6/2 = 11.8 cm), in order to take fringing fields into consideration. The
fringing fields along the width can be modeled as radiating slots, and electrically the width of
the microstrip antenna looks greater than its physical dimensions. Thus, the relation between
73

the physical length, L, and the effective length, L eff , can be written as (Chen and Chia, 2006)
L eff = L + ∆L (5.1)
where ∆L (length extension) stands for the effect of the fringe fields. For a given resonance
frequency f 0 , the effective length can be calculated as
L eff =
c
2f 0
√
εeff
(5.2)
Here, c and ε eff are the speed of light and effective dielectric constant, respectively. Empirically
the value of ∆L is given as (Hammerstad, 1975)
∆L = 0.412h (ε eff + 0.3) ( h
W + 0.262)
(ε eff − 0.258) ( (5.3)
h
W
+ 0.813)
where W is the width of the microstrip patch and h is the substrate thickness: besides, ε eff is
given as
ε eff = ε r + 1
2
+ ε r − 1
2
[
1 + 12 h ] − 1
2
W
(5.4)
where ε r is the dielectric constant of the substrate.
In this work, we have employed a substrate (NPC-H220A, Nippon Pillar Packing) having a
thickness 3.2 mm, a dielectric constant ε r = 2.17 and a loss tangent δ = 0.0005. By entering
the width of patch W = 93.82 mm and resonance frequency f 0 = 1.27 GHz, the effective length
(L eff ) and effective dielectric constant (ε eff ) can be obtained 77.18 mm and 2.08, respectively.
The basic components of the present antenna array are corner-truncated square patches,
a split-T and a 3-way circular-sector-shaped power divider. The array is composed of three
blocks, each of which consists of 2×2 patches (Figure 5.12). The circular-sector divider is used
to divide the feed energy into the three blocks, and a split-T power divider is employed to
divide the power equally into two parts at each division point. The optimal performance of
the circular-sector divider can be achieved when the sector angle, α, is equal to 90 o or 180 o
(Abouzahra et al., 1988). The power from a 50 Ω feed line is divided symmetrically in the
74

sector by placing each feed at an angular distance of θ = 22.5 o , with a width (W ) of the feed
of 6.8 mm and radius (R) of the circular-sector of 24 mm. Proposed antenna geometry with
thickness h 1 = h 2 = 1.6 mm and a photograph of fabricated antenna is shown in Figure 5.13
and Figure 5.14.
Figure 5.12: Configuration of the antenna array consisting of three blocks, each block having
2×2 element patches: (a) top view, and (b) side view.
The geometrical design of the antenna array is optimized using method of moment (the IE3D
simulation software) by assuming a finite ground plane model. By adjusting several parameters
indicated in Figure 5.14, the optimum parameters are obtained as listed in Table 5.2. In Table
Table 5.2: Geometry parameters (mm) of circularly polarized array antenna.
Parameters Size (mm) Parameters Size (mm)
L 79.18 Sv 50.00
C 9.00 R 24.00
D 14.00 W 1 6.80
F, L h 40.00 W 2 5.00
L v 38.00 L ground 810.00
S h 54.00 W ground 325.00
5.2, L ground and W ground are the total length and width of the ground plate. These dimensions
(810 mm×325 mm) are equal to the size of the fabricated antenna as shown in Figure 5.13.
75

Figure 5.13: Photograph of the fabricated antenna: (a) feed network and (b) 2 × 6 radiation
patch.
76

Figure 5.14: Geometry layout of a single piece of antenna array.
5.2.2 Input characteristics
Figure 5.15 shows the frequency dependence of S 11 measured for the fabricated antenna. The
simulated and measured curves show similar behavior, exhibiting reflection minimum around
1.270-1.275 GHz, though somewhat higher resonance frequency is seen for the measurement.
A slight difference of the working frequency between the simulation and measurement is presumably
due to the effect of alignment errors in the fabrication of the antenna in the antenna
range measurements and the effect of ground plane size. The difference in the minimum values
of the reflection coefficient (-60 vs. -30 dB) can also be considered to the degradation during
the fabrication process. Nevertheless, the measured value of the 10-dB impedance bandwidth
is 77 MHz, approximately 6.1% of the resonance frequency of 1.275 GHz. This result is by 14
MHz wider than the simulated value of 63 MHz (5.0 %). This dissimilarity result is probably
attributed by the difference in the patch geometry and feed line size (between simulated and
fabricated model) due to the substrate variation.
In Figure 5.16, the real input impedance is plotted against the frequency. The measured value
at the working frequency of 1.27 GHz is 48.07Ω, about 3.9% smaller than the simulated value
77

Figure 5.15: Simulated and measured reflection coefficient plotted as a function of frequency.
Figure 5.16: Simulated and measured real input impedance (Z in ) plotted as a function of
frequency.
78

of 50Ω. In Figures 5.15 and 5.16, slight differences between the simulated and measured results
are presumably due to the resistivity offering from milling process, a connector and soldering.
In the feed network, the length of each element from the patch to connector must be fixed at
nλ/4(n=1, 3, 5 ...) to achieve the optimal current intensity. The deviation of the length from the
designed value can also contribute to the difference between the simulation and measurement.
5.2.3 Radiation characteristics
The relation between the AR and frequency is plotted in Figure 5.17, showing the polarization
characteristics of the fabricated antenna. The 3-dB AR bandwidth achieved at the direction of
θ = 0 o (i.e., the AUT is set perpendicular to the standard antenna) is about 13 MHz, which
corresponds to 1.02 % of the operation frequency of 1.27 GHz. In the simulation, on the other
hand, the value is 11 MHz, or around 0.87% of the operation frequency. The minimum axial ratio
of the measured data is around 1.13 dB, exhibiting a difference of around 0.6 dB as compared
with the simulation result. Inaccuracies in the implementation of the corner-truncated patches
are one possible cause of differences between the simulated and measured results. This result
is mostly acceptable for the satellite-borne CP-SAR antenna, but still insufficient for the UAVborne
antenna with a target specification of 233 MHz. In our future work, more considerations
will be needed to further extend the 3-dB AR bandwidth.
Figure 5.17: Simulated and measured axial ratio (AR) plotted as a function of frequency.
79

