The DLR SAR calibration center is a facility for efficient and robust SAR system calibration – enabling extended field campaigns – for multimode SAR missions as successfully executed for TerraSAR-X and TanDEM-X as well as for Sentinel-1A and Sentinel-1B.
It is well equipped with a large number of accurate passive and active reference targets (see below) and includes several analysis and evaluation tools (see below) based on precise algorithms. Within an area of 120 km x 40 km, 37 targets are deployed and maintained in South Germany at different sites. Within this DLR calibration site (see Figure 4) six of these test sites are remotely controlled and operated from Oberpfaffenhofen.
Several software tools have been developed and established for analyzing and evaluating the different measurements executed for the calibration and the verification of spaceborne SAR systems. This comprise both: the derivation of various calibration parameters like the absolute calibration factor, antenna patterns or instrument drift parameters, and the determination of several image parameters like the resolution, PSLR or ISLR for calibrating and assessing the quality of the SAR data products respectively.
The multitude of analysis tools are summarized by:
One example of these tools is given in Figure 5, a screenshot of CALIX showing the impulse response function of a point target within a SAR image and extracted in range and in azimuth. Based on this point target analysis measurements of several impulse response function parameters can be derived, like the integrated point target energy (for deriving the absolute calibration factors), target/clutter ratios (to weight calibration factor estimates), as well as peak amplitude and phase estimates in the case of multi-channel (e.g. quad-pol) data. Beyond that CALIX offers tools for distributed target analysis and for geometric calibration (internal delay and pixel localization accuracy derived from accurately surveyed targets).
Man-made point targets are an essential part of external calibration and serve as absolute radiometric and geometric reference. They are mainly deployed during the commissioning phase for the proper calibration and verification activities. Strong requirement for the SAR system on radiometric and geometric accuracy can only be achieved with highly accurate and stable point targets, and their position on Earth surface have to be precisely surveyed.
Point targets are generally distinguished in passive and active targets depending on the manipulation of the received signal: passive targets like corner reflectors only reflect the received signal while active targets, so called transponders are able to modify the retransmitted signal (e.g. by amplification, adding a delay to it or changing the polarization).
The Institute utilizes corner reﬂectors for different purposes and of various sizes. In 2014, the large number of corner reflectors was extended by the first remote-controlled corner reﬂectors with a leg length of 2.8 m and a form tolerance of 1 mm. In between data takes the remote controlled corner reflector is turned upside down (as shown in Figure 6 and Video 1), in a parking position in order to protect the opening from precipitation. As part of the DLR calibration site these targets are located in Southern Germany and are available to support SAR missions worldwide.
Over the years the Calibration Group has developed and built highly accurate transponders in various frequency bands including X, C, and L-band for different SAR missions. The development of the so called “Kalibri” transponders, which have been operated successfully for ESA’s Sentinel-1 mission in the frame of the European Copernicus program since 2014, see Video 2, was focused on radiometric accuracy and autonomous, remote controlled outdoor operation for more than a decade.
For this purpose, the transponders are remotely aligned towards the satellite by a two-axis positioner. The transponder concept is based on a two-antenna design ensuring a small time delay between reception and re-transmission of the radar signal. Furthermore the complete electronic and both antennas are embedded in a temperature stabilized housing, as shown by the open device in Figure 7, in order to protect them from different weather condition. The superior radiometric stability of below 0.1 dB is maintained by an internal calibration loop. An FPGA-based subsystem allows for a digital recording of radar chirps.
The configuration of satellite overpasses and the corresponding data acquisitions are conducted in an automated fashion and can be conveniently scheduled and monitored by a web interface. The complete in-house development features a modular design, that allows to adapt the proven concept to different frequency bands: currently, front-ends for C- and X-band exist.
Based on the Kalibri design we developed two highly accurate C-band transponders for the RADARSAT Constellation Mission (RCM) of the Canadian Space Agency (CSA). Both transponders were deployed in Canada in 2017 and have been fully operational since then.
Through European and international projects we could well expand the radar calibration activities as well as strengthen our position as a DLR SAR Calibration Center. Beyond TerraSAR-X and TanDEM-X, the DLR SAR Calibration Center has been supporting ESA’s Sentinel-1A/-1B missions since 2014. All these missions enabled the achievement of a set of accuracies for the different calibration methods and techniques, which are summarized in Table 1, and which can be considered as benchmarks.
With our DLR SAR Calibration Center and the continuing effort on developing new calibration techniques, targets as well as analysis and evaluation tools, the Institute is well prepared for the challenges of the next generation of SAR missions.