A noise based transmitter system for the HITCHHIKER projectSimon Reuter, Florian Behner, Holger Nies, Otmar Loffeld Center for Sensorsystems (ZESS), University of Siegenhitchhiker@zess.uni-siegen.de Paul-Bonatz-Strasse 9-11, 57076 Siegen, Germany

Abstract

The HITCHHIKER project was started in 2009 with the goal to develop a passive high resolution receiver system toperform a series of bistatic experiments using the German TerraSAR-X system as illuminator and demonstrating the fea-sibility of high resolution bistatic imaging with a stationary receiver. This paper describes valuable extensions augmentingthe system to a fully functional bistatic one pass interferometer to acquire high resolution DEMs on the one hand and fur-ther integrating a modular and flexible high bandwidth transmitter component with a thermal noise based signal sourcefor waveform generation to upgrade the system to a fully operational active bi- or monostatic (standalone) SAR sensor.A first experimental result with a noise based transmitter and the HITCHHIKER receiver and an outlook on upcomingexperiments with this new system in 2012 completes the work.

1 Introduction

The HITCHHIKER project, inspired by the Spanish SAB-RINA experiment [1], was started in 2009 [2] with thegoal to develop a receiver system to perform a series ofbistatic experiments with DLR’s spaceborne TerraSAR-Xsystem demonstrating the feasibility of high resolution pas-sive bistatic SAR imaging with a stationary receiver. Thehigh bandwidth of the transmitter and especially its spot-light SAR mode extending the synthetic aperture used bythe stationary receiver system promised interesting results[3].Basically designed as a two channel receiver system, withone channel as a reference channel, capturing the directillumination signal and triggering the acquisition systemfrom pulse to pulse, the second channel captured the targetor scene signal. Based on this hardware setup new process-ing algorithms were developed and experimentally verifieddelivering high resolution (1m) SAR images. After per-forming several multi pass bistatic interferometric exper-iments using TerraSAR-X it was observed that temporaldecorrelation was severe in vegetated or forested areas[4].Hence the decision was taken to extend the receiver to fourreceiving channels. The additional receiving channels alsoallow for receiver side polarimetry which enables us to ac-quire full polarimetric images in combination with the dualpolarimetric imaging mode of the transmitter.To achieve a high resolution Radar, it is necessary to workwith high bandwidth signals, whereas the type of wave-form is nearly irrelevant. In the past, the generation of suchhigh bandwidth signals was performed by sweeping thefrequency of a sinusoidal waveform generator. This princi-ple leads to deterministic waveforms. The pulse compres-sion is done digitally or in an analogue way.However, it is not necessary to use deterministic signalwaveforms to compress a high bandwidth signal to a very

short pulse. Recording the transmitted signal enables toperform a pulse compression digitally and allows to usethermal noise as a high bandwidth signal source. This isshown in [5], [6] and called Noise Radar.

This paper describes a new noise based transmitter sys-tem as an extension for the HITCHHIKER project. Thebandwidth of the used noise source is in X-band between8 to 12GHz. But it is limited to 500MHz to a center fre-quency of 9.65GHz, due to the specifications of the exist-ing HITCHHIKER receiver system. It is planned to extendthe bandwidth in the future by replacing the bandpass fil-ters.

In the following sections the extended receiver setup is dis-cussed and the designed transmitter system is presented.The paper concludes with an experimental verification ofsuch a noise based transmitter system as an extension forthe HITCHHIKER project and an overview of the experi-ments planned for 2012.

2 Extended Four Channel Receiver

System

In extending the receiver frontend to four channels the al-ready existing two channels had to be reused, also leavingthe form factor of the receiver unchanged. In addition tothe direct receiving channel, which continues to be be usedfor synchronisation and capturing the reference signal wehave three receiving channels that are used for imaging.Two channels are connected to an interferometric pair ofantennas while the remaining channel is used with a differ-ent polarization for polarimetric acquisition.

EUSAR 2012

235

Figure 1: Interferometric antenna setup of the stationaryreceiver.

To ensure stable baselines for interferometry and in orderto know the positions of the antennas, the receiver wasequipped with an new interferometric antenna setup. Thefour antennas are fixed to a reference geometry which ismounted on a tripod where the geometry is levelled usinga tribrach. Position and heading of the setup is determinedusing two differential GPS receivers. Figure 1 shows aphotograph of the antenna setup.The phase synchronization between the bistatic platformsis done by matched filtering with the direct channel of thereceiver. Additionally the clocks and oscillators are syn-chronized using GPS for the bistatic case.

3 Transmitter System

To be more flexible in configuration and performance ofexperiments also in order to achieve standalone capabilitiesit was further decided to develop and operate an own trans-mitter system. In conventional Radar systems for SARimaging the common waveform is deterministic like thelinear frequency modulated "Chirp" signal. While this sig-nal is easy to describe it is technologically demanding togenerate high bandwidth signals digitally. Encouraged byLukin’s research and the fact that the receiver system im-plicitly captures a replica of every transmit pulse, the ideacame up to develop a transmitter based on white thermalnoise [5]. As the analogue signal is directly generated inthe X band, the transmitter system design for the noisewaveform is simple and shown as "Noise Waveform" blockin figure 2.To evaluate the system and for a first test, we performed anindoor experiment with the transmitter prototype which isdiscussed in section 4.Besides transmitting a noise waveform, the system alsois intended to be capable of transmitting the conventionalchirp waveform. The chirp waveform is generated bymeans of a DDS in complex baseband, then upconverted,amplified and transmitted by the same power amplifierstage used for the noise waveform. The DDS is imple-mented in a "commercial off-the-shelf" integrated circuitand with its sampling rate of 1GSPS capable of generat-ing chirp waveforms with a bandwidth of up to 800MHz.The proposed system is shown in figure 2.As the system is intended for outdoor use it is planned toperform a series of experiments imaging an area of 3-6 kmin range. In combination with the HITCHHIKER receiversystem the transmitter will be designed to expect a noiseequivalent sigma zero of −42 dB like in the TerraSAR-X– HITCHHIKER experiments [2].