In Figure 5.18, the antenna gain within the CP bandwidth is plotted as a function of
frequency. The antenna has a high gain, with the measured peak gain of about 16.07 dBic
at 1.27 GHz.
Although this value is slightly lower by around 0.43 dBic as compared with
the simulation, the observed value is mostly above the targeted value of 14.32 dBic. Such a
difference between the simulated and measured results can probably be ascribed to the slight
imperfection in size of patch and feed network offering from milling process, connector, etc.
Figure 5.18: Relationship between antenna gain and frequency at θ angle = 0 o .
Figure 5.19 shows the radiation characteristics produced from the array antenna in the theta
plane (Az = 180 o and 0 o or Azimuth direction of CP-SAR) at 1.27 GHz. Figure 5.19(a) shows
the distribution of the antenna gain, while Figure 5.19(b) shows the distribution of the AR.
In Figure 5.19(a), good agreement is seen between the simulated and measured results, both
showing that the width of the major lobe exceeding the target gain of 14.3 dBic is around 10 o .
The first side lobes appear at θ = −24 o with a peak amplitude of 5.85 dB and at θ = 23 o with
a peak amplitude of 4.71 dB. Thus, the differences in amplitude between the main and side
lobes are around 10.22 and 11.36 dBic for θ = −24 o and 23 o , respectively. When the antenna is
applied to a platform with altitude higher than 2 km, the side lobe level should be lower than
10 dB of the main beam. Thus, the present result satisfies this requirement. In Figure 5.19(b),
the measured width of the 3-dB AR is 24 o , while it is 60 o from the simulation. The measured
beamwidth is narrower than the simulated result, but still this satisfies the targeted beamwidth
80

of 6.77 o .
Figure 5.19: Array antenna characteristics in the theta plane (negative theta for Az = 180 o
and positive for Az = 0 o ) (x − z plane) at f = 1.27 GHz: (a) gain versus theta angle, and (b)
axial ratio versus theta angle.
The antenna characteristics in the theta plane (range direction of CP-SAR) with Az = 270 o
and 90 o are shown in Figure 5.20. The angular distributions of the gain and AR are plotted
in Figure 5.20 (a) and (b), respectively.
In Figure 5.20 (a), similar curves are seen for the
simulated and measured results. The beamwidth at 14.3 dBic is around 36 o , from -19 o to 17 o .
In Figure 5.20 (b), on the other hand, the measured 3-dB AR beamwidth is around 34 o , while
the simulation result is 75 o . These results indicate that the targeted beamwidth is achieved,
satisfying our system requirement.
81

Figure 5.20: Array antenna characteristics in the theta plane (negative theta for Az = 270 o
and positive for Az = 90 o ) (y − z plane) at f = 1.27 GHz: (a) gain versus theta angle, and (b)
axial ratio versus theta angle.
82

Figure 5.21: Cross polarization (E-left and E-right) of array antenna at f = 1.27 GHz.
The cross polarization performance of the array antenna is shown in Figure 5.21. Based on
this graph, the measured peak gain of E-left is about 16.07 dBic at θ = 0 o , about 23.53 dBic
higher than measured E-right gain value of -7.46 dBic. From this comparison result can be
inferred that the sense of polarization of the array antenna is left-handed circularly polarized
(LHCP).
5.3 RHCP array microstrip antenna
To realize the circular polarization, the CP-SAR system is composed by Left Handed Circularly
Polarized (LHCP) and Right Handed Circularly Polarized (RHCP) sub array antenna, where
the transmission (Tx) is working in RHCP or LHCP, and reception (Rx) is working in both
RHCP and LCHP. In previous research (Yohandri et al., 2011), the LHCP array antenna has
been reported, and in this work, the RHCP array antenna will be discussed.
5.3.1 Array antenna configuration
In general, geometry design of the RHCP array antenna is similar to the LHCP design. The
proposed antenna is developed using simple corner-truncated square-patch elements, which opposite
corner with the LHCP. Identic feeding network is implemented in this proposed antenna.
83

Figure 3 shows the design and photograph of the antenna, made using a substrate with a
permittivity ɛ r = 2.17 and a loss tangent δ = 0.0005.
Figure 5.22: RHCP array antennas: (a) configuration design and (b) photograph.
5.3.2 Input characteristics
In figures 5.23, the characteristic of input impedance of the proposed antenna is presented. In
general, good agreements are shown between the simulated and experimented results, indicating
that the antenna properties mostly satisfy the target specification as a CP-SAR array antenna.
Based on Figure 5.23, the measured value of the 10-dB impedance bandwidth is 79.1 MHz,
approximately 6.1% of the resonance frequency of 1.275 GHz. This result is by 15 MHz wider
than the simulated value of 64 MHz (5.0 %). This dissimilarity result is probably attributed
by the difference in the patch geometry and feed line size (between simulated and fabricated
model).
84

