) 65 (2011) 6267

Int. J. Electron. Commun. (AEU

Contents lists available at ScienceDirect

)
Int. J. Electron. Commun. (AEU
journal homepage: www.elsevier.de/aeue

Elliptical microstrip antenna for circularly polarized synthetic aperture radar

Merna Baharuddin , Victor Wissan, Josaphat Tetuko Sri Sumantyo, Hiroaki Kuze
Microwave Remote Sensing Laboratory, Center for Environmental Remote Sensing, Chiba University, 1-33 Yayoi-cho, Inage-ku, Chiba 263-8522, Japan

a r t i c l e in f o

a b s t r a c t

Article history:
Received 8 September 2009
Accepted 19 December 2009

We propose a new circularly polarized elliptical microstrip antenna intended for use in circularly
polarized synthetic aperture radar (CP-SAR) systems operated in L-Band (1.27 GHz). The design and
fabrication process aim to fulll the challenging specication for our CP-SAR onboard unmanned aerial
vehicle (UAV). We apply the proximity-coupled, single feeding method by a microstrip line positioned
451 counterclockwise to the semi-major axis of the ellipse to generate the left-handed circularly
polarized wave radiating from the patch. Both simulated and measured results are presented. For a
single-element model, the simulation gives an axial ratio bandwidth (o 3 dB) of about 0.85%
(10.8 MHz), which is consistent with the bandwidth measured for the measured fabricated model of
0.82% (10.4 MHz) and 0.91% (11.6 MHz) simulated for an antenna array model. These results satisfy the
specication for our CP-SAR sensor onboard UAV.
& 2010 Elsevier GmbH. All rights reserved.

Keywords:
Elliptical microstrip antenna
Circular polarization
Synthetic aperture radar

1. Introduction
Synthetic aperture radar (SAR) is a multipurpose sensor that
can be operated in all-weather and both daytime and nighttime. A
circularly polarized SAR (CP-SAR) onboard micro-satellite is
currently developed in the Microwave Remote Sensing Laboratory
(MRSL) of the Center for Environmental Remote Sensing, Chiba
University, Japan and will be launched in 2014. As part of the
project, an airborne CP-SAR development is also undertaken in
order to obtain sufcient knowledge of CP-SAR sensor systems.
An L-band CP-SAR system is also being developed to be operable
on an unmanned aerial vehicle (UAV). Conventional SAR systems
have been based on linearly polarized (LP) antenna systems [1,2].
However, there are limitations due to the propagation phenomena such as the variation of geometric distance between the radar
system and the Earth, the occurrence of a phase shift when
microwave strikes smooth, reective surfaces, etc. These phenomena are suppose to lead to unwanted modulation of backscatter signals, random redistribution of returned signal-energy
and in the end, the formed image would encounter a spatially
variant blurring and defocusing as well as ambiguous identication of different low-backscatter features in a scene. Especially for
space propagation, electromagnetic waves propagating through
the ionosphere interact with electrons and magnetic elds. As a
result, the polarization vector of the electric eld is rotated by the
Faraday rotation effect [3]. Since the peak ionospheric density
occurs at about 400 km altitude, an L-band SAR onboard a micro Corresponding author. Tel.: + 81 43 290 3840; fax: +8143 290 3857.

E-mail addresses: merna5@graduate.chiba-u.jp, merna.baharuddin@gmail.com

(M. Baharuddin).
1434-8411/$ - see front matter & 2010 Elsevier GmbH. All rights reserved.
doi:10.1016/j.aeue.2010.01.012

satellite orbiting at 500700 km altitude will transmit and receive

radar signals through most of the ionosphere. As compared with
the conventional linear polarization SAR, most of the abovementioned effects can be alleviated through the use of CP-SAR. In
addition, it has been pointed out [4] that a full characterization of
SAR signals backscattered from a random object can only be
possible through the use of circular polarization. In this way, the
measurement potential of a space-based SAR can be maximized.
Thus, a CP-SAR sensor would provide a greater amount of
information about scenes and targets being imaged than a linear
SAR sensor. According to a study [5] that analyzed quadraturepolarized synthetic aperture radar data from sloping terrain, the
results of synthesis of circularly polarized data are far better
compared to those constructed from the conventional linearly
polarized data. In addition, more sensitive measurements can
possibly be obtained using circular polarization when surface
roughness is studied using polarimetric SAR data [6]. The present
work focuses on the design of an L-band CP-SAR antenna. We
consider the SAR system requirements to achieve an excellent
performance of the overall CP-SAR system, including optimization
of the single element patch and array designing.

