TanDEM-X Performance Optimization

Kurzfassung

Abstract TanDEM-X (TerraSAR-X add-on for Digital Elevation Measurement) is a space borne earth observation mission with the goal of generating a global Digital Elevation Model (DEM) with an unprecedented accuracy corresponding to the HRTI-3 specifications. TanDEM-X will fly in a special orbit formation (HELIX formation) with the fully compatible TerraSAR-X (TSX-1) satellite in order to collect interferometric SAR data. It is planned to cover the earth surface twice with different baselines. Presented in this paper is an optimization strategy, with the goal of a quasi-homogeneous height accuracy, which meets the HRTI-3 standard. The optimization has a major impact on the data acquisition plan. The acquisition of a DEM on a global scale requires an optimized mission plan, since the time, where both satellites are in orbit simultaneously, is limited to three years. Within this time the landmass is imaged in strip map mode twice, which allows for multi baseline phase unwrapping techniques. Using 167 satellite orbits, the region to be imaged has an across-track extension of about 240km at the equator. This region is subdivided into several swaths, so called data takes. The swath-width is determined by the timing, but also by the pattern of the SAR antenna. The shape of the antenna beam and SAR system parameters, like the pulse repetition frequency (PRF) or the transmitted bandwidth, are in the focus of the optimization. The quantity, which is a measure for the quality in any DEM, is the height accuracy. Therefore the optimization concentrates directly on this parameter, rather than trying to maximize a single parameter, e.g. the signal to noise ratio (SNR). In the first part of the paper, a height performance model is presented, taking into account several error sources, like thermal decorrelation, quantization errors, range- and azimuth-ambiguities, misregistration as well as volume decorrelation. The analysis is extended to different terrain-types and terrain-slopes, showing a significant impact on the height accuracy. The second part is dedicated to the optimization strategy. It has to be mentioned, that this optimization approach is a heuristic method, rather than a mathematical procedure with a cost function to be minimized. Since the parameters under consideration are in complex connection, the main challenge is to reduce the number of variables and still yield an accurate optimization result. Moreover a method is presented, how to fuse several independent interferometric data acquisitions to a single DEM, further increasing the height accuracy.