Figure 5.23: Simulated and measured reflection coefficient plotted as a function of frequency.
5.3.3 Radiation characteristics
As can be seen in Figure 5.24 the antenna has an axial ratio less than 3dB from 1.276 GHz
to 1.290 GHz. This accounts for 1.1% 3dB axial ratio bandwidth at a center frequency of 1.28
GHz. The value of the axial ratio at the center frequency is 0.59 dB. The working frequency
of this antenna is sifted to the higher frequency about 1 MHz. It assumed contribution of the
inconsistency in the fabrication process due to multi section of the antenna. Figure 5.25 shows
that the gain of the antenna is varied around 13 dBic. In other hands, the simulated result
fairly constant around 16 dBic.
Figure 5.26 shows the radiation characteristics produced from the array antenna in the theta
plane (Az = 180 o and 0 o or azimuth direction of CP-SAR) at 1.28 GHz. Figure 5.26 (a) shows
the distribution of the antenna gain, while Figure 5.26 (b) shows the distribution of the AR.
In Figure 5.26a, good agreement is seen between the simulated and measured results, both
showing that the width of the major lobe exceeding the target gain of 14.3 dBic is around 10 o .
The first side lobes appear at θ = 25 o with a peak amplitude of 6 dB and at θ = 23 o with a
peak amplitude of 5.7 dB. In Figure 5.26(b), the measured width of the 3-dB AR is 20 o , this
result almost similar with the simulated one 23 o .
85

Figure 5.24: Simulated and measured axial ratio (AR) plotted as a function of frequency.
Figure 5.25: Relationship between antenna gain and frequency at θ angle = 0 o .
86

Figure 5.26: Array antenna characteristics in the theta plane (negative theta for Az = 180 o
and positive for Az = 0 o ) (x − z plane) at f = 1.27 GHz: (a) gain versus theta angle, and (b)
axial ratio versus theta angle.
87

5.4 Broadband circularly polarized microstrip antenna
In previous research, a number of CP microstrip antennas for CP-SAR have been developed
(Yohandri et al., 2012 and Baharuddin et al., 2011). Nonetheless, the axial ratio bandwidth
of these antennas is quite narrow (≤ 1%). In CP-SAR system onboard UAV, the broadband
axial ratio antenna is required to maintain the fine resolutions of the sensor. Therefore, the
purpose of the present work is to describe the design of a broadband CP microstrip antenna
for CP-SAR installed on UAV.
5.4.1 Geometry design
The geometry of the proposed antenna is shown in Figure 5.27, where Figure 5.27 (a) gives
the top and side view structure, and 5.27 (b) show the 3D view of the antenna structure.
The circular microstrip antenna is designed on two layers substrate (NPC-H220A, Nippon
Pillar) having a permittivity ɛ r = 2.17 and a loss tangent δ = 0.0005. In addition, to obtain
a broadband antenna, a Wilkinson power divider is implemented on the feed structure. The
proposed antenna is optimized using Ansoft High Frequency Structure Simulator (HFSS). Based
on the simulated result, the optimum geometry parameters of the antenna are the following: L
= 148 mm, W = 124 mm, h = 1.60 mm, W f = 4.70 mm, L f = 40.0 mm, R = 46.0 mm, R f =
25.7 mm, and d = 17.7 mm.
5.4.2 Antenna performances
Figures 5.28 to 5.33 show the reflection coefficient (S 11 ), axial ratio (AR), gain (G), and radiation
pattern of the antenna. The broadband CP antenna characteristic can be achieved as
shown in Figure 5.28 and Figure 5.29. The S 11 bandwidth of the proposed antenna can be
achieved about 580 MHz or 45.7 %. In addition, the axial ratio 3dB bandwidth is obtained
around 548 MHz or 66.9 %.
These broadband characteristics are satisfied for the CP-SAR
onboard UAV requirements.
The gain of proposed antenna is shown in Figure 5.30. The antenna has a high gain, with
the measured peak gain of about 6 dBic at 1.27 GHz. To occupy the requirement gain of the
CP-SAR system, this single element antenna should be arranged in array configuration.
88

Figure 5.27: Geometry of the proposed antenna: (a) top and side view and (b) 3D view.
Figure 5.28: Reflection coefficient plotted as a function of frequency.
89

Figure 5.29: Axial ratio (AR) plotted as a function of frequency.
Figure 5.30: Relationship between antenna gain and frequency at θ angle = 0 o .
90

Figure 5.31: Radiation pattern of the antenna at f = 1.27 GHz.
Figure 5.32: Axial ratio plotted as a function of theta angle.
91

