A Concept for High Performance Reflector-Based Synthetic Aperture Radar

Abstract

The success of current spaceborne Synthetic Aperture Radar (SAR) is boosting the performance requirement of next generation systems. In order to cope with the evolution of SAR, the design of the new systems will need to meet higher requirements for spacial and radiometric resolutions together with an increased availability. This tendency is recognized nearly independently of the application area and manifests itself through several study programs from different space agencies aiming at the design of future SAR systems. In this context the use of large reflectors combined with digital feed arrays for SAR is a highly attractive concept. This paper suggests a spaceborne SAR system utilizing a a deployable reflector together with a digital feed array and elaborates into the various design and operation issues together with the resulting instrument SAR performance.

Full text

Page 1

A CONCEPT FOR HIGH PERFORMANCE
REFLECTOR-BASED SYNTHETIC APERTURE RADAR
Marwan Younis, Anton Patyuchenko, Sigurd Huber, Gerhard Krieger, and Alberto Moreira
German Aerospace Center (DLR), Microwaves and Radar Institute
ABSTRACT
The success of current spaceborne Synthetic Aperture Radar
(SAR) is boosting the performance requirement of next gener-
ation systems. In order to cope with the evolution of SAR the
designofthe newsystems will needto meethigherrequirements
for spatial and radiometric resolution together with an increased
availability. This tendency is recognized nearly independently
of the applicationarea and manifests itself throughseveral study
programs initiated by space agencies aiming at the design of fu-
ture SAR systems. In this context the use of large reflectors
combined with digital feed arrays for SAR is considered a pos-
sible alternativeto planararrayantennas. This papersuggests an
X-band spaceborne SAR system utilizing a deployable reflector
together with a digital feed array, analyzes its performance and
highlights its advantages compared to other systems based on
direct radiating arrays.
1. INTRODUCTION
A review of several ongoing studies for the conception of next
generationSAR systems, reveals the shared characteristic of be-
ing multi-channel systems utilizing digital beamforming (DBF)
techniques. Examples are the High Resolution Wide Swath
(HRWS) SAR [1, 2, 3] and Tandem-L/DESDnyI mission [4, 5].
The common purpose of using multi-channel systems is to
simultaneously obtain high spatial resolution and wide swath.
For DESDnyI/Tandem-L a reflector-based SAR system was
firstsuggested,whichwas laterextendedto ahybridarchitecture
through a digital feed [6]. With this system it would be possi-
ble to image a swath width of 300 − 400km [4]. Such a hybrid
architecture has the potential to combine both the flexibility and
the capabilities of DBF with the high antenna gain provided by
a large reflector aperture. To lower the stowed satellite volume
andweight, and thereforethe launchcosts, the reflectorcouldbe
deployable. Unfurlable reflector antennas are a mature technol-
ogy with extensive flight heritage in space telecommunications;
satellites with lightweightmesh reflectorsspanningdiameters of
> 20m are already deployed in space [7].
From the above it seems reasonable to consider reflector-
based SAR systems for the future. In [8] a planar and a reflector
system were designed to a common set of performance param-
eters; the comparison revealed that the reflector system can be
realized with a simpler hardware and shows a performance ad-
vantageof several dBs in terms of ambiguityandsignal-to-noise
ratio.
The paper addresses this issue by suggesting a SAR sys-
tem utilizing a reflector in conjunction with a digital feed array.
Keeping future follow-up systems for the German TerraSAR-X
andTanDEM-XSAR satellites in mind,thereflectorsystemwill
bedesignedforX-bandoperationwithperformancerequirement
possibly exceeding those of HRWS [1]. In this paper empha-
sis will be given to the various operation modes and the perfor-
mance; the antenna design is detailed in [9], while [10] elab-
orates on the performance improvement using dedicated DBF
techniques and [11] addresses the issue of imaging gap removal
by varying the pulse repetition frequency (PRF).
2. ARCHITECTURE AND OPERATION
In 1981 Blyth [12] suggested a basic approach for analog beam-
steering for a reflector system such that the receive beam moves
over the swath in accordance with the direction of reflection.
About twenty years later, his idea finds a more detailed descrip-
tion and justification in two independent and almost contempo-
rary works [13, 14]. Digital beamforming techniques in eleva-
tion and azimuth for a reflector are presented for the first time in
[6].
In the following the digital beamforming technique and the
corresponding system architecture is addressed. For clarity this
will begivenseparatelyfortheelevationandazimuthdirections.
Nel
1
2
signal gen.
reflector
flight
direction
slant
range
ADC
ADC
ADC
feed
elements
AMP
AMP
AMP
DBF
T/R-Module
Fig. 1. System architecture for reflector system; shown here for
a single azimuth but multiple elevation channels.