2. Circularly polarized SAR antenna requirements

The capability of a SAR antenna can be described by its
sensitivity, spatial resolution in range and azimuth directions,
image quality, ambiguities, and swath coverage [7]. Table 1 shows
the specications and targets desired for the present CP-SAR
system, which in turn inuence the specication of the L-Band
CP-SAR antenna.

 ) 65 (2011) 6267
M. Baharuddin et al. / Int. J. Electron. Commun. (AEU

Table 1
Specication of CP-SAR onboard unmanned aerial vehicle.
Parameter

Specication

Frequency f
Chirp bandwidth D f
Polarization

1.27 GHz (L band)

10 MHz
Transmitter: RHCP or LHCP
Receiver: RHCP LHCP
4 20 dBic
o 3 dB (main beam)
2 m (azimuth) 0:7 m (range)
83 (azimuth) 253 (range)
 1:4 m (azimuth) 20 m (range)
3  10 km

Gain G
Axial ratio AR
Antenna size
Beam width
Resolution
Altitude range

Fig. 1. Conguration of a CP-SAR antenna array consisting of 5  11 microstrip

elements.

The operation frequency of 1.27 GHz (L-band) has been chosen,

since its relatively longer wavelength (l 23:6 cm) ensures better
penetration through vegetation canopies and snow, providing
useful information to distinguish characteristics of the Earth
surface. This band also provides strong returns for larger objects
and surface features [8]. The drawback associated with this
choice, however, is the relatively large dimension of microstrip
elements. The requirements for the range resolution (15 m)
determine the antenna bandwidth of 10 MHz, or less than 1% of
the operation frequency of 1.27 GHz. This 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. The primary
considerations in the design and subsequent fabrication processes
are low cost, light weight and ease of manufacturing. The CP-SAR
antenna consists of an array of single antenna elements, each
being a microstrip antenna for circular polarization. The single
element patches which have been optimized are then spatially
arranged to form a planar array (see Fig. 1 for illustration). The
planar array conguration is widely employed in radar systems
where a narrow pencil beam is needed [9]. A better control of the
beam shape and position in space can be achieved by correctly
arranging the elements along a rectangular grid to form a planar
array. The antenna gain is mostly determined by the aperture size
and inter-element separation.
Feed network is implemented in a separate substrate as the
feeding method is proximity coupled. The feeding array is parallel
to the antenna array, corresponding to the scheme of proximitycoupled, corporate feeding. This type of feed method allows better
optimization of both array structures individually. In the present

63

paper, we focus on the design, fabrication, and test of an antenna

element. In addition, a numerical simulation is carried out for an
array consisting 5  11 microstrip elements, in accordance with
the array conguration planned for an airborne CP-SAR sensor.

3. Analysis and design of the radiating elements

The requirement of a patch element for generating a circularly
polarized radiation is that the patch must have orthogonal (inphase and quadrature) elds of equal amplitude. Slightly elliptical
patch can have a circular polarization radiation with a single
feeding [10,11]. In addition, an elliptical antenna element
generally has an elliptically polarized radiation, but it has lefthanded (or right-handed) circularly polarized (LHCP/RHCP)
radiation when the feed point of the antenna element is located
on the radial line rotated 451 counterclockwise (or clockwise) to
the semi-major-axis of the ellipse [11]. Also according to [11], the
best circular polarization radiation may be achieved by limiting
the eccentricity of the ellipse to a range of 1020%. The
conguration of the radiating element together with the microstrip line feed and ground plane is shown in Fig. 2(a), where
important parameters are labeled. Although prior elliptical
patches were based on the probe-feed method [10,11], in this
paper we adopt the proximity-coupled feeding method [12].
This approach has the advantage of easier adjustment in
the design and fabrication processes, especially in producing
good circular polarization with good impedance matching. Also,
bandwidth enhancement and reduced parasitic radiation of the
feeding network is achieved compared with other direct
feeding methods. The dimensions of the radiator, and the
ground plane for the elliptical patch are a 45:9 mm, b 44:5
and la  lr 120  126:65 mm. Side view is depicted in Fig. 2(b).
The geometry model is implemented on two substrates, each
with thickness t 1:6 mm, conductor thickness tc  0:035 mm,
relative permittivity er 2:17 and dissipation factor 0.0005.
The parameters of the microstrip line feed are w 4:8 mm,
d 10:8 mm, l 48:7 mm, ls 7 mm, and ws 7 mm. With the
width of the microstrip line of 4.8 mm, the characteristic
impedance is approximately 50:9 O.

Fig. 2. Conguration of equilateral triangular patch antenna with proximity

coupled feed; (a) top view and (b) side view.