Figure 5.33: A 3D beam pattern of the antenna at f = 1.27 GHz.
5.5 Low sidelobe level array Antenna
In many applications, be useful to have a pattern with an optimum compromise between
beamwidth and side-lobe level. In other words, for a specified beamwidth the side-lobe level
would be as low as possible; or vice versa, for a specified side-lobe level the beamwidth would
be as narrow as possible. For SAR applications, a low side lobe antenna as well as narrow main
beam is desirable to maintain the data quality.
In previous work, an array antenna for CP-SAR application has been developed (Yohandri
et al., 2011).
However, the side lobe level is not too satisfy for CP-SAR system onboard
UAV. There are several techniques that can be used to improve the side lobe level of the array
antenna. In this section, a Chebyshev synthesis method is adopted for linear arrays with halfwavelength
spacing. This method will be implemented to manage the power distribution in the
feed network.
5.5.1 Dolph-Chebyshev synthesis
In this design, the array antenna consists of 5 elements patch with uniform space (d = λ 0 /2)
and side lobe level 20 dB. The power distribution in the feed network is arranged based on
excitation coefficient of the array antenna. For five elements array (P=5, N= 2), the array
92

factor (AF) can be derived as (Stutzman and Thiele, 1998).
AF (u) = a 0 + 2a 1 cos u + 2a 2 cos 2u where u = 2π(d/λ) cos θ (5.5)
For d = λ 0 /2, u = π cos θ. Using the Chebyshev polynomial (3.44) and drive the R from the
following equations
SLL = −20 log R (dB), (5.6)
and solving for z 0
( )
1
z 0 = cosh
P − 1 cosh−1 R , (5.7)
the excitation coefficient can be obtained and the array factor of the five elements antenna can
be written as
AF (u) = 2.6978 + 2(2.2465) cos u + 2(1.3975) cos 2u (5.8)
Normalized to the center of element (a 0 ) the array factor of the antenna is
AF (u) = 1 + 2(0.8327) cos u + 2(0.518) cos 2u (5.9)
Based on the AF equation, the array factor as function of the θ can be plotted as shown in
Figure 5.34.
As shown in Figure 5.34, the side lobe level of the array antenna is around 20 dB. The first
side lobe is appear at 40 o for the both sides.
5.5.2 Geometry design of the antenna
The square microstrip antenna with corner truncated and proximity feed are employed in the
proposed antenna. To occupy the power distribution based on the array factor calculation, a
corporate feed network is implemented in this design. The geometry design of the antenna is
93

Figure 5.34: Dolph-Chebyshev array factor for five elements and λ 0 /2 spaced.
shown in Figure 5.35.
Figure 5.35: Geometry design of the Dolph-Chebyshev array antenna.
The geometrical design of the antenna array is optimized using method of moment (the IE3D
simulation software) by assuming a finite ground plane model. By adjusting several parameters
indicated in Figure 5.35, the optimum parameters are obtained as listed in Table 5.3.
In this work, we have employed a substrate (NPC-H220A) having a thickness 1.6 mm, a
dielectric constant ε r = 2.17 and a loss tangent δ = 0.0005. The photograph of fabricated feed
network and the radiation patch are shown in Figure 5.36.
94

Table 5.3: Optimum parameters of the proposed antenna.
Antenna parameters Value Z ij (Ω)
Z 11 109.6
Z 12 63
Z 21 94
Z 22 69.6
Z 31 64
Z 32 77
Z 1 42
Z 2 40
Z 3 40
Z 42
L
79.55 mm
L g
593.75 mm
170.48 mm
W g
Figure 5.36: Photograph of fabricated feed network and radiation patch.
5.5.3 Simulated and measured results
The comparisons of simulated and measured results are presented in Figure 5.37 to 5.42. In
input characteristics, the reflection coefficient and Voltage Standing Wave Ratio (VSWR) is
plotted as function of the frequency. Meanwhile, the axial ratio, gain and radiation pattern of
the antenna is demonstrated on radiation characteristics.
Input characteristics
Simulated and measured return losses of the proposed antenna are shown in Figure 5.37. The
return losses at 1.27 GHz are lower than 29.21 dB and 16.29 dB in the simulation and the
95

measurement, respectively. Such degradation is probably caused by the transitions mismatch.
The measured S 11 characteristics show slightly broadened characteristics from 1.242 GHz to
1.312 GHz of 5.51%. Meanwhile, the simulataed S 11 show slightly nerrower than measured
result from 1.244 GHz to 1.300 GHz of 4.4%. Moreover, VSWR characteristics is a measure of
how well matched antenna is to the cable impedance. Simulated and measured results of the
VSWR versus frequency are compared in figure 5.38. The measured impedance bandwidth for
VSWR ≤ 2 is 6.45% ranging from 1.217 GHz to 1.299 GHz, while the simulation result show
a 4.57% bandwidth achieved from 1.243 GHz to 1.301 GHz. Good agreements are observed
between the simulated and the measured results throughout the whole operating bandwidth.
Figure 5.37: Simulated and measured reflection coefficient plotted as a function of frequency.
Radiation characteristics
The AR characteristics are slightly shifted to the higher frequency end, but the bandwidth remains
almost the same in both simulation and measurement. A simulated axial ratio bandwidth
of 0.9% is obtained from 1.264 GHz to 1.275 GHz where the measured results show a bandwidth
of 0.83% from 1.268 GHz to 1.278 GHz. The measured and simulated gain characteristics of
the antenna shown in Figure 5.40 are in good agreement. A maximum gains of 11.9 dBic in
simulation and 10.4 dBic in measurement are obtained.
Normalized array antenna characteristics for x − z and y − z plane at frequency 1.272 GHz
96

Figure 5.38: Simulated and measured VSWR plotted as a function of frequency.
Figure 5.39: Simulated and measured axial ratio plotted as a function of frequency.
97