Page 2

2.1. Digital Beamforming in Elevation
The system (in elevation) consists of a parabolic reflector and a
feed array of Nelantenna elements fed through transmit/receive
modules, where an analog-to-digital converter (ADC) is placed
after each T/R-module as shown in Fig. 1.
The SCan-On-REceive (SCORE) mode of operation [14],
which is also suggested here, is primarily based on generating
a wide transmit beam that illuminates the complete swath and a
narrow, high gain receive beam that follows the pulse echo on
the ground. SCORE results in an increased signal-to-noise ratio
compensatingthelowgainofthetransmitantennaandsuppress-
ing range ambiguities.
Activating all elements on transmit will generate a narrow
beam illuminating a small portion of the reflector and by this
gives a wide low gain beam illuminating the complete swath.
On receive, the energy returned from a narrow portion of the
groundilluminates the entire reflector and is focusedon individ-
ualelementsofthefeedaperture. Eachradarpulsetraversingthe
swath causes the focused energy to sweep through all the feed
elements within the time period of one pulse repetition interval
1/PRF reducedbythetransmitpulse duration. Fora highPRF
multiple portions of the swath are illuminated instantaneously
but each will activate a different subset of feed elements due to
the different angle-of-arrival. Here, DBF consists of a complex
summation of one or more subsets of the feed elements [10].
2.2. Digital Beamforming in Azimuth
Areflectorsystemofmultipleazimuthphasecenterswill require
multiple feeds separated in the along track direction as shown in
Fig. 2. Here each azimuth element “looks” at a different angle
and by this covers a distinct angular (Doppler) segment. Thus
each element samples a narrow Doppler spectrum correspond-
ing to the half-power-beamwidth of the corresponding pattern.
The PRF must be high enough such that the spatial sampling
for each channel is adequate. This is approximately given by
PRF > 2·V/D with the platform velocity V and the reflector
diameter D. If the Doppler spectra of the elements are contigu-
ous,theyjointlyyielda higherazimuthresolution≈ D/(2Naz).
EachoftheNazRxchannelscarriesnon-redundantinformation.
Doppler
span 4
beam 3 beam 1beam 2 beam 4
Doppler
span 1
Doppler
span 2
Doppler
span 3
El 1 El 2 El 3 El 4
flight
direction
Fig. 2. Concept of digital beamforming in azimuth.
3. SYSTEM DESIGN AND REQUIREMENTS
The system is not designed for a specific set of user require-
ments which causes some difficulty in quantifying the SAR per-
formance requirements. To overcomethis difficulty we state the
requirements based on state-of-the-art SAR systems. The later
analysis will then give more insight into the actual performance
and the possible compromises. The requirements are a resolu-
tion of≈ 1 x 1m, an ambiguity-to-signalratio of−20dB, anda
noise-equivalentsigma zeroof−20dB. We assumean available
average power of 2kW which matches the value for the HRWS
system [1].
3.1. Orbit Selection
The orbit selection is closely related to the intended application
area and plays a crucial role for the reflector and feed design.
Increasing the orbit height allows the reduction of the incidence
angle range for the same swath width. It is noted that for direct
radiating arrays increasing the orbit height is disadvantageous
because of the degraded SNR or increased antenna size; in the
case of a reflector system, effective antenna areas in the order
of 15 − 40m2can be realized, which makes higher orbits an
attractive option since it releases the elevation scanning require-
ments.
For the mixed scientific and commercial application con-
sidered here the access range should be such that any arbitrary
region on the Earth can be imaged at least once during the re-
peat cycle, however, global coverage within one repeat cycle is
even preferable. The table below shows the access range and
repeat cycle time for valid orbit configurations for orbit heights
in the order of 750km. Global coverage within one repeat cy-
cle is possible if the contiguous imaged swath of a single pass
is equal to the access range given in the table. The reflector
system is designed to image any sub-swath within the access
range for any configuration. The instantaneous imaged swath
will be310kmthusallowingaglobalcoveragewithin≥ 9days.
repeat cycle day
access range km
78911
252
12
232397 349309
3.2. Reflector and Digital Feed
The antenna and feed design is elaborated in [9]. Here the de-
scription is restricted to reporting the values of the system used
for the performance analysis. A circular rim reflector of diam-
eter D = 6m, focal length F = 6.2m, and vertical (elevation)
feed offset of 0.82m is designed. The feed array consists of 4 x
36 digital channels in azimuth and elevation, respectively. Each
channel (connected to an ADC) may consist of one or more ra-
diators.
4. MODES OF OPERATION
The digital feed allows operation in various modes which basi-
cally differ in the number of imaged sub-swathes and the way