64

) 65 (2011) 6267
M. Baharuddin et al. / Int. J. Electron. Commun. (AEU

Fig. 3. Simulation results showing the frequency dependence of the axial ratio
(AR) of the elliptical microstrip antenna for various values of (a) the semi-major
axis a and (b) semi-minor axis b.

The elliptical radiator will generate LHCP by rotating the patch

by  451 around the center of the ellipse. Simulations with a nite
ground plane model have been undertaken to optimize the size
parameters using a full wave analysis tool (IE3D Zeland software)
based on the method of moment (MoM) algorithm. During the
optimization process of the elliptical patch conguration, it was
observed that the parameters a (semi-major axis) and b (semiminor axis) have a strong inuence on both the CP frequency and
the AR of the antenna. Fig. 3 shows the result of the simulation, in
which the frequency dependence of the axial ratio is plotted for
various values of the parameters a and b while keeping the other
parameters unchanged (with the optimized values a 45:9 mm
and b 44:5 mm). The best circular polarization radiation is
achieved for the eccentricity ranging from 19% to 28%. The
difference between the present result (1928%) and that in a
previous work (1020%) [11] is presumably due to the difference
in the feeding method applied to the elliptical patch.

20

40

60
mm

80

100

120

Fig. 4. Fabricated elliptical microstrip antenna.

4. Measurement of antenna performance

Fig. 5. Schematic of the measurement system.

An elliptical microstrip antenna (Fig. 4) was actually

fabricated, and its properties were tested in comparison with
the simulation results. Careful and precise fabrication process is
required to produce radiating behavior similar to the simulated
model.
The reection coefcient and input impedance were measured
with a RF Vector Network Analyzer (Agilent, E5062A, ENA-L).
The antenna gain, AR, and radiation patterns were measured
inside the anechoic chamber of MRSL, having a dimension of
4  8:5  2:4 m3 . The measurement system is schematically
shown in Fig. 5 [13]. The frequency dependence of AR and gain,
AR pattern, and gain pattern are measured using conical log spiral
LHCP, RHCP and dipole antennas as the standard reference.
Precise alignment between antenna under test (AUT) and the
conical log spiral antenna is indispensable in order to obtain
accurate measurement results.

5. Simulation and measurement results and discussion

The experimental results are shown in Figs. 610 in
comparison with the simulation. Fig. 6 shows the frequency
dependence of the S11 parameter (reection coefcient).
Although the measured parameter takes a minimum value
at 1.256 GHz, somewhat smaller than the operation frequency of
f 1:27 GHz, an impedance bandwidth (S11 o10 dB) of more
than 20 MHz is attained around the operation frequency, in spite
of the difference between the measured and simulated curves.
The input resistance and reactance are shown in Fig. 7. The
measured values of these parameters are 39:1 O for input
resistance and 3:5 O for the input reactance at f 1:27 GHz. The

 ) 65 (2011) 6267
M. Baharuddin et al. / Int. J. Electron. Commun. (AEU

65

Fig. 6. Reection coefcient vs. frequency.

Fig. 9. Gain and AR vs. theta angle (radiation pattern) in the theta plane (Az 03
and 1803 ) (xz plane) at f 1:27 GHz.

Fig. 7. Input impedance vs. frequency.

Fig. 10. Gain and AR vs. theta angle (radiation pattern) in the theta plane (Az 903
and 2703 ) (yz plane) at f 1:27 GHz.

Fig. 8. Gain and AR vs. frequency at yangle 03 .

input resistance value of the fabricated antenna shows a smaller

value by about 12 O than the simulated model, presumably due to
the resistivity contribution from the connector and solder
material and variation of the substrate parameter (dielectric
constant, loss tangent d, thickness, etc.).
In Fig. 8, the antenna gain and AR at y 03 are plotted against
the frequency. From this gure, it can be seen that whereas the
gain of the antenna is simulated to be 6.96 dBic at 1.27 GHz, the
experimental result shows a smaller value by about 0.5 dB. Such a
difference between the simulation and experimental results, as
also seen in other curves in Figs. 610, can probably be ascribed to
the fabrication imperfections (such as inaccuracy in the milling
and etching processes, connector soldering and holes with plastic
screws), the substrate loss and cable loss, and also the innite