Figure 5.40: Relationship between antenna gain and frequency at θ angle = 0 o .
are plotted in Figure 5.41 and 5.42, respectively. In Figure 5.41a, the good agreement between
calculation, simulation and measurement result are presented. The 3 dB beamwidth is 24 degree
for calculation, 23 degree for measurement and 22.25 degrees for simulation. The first side lobe
appear at -36 o and 32.4 o with side lobe level 20.8 dB and 19.5 dB for the simulation, for the
measured are -58.7 o and 36 o with side lobe level 20.5 dB and 15.5 dB, respectively. The 3 dB
axial ratio beamwidth is 44.5 o for simulated and 51 o for measured as shown in Figure 5.41b.
In y − z plane, the good agreement is achieved between measured and simulated with 3 dB
beamwidth is 78 o and 77 o , respectively. The 3 dB axial ratio beamwidt at this plane is 115 o
for simulated and 56 o for measured result.
98

Figure 5.41: Array antenna characteristics in the theta plane (negative theta forAz = 180 o and
positive for Az = 0 o ) (x − z plane) at f = 1.272 GHz, (a) normalized gain versus theta angle,
and (b) axial ratio versus theta angle.
99

Figure 5.42: Array antenna characteristics in the theta plane (negative theta forAz = 270 o and
positive for Az = 90 o ) (y − z plane) at f = 1.272 GHz, (a) normalized gain versus theta angle,
and (b) axial ratio versus theta angle.
100

BIBLIOGRAPHY
Bibliography
[10] Abouzahra M.D., “Multiport power divider-combiner circuits using circular-sector-shaped
planar components,” IEEE Trans. On Antenna Propagation, Vol. 36, No. 12, 1747–1752,
1988.
[4] Baharuddin M., Wissan V., Sri Sumantyo J.T. and Kuze H.,“Equilateral triangular microstrip
antenna for circularly-polarized synthetic aperture radar,” Progress in Electromagnetics
Research C , Vol. 8, 107–120, 2009.
[4] Baharuddin M., Wissan V., Sri Sumantyo J.T. and Kuze H.,“Development of an elliptical
annular ring microstrip antenna with sine wave periphery,” Progress in Electromagnetics
Research C , Vol. 12, 27–36, 2010.
[4] Baharuddin M., Wissan V., Sri Sumantyo J.T. and Kuze H.,“Elliptical microstrip antenna
for circularly polarized synthetic aperture radar,” International Journal of Electronics and
Communications (IJEC), Vol. 65, No. 1, 62–67, 2011.
[5] Chen L., Fu S.Z., Yong C.J., Fan Z. and Xin X., Siquiera P. and Curlander J., “A threefed
microstrip antenna for wideband circular polarization,” IEEE Antennas and Wireless
Propagation Letter, Vol. 9, 359–362, 2010
[15] Chen Z.N. and Chia M.Y.W., Broadband planar antennas: design and applications, John
Wiley & Sons Inc., England, 2006.
[12] Faiz M.M. and Wahid P.F., “A high efficiency L-band microstrip antenna,” Antennas and
Propagation Society International Symposium, Vol. 1, Page 272–275, Florida, August 1999.
[8] Girish K. and Ray K.P., Broadband Microstrip Antennas, Arteck House, Norwood, MA,
2003.
[12] Hammerstad E.O., “Equations for microstrip circuit design,” 5th European Microwave
Conference, Page 268–272, Hamburg, Germany, September 1975.
[10] Kim B., Pan B., Nikolaou S., Kim Y.S., Papapolymerou J. and Tentzeris M.M., “A novel
101

BIBLIOGRAPHY
single-feed circular microstrip antenna with reconfigurable polarization capability,” IEEE
Trans. On Antenna Propagation, Vol. 56, No. 3, 630–638, 2008.
[11] Pozar D.M. and Kaufman B., “Increasing the bandwidth of a microstrip antenna by proximity
coupling,” Electronics Letters, Vol. 23, No. 8, 368–369, 1987.
[12] Rahim M.K.A., Masri T., Ayop O. and Majid H.A., “Circular polarization array antenna,”
Asia-Pacific Microwave Conference, Page 1–4, Hong Kong, December 2008.
[14] Raul R.R., Franco D.F. and Nicolaos G.A., “Single-feed circularly polarized microstrip ring
antenna and arrays,” IEEE Trans. On Antenna Propagation, Vol. 48, No. 7, 1040–1047,
2000.
[14] Sri Sumantyo J.T., Ito K., Delaune D., Tanaka T., Onishi T. and Yoshimura H., “Numerical
analysis of ground plane size effects on patch array antenna characteristics for mobile
satellite communications,” International Journal of Numerical Modelling: Electronic Networks,
Devices and Fields, Vol. 18, No. 2, 95–106, 2005.
[15] Stutzman W.L. and Thiele G.A., Antenna theory and design, John Wiley & Sons Inc.,
New York, 1998.
102