Page 3

they are combined into one larger swath. These modes can be
dividedintofourcategories,whereFig.3showsanexampletim-
ing diagram of each category:
• Single Stripmap This mode is well known from conven-
tional SAR, where any sub-swath within the access range
is imaged with a single burst and constant PRF. As men-
tioned in section 2.2 the azimuth processing needs to be
adapted to the fact that each channel samples a narrow
Dopplerspectrumwhichare combinedduringprocessing.
• Multi Stripmap Here multiple sub-swathes of the same
PRF are imaged simultaneously allowing an increase of
the total swath up to the access range. However, the im-
aged swath contains gaps (c.f. Fig. 3(b)) caused by the
transmit instances. The gaps width is proportional to the
pulse duty cycle. This mode takes advantage of DBF,
since several SCORE beams are generated each one fol-
lowing the receive echo within one sub-swath.
• ScanSAR In this mode multiple bursts are used to in-
crease the swath width. For the system shown here, a
total of six to seven bursts would be required to cover
the complete access range. An alternative would be to
use ScanSAR to fill the gaps of the Multi-Stripmapmode;
this would allow operation with only two bursts to image
the complete access range. In any case the ScanSAR re-
quires an adaption of the azimuth processing, which can
become a challenge.
• PRF Variation Multiple sub-swathes are imaged at the
same time but in additionthe PRF is varied from pulse to
pulse. Bythis thegapsoftheMulti-Stripmapmodecanbe
avoided. This mode offers a highly attractive way to im-
agean ultrawideswath butrequiresinnovativeprocessing
approaches [11].
5. SAR PERFORMANCE
In the following the performanceof the Multi-Stripmap mode is
shown, since it is the most attractive one. The performance of
the Single-Stripmap is identical to that of any single sub-swath.
For the impact of the PRF-Veriation we refer to the separate
investigation in [11].
Therangeperformancegivenintermsoftherange-ambiguity-
to-signal ratio (RASR) is shown in Fig. 4. Note that the echo
signals from all the sub-swathes arrive simultaneously, some-
thing that typically causes range ambiguities in a conventional
SAR (see Fig. 3(b)), here the narrow Rx SCORE pattern allows
adequate range ambiguity suppression. The RASR is below
−40dB which is possible because of the narrow low sidelobe
SCORE Rx patterns shown exemplarily in Fig. 4(b).
The azimuth performance is given in terms of the azimuth-
ambiguity-to-signal ratio (AASR) shown in Fig. 5. The AASR
suffers from the degradation of the azimuth patterns at the near-
range boundaryof the access range. This is basically an antenna
(a) Single Stripmap(b) Multi Stripmap
(c) ScanSAR(d) PRF Variation
Fig. 3. The timing diagram for different operation modes for a
745km orbit and a pulse duty cycle of 10%.
(a) RASR
(b) elevation pattern
Fig. 4. Elevation ambiguity performance of reflector system.
and feed dimensioning issue which involves a compromise be-
tween the allowable maximum size and the performance. Here
the AASR computed for a single azimuth channel is representa-
tive for the overall AASR as explained in section 2.2.
(a) AASR
(b) azimuth pattern versus Doppler
Fig. 5. Azimuth ambiguity performance of reflector system.
The noise-equivalent sigma-zero for a total average trans-
mit power of 2kW, a 2-way system loss of 3dB, and a system
noise temperature of 460K is shown in Fig. 6(a). One charac-
teristic of the reflector system with a transmit feed connected
to T/R-modules (TRM) is that the power density on the ground
does not decrease when illuminating a wider swath. As such the
NESZ shown in Fig. 6(a) is valid independently of the number