lateral extension of the substrate in the IE3D simulation while the

fabricated one is a nite substrate with the same size as the
ground plane. As for the frequency dependence curves of AR, a
crucial parameter for circularly polarized antenna operation, Fig. 8
shows that the 3-dB AR bandwidth of the simulation is 10.8 MHz
and from observation it is 10.4 MHz, ranging from 1.2658 to
1.2762 GHz. The AR bandwidth of the simulated model has
satised the target specication (10 MHz) of CP-SAR, but the
range is slightly shifted from the ideal range of 1.2651.275 GHz.
Again, this shift is possibly caused by fabrication imperfections.
Figs. 9 and 10 show the radiation pattern in terms of the gain and
AR at an azimuth angle Az 03 (and 1801, xz plane) and 901
(and 2701, yz) plane and at the frequency of f 1:27 GHz. In
Fig. 9, a difference of around 0.5 dB is seen between the simulated
and measured results. Fig. 9 also shows that the simulated
and measured 5-dBic gain beamwidths and the simulated and
measured 3-dB AR beamwidths are 601, 541, 1501 and near 1801,
respectively. In the 901 azimuth (Fig. 10) the simulated and
measured 5-dBic gain beamwidths and the simulated and
measured 3-dB AR beamwidths are 601, 551, 1451 and 1691,
respectively. From Figs. 9 and 10, it is apparent that the measured
AR performance is better in terms of AR beamwidth than the
simulated one. This may be inuenced by the imperfection effects
mention above. It can also be noticed that there is a slight shift of
gain pattern, and particularly the 901 azimuth gain pattern has a
little uctuation. This is possibly due to the measurement system,
i.e. the slight variations in antenna alignment during rotation.

66

) 65 (2011) 6267
M. Baharuddin et al. / Int. J. Electron. Commun. (AEU

Acknowledgments
We thank to Basari, Muhammad Fauzan, Prilando R. A., Ilham
A. and Sugianto for assisting in the antenna fabrication and
measurement; the Japan Society for the Promotion of Science
(JSPS) for Grant-in-Aid for Scientic Research - Young Scientist (A)
(no. 19686025); National Institute of Information and Communication Technology (NICT) of Japan for International Research
Collaboration Research Grant 2009 and this work was supported
in part by the research grant for Mission Research on Sustainable
Humanosphere from Research Institute for Sustainable Humanosphere (RISH), Kyoto University, Japan.
References

Fig. 11. Simulated radiation pattern of a 5  11 elliptical microstrip array at

f 1:27 GHz.

Fig. 11 shows the antenna gain patterns (Az 03 and 903 ) for a
5  11 elliptical microstrip element array simulated using Zeland
IE3D on an innite ground, with each elliptical element fed
individually. The antenna arrangement is as shown in Fig. 1. The
elevation spacing between elements is 125 mm and the azimuth
spacing is 140 mm. Fig. 11 shows the resulting co- and crosspolarization composition patterns from RHCP elements. From this
gure, the gain at y 03 is 17.6 dBic, somewhat lower than the
target gain of 20 dBic (Table 1). However, the beam patterns are
mostly satisfactory for CP-SAR operation in both the azimuth and
elevation planes. The nearest side lobe is suppressed by about
13 dBic. From the simulation, it has also turned out that the array
achieves a 3-dB AR bandwidth of 11.58 MHz with the center
frequency of 1.27 GHz. Mutual coupling between elements in the
array can be inspected by observing the active element pattern and S
parameters from the IE3D simulation. The result has shown that no
scan blindness occurs at the main beam (with a beamwidth of
approximately 1001) and mutual coupling between elements are
lower than  17 dB. Further implementation of the array
conguration is in progress in our laboratory.

6. Conclusion
A circularly polarized antenna has been developed for
implementing an antenna for circularly polarized synthetic
aperture radar (CP-SAR) operated in L-band. The design and
optimization process was carried out using a MoM analysis
software. The single-element model was actually fabricated and
experimentally tested in our MRSL. From both of the simulation
and measurement results, it can be inferred that the elliptical
patch has mostly satised the target specication for a CP-SAR
antenna element to be used onboard UAV. The slight difference
between the simulated model and the measurement results is
probably due to the fabricated antenna conguration with holes
for plastic nuts, connectors, etc. The performance of an elliptical
microstrip array has also been examined with good results. On the
basis of knowledge on the antenna element and array conguration described in this paper, the array conguration will be
further designed, simulated, and tested in order to realize the CPSAR antenna onboard UAV.