Chapter 6
Conclusions
6.1 Summary
In this dissertation, we have discussed thoroughly the circularly polarized microsrip antenna
for CP-SAR systems.
These antennas will be mounted on an unmanned aerial vehicle for
CP-SAR experiments. In Chapter 1, the brief description about background and motivation
of the research has been presented. The main target of this project is to realize the CP-SAR
system onboard UAV. In this research, we have introduced the new model of circularly polarized
microstrip antenna and array antenna configuration for the CP-SAR system.
In Chapter 2, the review on the concept of circularly polarized synthetic aperture radar
has been discussed. From basic principles of radar, polarization of electromagnetic wave and
main differences of CP-SAR as compared with LP-SAR is explained. Furthermore, the design
of CP-SAR system onboard UAV consists of system and parameters design can be found on
the last section in this chapter. Based on this review, we expect the CP-SAR system can be
realized and present the new sensor for remote sensing.
One important part in realizing this new sensor is circularly polarized antenna.
In this
chapter, the technical features and performance of circularly polarized microstrip antennas
have been described as well. The related feature discussed is the circular polarization natures
of the microstrip antenna, technique for generating the circular polarization wave, the feeding
method of proximity-coupled and concept of microstrip array antenna. The technical reviews
about microstrip antenna in this chapter are contributed in developing the circularly polarized
103

antenna for CP-SAR sensor.
In the chapter 4, development process of the CP microstrip antenna has been described.
First stage is the design and numerical calculation by using Method of Moment (MoM) employing
the Zeland IE3D software. For an antenna with complex geometry design, the Finite
Element Method (HFSS) is employed. The fabrication process can be done by two ways, which
are using high-precision machine and dry film photoresist with UV exposure. The next stages of
these ways are masking, etching, and installing the SMA connector. Measurement is conducted
to verify the simulated result with the experiment. To achieve the good measured result, careful
setting of all equipments is very important. In this works, three CP microstrip antennas
for CP-SAR have been fabricated and characterized in an anechoic chamber. The first is the
circular microstrip antenna which is a triple-fed proximity coupled. The other two antennas
are array configuration of LHCP and RHCP circularly polarized microstrip antennas.
The measured results of the triple-fed circular microstrip antenna, array configuration of
LHCP and RHCP microstrip antenna have been presented and discussed in Chapter 5. Based
on measured results of the new triple-fed circular microstrip antenna, good CP performance
has been attained over a 3-dB axial ratio bandwidth of around 8.7 MHz, with fairly high gain
of about 7.11 dBic. In general, numerical analyses using the MoM software can lead to a good
agreement with experimental results.
On the other hand, the characteristics of an L-band
circularly-polarized antenna array both LHCP and RHCP have been investigated. Excellent
agreements have been found between the simulated and measured results, indicating that the
antenna properties mostly satisfy the target specification as a CP-SAR antenna array. To satisfy
the requirement of CP-SAR system onboard UAV, also the design of a broadband CP antenna
was introduced in this chapter. The performance of this broadband antenna in the numerical
analysis is promising for fulfilling the specification of the CP-SAR system. To guarantee the
quality of the radar data, a low side lobe level array antenna is developed and the antenna
characteristics are promising for the CP-SAR sensor system.
From the whole research work, there are some important findings that can be remarked.
Firstly, the new feed method, namely triple-fed proximity has been developed for generating
circular polarization on circular microstrip antenna. Secondly, new model of a power divider
that can be implemented in feeding network of array antennas for both odd and even number
104

of feed line. The flexibility of the power divider makes it easy to develop an array antenna
with limited space onboard UAV. In addition, by implemented short pins in new model feed
technique on circular microstrip antenna provides the broadband circularly polarized microstrip
antenna.
6.2 Future works
In order to implement the circularly polarized array antenna onboard the UAV, the broadband
antenna in terms of S 11 and axial ratio bandwidth should be realized. In future works, the
development of the broadband CP antenna still become our concern. New method and any
techniques to fulfill the specification requirement of CP-SAR onboard UAV will be explored.
In addition, the compact design, tiny size and light weigh microstrip antenna still considered.
The candidate of broadband single element antenna will be arranged in array configuration.
Dimension of the array antenna should be designed to occupy the available space for the
antenna onboard UAV. Moreover, the coupling effect of the four panel array antennas should
be investigated. Main consideration in designing a dual polarized CP-antenna array is: low
cross-polarization and low mutual coupling.
105

List of Publications
Peer-reviewed papers:
Yohandri, V. Wissan, I. Firmansyah, P. Rizki Akbar, J.T. Sri Sumantyo, and H. Kuze, ”Development
of Circularly Polarized Array Antenna for Synthetic Aperture Radar Sensor Installed
on UAV,” Progress in Electromagnetics Research C, Vol. 19, pp. 119-133, January 2011.
Yohandri, J.T. Sri Sumantyo, and Hiroaki Kuze, A new triple proximity-fed circularly polarized
microstrip antenna, AEU-International Journal of Electronics and Communications, Vol. 66,
Issue 5, Pages 395-400, May 2012.
Conferences:
Yohandri, Iman Firmansyah, Prilando Rizki Akbar, Josaphat Tetuko Sri Sumantyo and Hiroaki
Kuze, ”Design of Circularly Polarized Synthetic Aperture Radar (CP-SAR) onboard Unmanned
Aerial Vehicle,”The Institute of Electronics, Information and Communication Engineers-Space,
Aeronautical and Navigational Electronics Conference (IEICE), Technical Report SANE2010-
62, pp.11-16, Vol. 110, No. 173, 25 August 2010, Niigata University, Niigata, Japan.
Yohandri, Iman Firmansyah, Josaphat Tetuko Sri Sumantyo and Hiroaki Kuze ”Development
of CP-SAR Sensor Onboard Unmanned Aerial Vehicle,” The 50th Spring Conference of the
Remote Sensing Society of Japan, A18. Pp.45-46, 26-27 May 2011 Century Anniversary Hall,
106