Page 4

of imaged sub-swathes, however the average power consump-
tion increases with the number of imaged sub-swathes.
(a) NESZ
(b) pulse extension loss (PEL)
Fig. 6. Noise-equivalent sigma-zero and pulse extension loss of
the reflector system.
In Fig. 6(b) the pulse extension loss (PEL) is shown. This
performance parameter describes the loss due to the non-
vanishing pulse extension on the ground which is attenuated
by the narrow RX SCORE antenna pattern. For the 10% duty
cycle this loss is below 2dB.
6. CONCLUSION
Spaceborne SAR systems utilizing reflector antennas offer the
possibility to improve the SAR performance. This performance
improvement manifests itself through an increased swath width
and a higher signal-to-noise ratio. The digital feed of a reflec-
tor system uses a smaller number of T/R-modules and by this
require a higher average power per T/R-modules to radiate the
same total power. Further the imaging modes shows a high
potential for systems operating at low pulse duty cycle which
require a higher peak power. The power requirements are not
fulfilled with current T/R-modules at X-band and thus require
future technology development; here GaN technology seems to
be a promising candidate. On-going research on advanced dig-
ital beamforming techniques for reflector systems show a high
potential and should be continued.
7. REFERENCES
[1] C. Fischer, C. Schaefer, and C. Heer, “Technology de-
velopment for the HRWS (High Resolution Wide Swath)
SAR,” in International Radar Symposium IRS’07, Sep.
2007.
[2] F. Bordoni, M. Younis, E. M. Varona, N. Gebert, and
G. Krieger, “Performanceinvestigationon scan-on-receive
and adaptive digital beam-forming for high-resolution
wide-swath synthetic aperture radar,” in Proceedings In-
ternational ITG Workshop of Smart Antennas, Berlin, Ger-
many., Feb. 2009.
[3] C. Schaefer, M. Younis, and M. Ludwig, “Advanced sar
instrument based on digital beam forming,” in Proceed-
ings Advanced RF Sensors and Remote Sensing Instru-
ments (ARSI), Noordwijk, The Netherlands, Nov. 2009.
[4] A. Moreira, I. Hajnsek, G. Krieger, K. Papathanassiou,
M. Eineder, F. De Zan, M. Younis, and M. Werner,
“Tandem-L: Monitoring the earth’s dynamics with insar
and pol-insar,” in Proceedings International Workshop on
Applications of Polarimetry and Polarimetric Interferome-
try (Pol-InSAR), Frascati, Italy, Jan. 2009.
[5] A. Freeman, G. Krieger, P. Rosen, M. Younis, W. Johnson,
S. Huber, R. Jordan, and A. Moreira, “SweepSAR: Beam-
forming on receive using a reflector-phased array feed
combination for spaceborne SAR,” in Proceedings IEEE
RadarConference(RadarCon’09),Pasadena,U.S.A.,May
2009.
[6] G. Krieger, N. Gebert, M. Younis, and A. Moreira, “Ad-
vanced synthetic aperture radar based on digital beam-
forming and waveform diversity,” in Proc. IEEE Radar
Conference (RadarCon), Rom, Italy, May 2008.
[7] (2009, Oct.) Introducing terrestar’s integrated satellite-
terrestrial system. TerreStar Corporation. [Online]. Avail-
able: http://www.terrestar.com/technology.php
[8] M. Younis,
doni, and G. Krieger, “Performance comparison of
reflector-andplanar-antenna
forming SAR,” Int. Journal of Antennas and Prop-
agation, vol. 2009, Jun. 2009. [Online]. Available:
http://www.hindawi.com/journals/ijap/2009
S. Huber,A. Patyuchenko,F. Bor-
baseddigitalbeam-
[9] A. Patyuchenko, M. Younis, S. Huber, and G. Krieger,
“Optimization aspects of the reflector antenna for the digi-
tal beam-forming SAR system,” in Proc. European Con-
ference on Synthetic Aperture Radar EUSAR’2010 (ac-
cepted), 2010.
[10] S. Huber, M. Younis, A. Patyuchenko, and G. Krieger,
“Digital beam forming techniques for spaceborne reflector
SAR systems,” in Proc. EuropeanConference onSynthetic
Aperture Radar EUSAR’2010 (accepted), 2010.
[11] N. GebertandG.Krieger,“Ultra-wideswathSAR imaging
with continuous PRF variation,” in Proc. European Con-
ference on Synthetic Aperture Radar EUSAR’2010 (ac-
cepted), 2010.
[12] J. H. Blythe, “Radar systems,” United States Patent Patent
4253098, Feb., 1981.
[13] J. T. Kare, “Moving receive beam method and apparatus
for synthetic aperture radar,” United States Patent Patent
6175326, Jan., 2001.
[14] M. Suess and W. Wiesbeck, “Side-looking synthetic aper-
ture radar system,” European Patent Patent EP 1 241 487,
Sep., 2002.