[1] Nemoto Y, Nishino H, Ono M, Mizutamari H, Nishikawa K, Tanaka K. Japanese

Earth Resources Satellite-1 synthetic aperture radar. Proc IEEE 1991;79:
8009.
[2] Raney RK, Luscombe AP, Langham EJ, Ahmed S. RADARSAT. Proc IEEE
1991;79:83949.
[3] Rignot Eric JM. Effect of faraday rotation on L-band interferometric and
polarimetric synthetic-aperture radar data. IEEE Trans Geosci Remote Sensing
2000;38:38390.
[4] Raney RK. Hybrid-polarity SAR architecture. IEEE Trans Geosci Remote
Sensing 2007;45:3397404.
[5] Lee J, Schuler DL, Ainsworth TL. Polarimetric SAR data compensation for
terrain azimuth slope variation. IEEE Trans Geosci Remote Sensing
2000;38:215363.
[6] Mattia F, Toan TL, Souyris J, De Carolis G, Floury N, Posa F, et al. The effect of
surface roughness on multifrequency polarimetric SAR data. IEEE Trans
Geosci Remote Sensing 1997;35:95466.
[7] Pokuls R, Uher J, Pozar DM. Dual-Frequency and dual polarization microstrip
antennas for SAR applications. IEEE Trans Antennas Propagation
1998;46:128996.
[8] Birk R, Camus W, Valenti E, McCandless WJ. Synthetic aperture radar imaging
systems. IEEE Aerosp Electronic Syst Mag 1995;10:1523.
[9] Garg R, Bhartia P, Bahl I, Ittipiboon A. Microstrip antenna design handbook.
Norwood MA: Artech House; 2001.
[10] Bailey MC, Deshpande MD. Analysis of elliptical and circular microstrip
antennas using moment method. IEEE Trans Antennas Propagation
1985;33:9549.
[11] Shen CS. The elliptical microstrip antenna with circular polarization. IEEE
Trans Antennas Propagation 1981;29:904.
[12] Pozar DM, Kaufman B. Increasing the bandwidth of a microstrip antenna by
proximity coupling. Electron Lett 1987;23:3689.
[13] Wissan V, Baharuddin M, Sri Sumantyo JT, Kuze H. Development of
measurement system for circularly polarized synthetic aperture radar. In:
The 16th remote sensing forumSICE, RESTEC, Tokyo, 2009. p. 1112 (in
Japanese).

Merna Baharuddin received her B.E. degree in electrical engineering specializing in Telecommunication
from Hasanuddin University, Indonesia, in 1999,
received her M.E. degree in Telecommunication Engineering from Monash University, Australia, in 2002.
She is a lecturer in Electrical Engineering Department,
Engineering Faculty, Hasanuddin University, Indonesia. She is now a Doctoral Student in Chiba University,
Graduate School of Advanced Integration Science.

Victor Wissan received his B.E. degree from Chiba

University Faculty of Engineering, in April 2009. He is
now a Master Student in Chiba University, Graduate
School of Advanced Integration Science.

 ) 65 (2011) 6267
M. Baharuddin et al. / Int. J. Electron. Commun. (AEU

Josaphat Tetuko Sri Sumantyo was born in Bandung,

Indonesia, 1970. He received the B.Eng. and M.Eng.
degrees in electrical and computer engineering from
Kanazawa University, Japan, in 1995 and 1997,
respectively, and the Ph.D. degree in articial system
sciences from Chiba University, Japan, in 2002. From
1990 to 1999, he was a Researcher with the Indonesian
Governmental Agency for Assessment and Application
of Technology (BPPT) and the Indonesian National
Army (TNI-AD). In 2000, he was with the Center for
Environmental Remote Sensing (CEReS), Chiba University, as a Research Assistant, and from 2002 to 2005,
he was a Post Doctoral Fellowship Researcher with the
Center for Frontier Electronics and Photonics, Chiba University. He has been a
director with the Remote Sensing Research Center, Pandhito Panji Foundation,
Indonesia since 2000. From 2005, he is permanently as Associate Professor with
the Microwave Remote Sensing Laboratory, CEReS, Chiba University. His main
interests include analysis of printed antennas for mobile satellite communications,
aircraft and microsatellite onboard synthetic aperture radar, scattering wave
analysis and its applications in microwave remote sensing. He is a Member of the
IEEE, IEICE, JSPRS, and RSSJ and has received many awards and research grants
related to his study and research.

67

Hiroaki Kuze nished the graduate school of University of Tokyo in 1982, obtaining the Ph.D. degree in
physics in the eld of quantum electronics. After
working as a post-doctoral researcher for the Institute
for Physical and Chemical Research (RIKEN), in 1984 he
became a research assistant and in 1986 an associate
professor in the physics department, faculty of liberal
arts, Shizuoka University. During 19871988, he
stayed at the Institute for Quantum Optics, Max-Planck
Institute in West Germany as a visiting scientist. From
1995 to 2004, he worked as an associate professor with
the Center for Environmental Remote Sensing (CEReS),
Chiba University, and in 2004 he became a professor.
His major research eld is optical remote sensing of the atmosphere.