Nihon University, Tokyo, Japan.
Yohandri, J.T. Sri Sumantyo, and H. Kuze, ” Circularly Polarized Array Antennas for Synthetic
Aperture Radar Sensor,” Progress in Electromagnetics Research Symposium (PIERS),
12-16 September 2011, Suzhou, China.
Yohandri, J.T. Sri Sumantyo, and Hiroaki Kuze, ”Design of a Broadband Antenna for CP-
SAR installed on Unmanned Aerial Vehicle,” The 17th CEReS International Symposium, P10,
Chiba Unversity, Japan, March 1, 2012
107

Curriculum Vitae
YOHANDRI
He was born in Bayur, Maninjau, West Sumatera, Indonesia, on July 25, 1978. He received his
B.Sc. degree in physics specializing in physics instrumentation from State University of Padang
(UNP), Indonesia, in 2001 and his M.Sc. degree also in Physics from Bandung Institute of
Technology (ITB), Indonesia, in 2005.
From 2005 up to now, he was with Physics Department,
Faculty of Mathematics and Natural Science, State University of Padang, Indonesia as
a lecturer. Since 2009, he become a Doctoral Student and expected to get the Ph.D degree in
Graduate School of Advanced Integration Science, Chiba University, Japan. Mr. Yohandri is
a member of the Institute of Electrical and Electronics Engineers (IEEE) dan Remote Sensing
Society of Japan (RSSJ).
108

Appendix A IE3D
Electromagnetic Simulation
Some users may have a geometry constructed using other tools. The MGRID can import and
export in GDSII and CIF formats in the standard version. The optional ADIX converter allows
a user to import and export geometry in AutoCAD DXF format (for 2D or 3D), ACIS format
(for 3D) and GERBER format. ADIX is fully integrated into MGRID. When the ADIX optional
is enabled, MGRID is able to import and export in GDSII, CIF, DXF, ACIS and GERBER
formats.
Figure A.1: The flow chart of a basic IE3D EM simulation.
109

To perform an electromagnetic simulation, a user starts from the layout editor MGRID. On
MGRID, we draw a structure as a group of polygons. After finish constructing the structure
as polygons and defining ports on it, we can invoke the simulation engine IE3D to perform
an electromagnetic simulation. The simulation result is saved into a file in the Agilent/EEsof
compatible format. The simulation result can also be displayed and processed using the post
processor MODUA of the IE3D package. We can also define lumped elements such as resistor,
capacitor, inductor, mutual inductor, open circuit, short circuit and ideal connection on
MODUA to do a mixed Electromagnetic and circuit simulation.
MODUA is automatically
invoked by IE3D to display the solved s-parameters after a simulation.
One of the major advantages of electromagnetic simulation is that the field and current
distributions from a simulated structure are accessible to the users. Information on the current
and field distribution in a structure can be valuable to circuit and antenna designers. On the
IE3D, we can optionally save the current distribution file in a simulation. The current distribution
file can be opened on MGRID in post-processing mode to display the current distribution
as colorful density plots or as vector current distribution plots. You can find the radiation patterns
and other parameters from the MGRID as post-processor. Finally, the radiation patterns
saved into files from MGRID can be displayed and post-processed on the PATTERNVIEW. We
can display the 3D patterns, 2D patterns, merge different patterns, find array radiation patterns,
and find the transfer functions between the transmitting (Tx) antenna and the receiving
(Rx) antenna. We can display and process the parameters of linearly polarized and circularly
polarized antennas.
On the MGRID as post-processor, we can also calculate the near field
distribution on the structure. Near field distribution visualization is missing from most MOM
simulators. On IE3D, there is a robust and flexible near field visualization.
The IE3D package consists of the six major application programs:
MGRID
It is the major layout editor for construction of a structure. It allows a user to create and edit a
structure as polygons and vertices. It has full control over the detail shapes and locations of geometry.
Starting from V14, MGRID is renamed as IE3D EM Design System. It has integrated
110

layout editing, s-parameters visualization and post processing, current distribution visualization,
near-field and far-field post processing and visualization. It also has FastEM Design Kit
for real-time full-wave EM tuning and optimization.
IE3D LIBRARY
The object-oriented schematic-layout editor for parameterized geometry modeling and editing.
With the introduction of FastEM Design Kit for real-time EM tuning, optimization and
synthesis, parameterization becomes necessarily needed and extremely important for IE3D fullwave
design. Parameterization is available on the major IE3D layout editor MGRID. However,
it is limited to vertices and polygons levels. High-level parameterization can be done on
IE3DLIBRARY. To make IE3DLIBRARY more flexible, we have introduced Boolean objects
and void objects. The new introduction makes IE3DLIBRARY much more capable in generating
sophisticated parameterized models. IE3DLIBRARY is relatively easy to use because no
many commands are involved. Detailed discussion on using IE3DLIBRARY can be found from
other electronic documentations.
AGIF
The IE3D-SI advanced geometry modeling tool to create full-3D IE3D models directly from
GDSII files, Cadence Virtuoso and Cadence Allegro.
IE3DOS
It is the EM simulator or simulation engine for numerical analysis. It is a DOS-style command
line application. It is called in the background by the IE3D dialog to perform an EM simulation.
It is normally hidden from the customers. IE3DOS supports Win32, Win64, Linux32 and
Linux64. The 64-bit editions allow users to solve large structures.
MODUA
MODUA is the schematic editor for parameter display and nodal circuit simulation. Most of
its capabilities are integrated into MGRID in V14. Mixed EM and circuit cosimulation is still
the unique feature on MODUA while other s-parameter display and post processing features
111

are integrated into MGRID.
PATTERN VIEW
Post processor for radiation pattern visualization and post processing. All functionalities of
PATTERNVIEW are integrated into MGRID in V14.
ADIX
It is the optional ACIS/DXF/GDSII/GERBER format converter. All functionalities of ADIX
are integrated into MGRID for those users choose the ADIX option.
112