kT

DDS I

DDS QDirect Signal OutReference Signal In

Pulse Modulation In

Noise Waveform Transmitter Stage

Chirp Waveform

Figure 2: Simplified block diagram of the transmitter system

236

Figure 3: Photograph of the scene

0 2 4 6 8 10 12

0

2

4

6

8

cross range in meter

rang

ein

met

er

Figure 4: Processed result

4 First Experiment

We performed an indoor experiment with a prototypetransmitter and the HITCHHIKER receiver. This proto-type consists of a thermal noise source in X-band, whichis amplified by a 42 dB low noise amplifier. The signal ispulsed to trigger the HITCHHIKER system with the directchannel. The resolution in range direction of about 30 cmdepends on the bandwidth of the transmitted signal, whichwas limited to 500MHz. Due to the gain of the antennasof 20 dBi the aperture angle is very large. This leads to ahigh resolution of about 5 cm in cross range direction. Thetarget area of this experiment was the staircase of the ZESSbuilding. We used a rolling cart to move the transmitter andreceiver (monostatic case) along the gallery, to be able toperform SAR processing techniques for cross range focus-ing. The system looked straight down to the stairs. Figure3 shows a photograph of the target area.In figure 4 an image of the processed data is shown, fo-cused with a SIFT processor. As a destinctive target a

mannequin is used, which is clearly visible in the focusedresult, due to the high resolution in cross range direction.Furthermore each stair can be discriminated.

5 Conclusion and Outlook

The straightforward design of the presented system leadsto a cost effective high bandwidth transmitter suitable forhigh resolution SAR imaging. Due to the direct channelof the HITCHHIKER system, recording of the transmittedsignal is done automatically for pulse compression. Thepromising results encourage us to carry on the develop-ment.For 2012 we have planned to perform a series of experi-ments with different configurations on different platforms.The first experiment is again a spaceborne–stationarysetup. We plan to acquire a DEM using the interferometricphase at the stationary receiver, which will be the first testof the new interferometric setup. While an interferometric

237

phase was already acquired in earlier experiments, the gen-eration of an usable DEM failed due to the unknown geo-metrical relation between the receiving antennas. The illu-minating transmitter in this experiment is the TerraSAR-Xsystem.In a series of monostatic experiments with a truck-mountedsensor we are planning to test the transmitter system andespecially the generation of a synthetic aperture using thenoise waveform. The expected resolution for these exper-iments is about 30 cm in range and 5 cm in cross range.The range resolution is planned to be extended later to7.5 cm by filter replacement and thus an increase of thesignal bandwith to 2GHz.Further it is planned to perform bistatic experiments withthe sensor system as it was originally designed for thebistatic configuration. This includes the configurationstruck–stationary and the airborne–stationary case.

6 Call for proposals

We want to encourage the reader to contact the authors inorder to propose new ideas for experiments and applica-tions. Requests for cooperation are also welcome if thereader is interested in working with data acquired withthe HITCHHIKER system. Please contact us by emailhitchhiker@zess.uni-siegen.de.

References

[1] J. Sanz-Marcos and P. Lopez-Dekker et al., “SAB-RINA: A SAR bistatic receiver for interferometric ap-

plications,” Geoscience and Remote Sensing Letters,IEEE, vol. 4, no. 2, pp. 307–311, April 2007.

[2] Florian Behner and Simon Reuter, “HITCHHIKER,hybrid bistatic high resolution SAR experiment usinga stationary receiver and TerraSAR-X transmitter,” in8th EUSAR, Aachen, Germany, 2010.

[3] Simon Reuter, Florian Behner, Holger Nies, OtmarLoffeld, Dietmar Matthes, and Joachim Schiller, “De-velopment and experiments of a passive SAR receiversystem in a bistatic spaceborne/stationary configura-tion,” in Geoscience and Remote Sensing Symposium(IGARSS), 2010 IEEE International, 2010, pp. 118 –121.

[4] Holger Nies, Florian Behner, Simon Reuter, OtmarLoffeld, and Robert Wang, “Polarimetric and inter-ferometric applications in a bistatic hybrid SAR modeusing TerraSAR-X,” in Geoscience and Remote Sens-ing Symposium (IGARSS), 2010 IEEE International,2010, pp. 110 –113.

[5] D. Tarchi, K. Lukin, J. Fortuny-Guasch, A. Mogyla,P. Vyplavin, and A. Sieber, “SAR imaging with NoiseRadar,” Aerospace and Electronic Systems, IEEETransactions on, vol. 46, no. 3, pp. 1214 –1225, 2010.

[6] K.A. Lukin, A.A. Mogyla, V.P. Palamarchuk, P.L. Vy-plavin, O.V. Zemlyaniy, Y.A. Shiyan, and M. Zaets,“Ka-band bistatic ground-based noise waveform SARfor short-range applications,” Radar, Sonar Naviga-tion, IET, vol. 2, no. 4, pp. 233 –243, 2008.

238