Appendix B HFSS
Electromagnetic Simulation
Model setup
First the model of the microstrip patch antenna has to be drawn in HFSS. It consists of
rectangular substrate and the metal trace layer as shown in Figure B.1. The dimensions of
antenna can be found in the HFSS simulation file.
Figure B.1: Patch antenna layout.
Lamped port setup
In order to excite the structure an excitation source has to be chosen. For this simulation a
lamped port will be used. In order to get an accurate result, the lamped port has to be defined
113

properly. After a rectangle is drawn, the Lamped Port excitation was assigned to it. The
lamped port setup is shown in Figure B.2.
Figure B.2: Lamped port.
Airbox and boundary conditions
An airbox has to be defined in to model open space so that the radiation from the structure
is absorbed and not reflected back.
The airbox should be a quarter-wavelength long of the
frequency of interest in the direction of the radiated field. In the directions where the radiation
is minimal, this quarter-wavelength condition does not have to be met and an air space may not
even have to be defined. The antenna with airbox and lamped port setup is shown in Figure
B.3.
Analysis/Sweep Setup
A Solution Setup is added to the analysis of the simulation as the following:
Solution Frequency: 1.27 GHz
Maximum number of Passes: 20
Maximum Delta S: 0.02
In addition, the field data is saved for each frequency point in the sweep; field data needs
to be saved in order to do any field post-processing.
114

Figure B.3: Patch antenna layout showing airbox and waveport.
Figure B.4: Solution setup for the simulation.
115

Figure B.5: Frequency sweep in simulation.
Figure B.6: Validation check of the proposed antenna.
116

Before running the simulation, the proposed antenna is checked to make ensure no errors in
the design. The validation check window is shown in Figure B.6.
Plotting results
The resulting return loss of the structure is shown in Figure B.7.
Figure B.7: Return loss and axial ratio of antenna from 0.9 GHz to 1.6 GHz.
To plot the far-field patterns of the antenna, a far-field setup has to be created. To create farfield
setup go to HFSS > Radiation > InsertFar − FieldSetup > InfiniteSphere. For the threedimensional
pattern, the default values can be used. Figure B.8 shows the three-dimensional
117

Figure B.8: Three-dimensional far-field patterns.
patterns.
118

Appendix C Antenna fabrication
Microwave artwork
In the microwave artwork process, the layout of the antenna model to be fabricated is made
in AutoCAD (dxf file). The artwork is done on the substrate material type NPC-H220A Pillar
PC-Clad microwave copper-clad laminates produced by Nippon Pillar Packing co.,Ltd, with
characteristics as follows : thickness 1.6 mm, relative permittivity ε r = 2.17 and dissipation
factor 0.0005. Generally the substrate must have a good shape stability, withstand high temperatures
during soldering and have a smooth and flat surface (no bend after etching) to reduce
losses. In this research the artwork is done by two ways, which are PCB Prototyping Machine
and using dry film photoresist.
PCB Prototyping Machine
The artwork done by a PCB Prototyping Machine consists of some steps, milling, drilling and
routing. Milling is performing drawing the geometry shape of the radiator antenna and feed
line on the surface of the material. Then next is drilling, to make holes for screw installation to
join the two layers of material. Last step is routing, to cut the edge of the material according
to the finite ground size of the antenna. The Seven Mini PCB Prototyping Machine and the
tools are shown in the Figure C1. The possible fabrication error contributed from this stage are
milling machine removes more or less copper than expected and imperfections in the cutting
the edge of the substrate in the routing process.
119

Figure C.1: (a) Seven Mini PCB Prototyping Machine (b) Tools used for microwave artwork
of the antenna.
120

Dry-film photoresist
The second way in fabricating antenna is using photoresist technique. There are some steps in
this technique, which is dry film laminating, UV exposure and film developing. The laminating
is used to cover the antenna substrate with dry film. UV exposure is used to transfer the image
of the circuit pattern with a film in a UV exposure machine onto to the photo resist laminated
board. This process will usually take 2 minutes. To transfer the antenna on a film, CAD
(Computer Added Drawing) software is used. By using this software, the actual value of the
simulated antenna can be transferred easily. Figure C2 below shows the UV exposure machine.
Figure C.2: UV exposure machine.
Etching
The final stage of the fabrication process is etching. Etching process which removes the unwanted
copper area, immediately followed by removal of solution by water. For the etchant,
Ferric Chloride is needed. This process usually takes a few minutes depend on the size and
temperature of the solution.
121

Figure C.3: Etching tank and chemical powder.
Bonding
Bonding has the purpose to produce good electrical contacts between metallic parts by joining
surfaces.
In this case it is done by soldering the SMA connector to the edge of microstrip
feed line. Figure B.3 illustrate the soldering process for the equilateral triangular microstrip
antenna.
Figure C.4: Bonding between SMA connector and microstrip feed lines for the equilateral
triangular antenna.
122

Figure C.5: Step by step fabrication the microstrip antenna with prototyping machine.
Figure C.6: Step by step fabrication the microstrip antenna with dry film photoresist.
123