Generation of Highly Accurate DEMs Over Flat Areas by Means of Dual-Frequency and Dual-Baseline Airborne SAR Interferometry

Abstract

In this paper, a dual-frequency and dual-baseline (DFDB) processing framework for the extraction of highprecision terrain information from airborne interferometric synthetic aperture radar (SAR) data is presented. Specifically, we propose the use of two single-pass data sets acquired simultaneously in two different frequency bands and two largebaseline repeat-pass data sets also acquired simultaneously in two frequency bands. The configuration profits from the stability of the single-pass derived elevation maps in relation to spatially correlated artifacts as well as from the increased sensitivity associated with large-baseline acquisitions. Moreover, the dualfrequency nature of the data set enables the tackling of the phase unwrapping issue, promoting the retrieval of unambiguous measurements. Several algorithms for the interferometric processing of the DFDB airborne data set are proposed, including the outline of multichannel phase calibration and unwrapping error correction strategies and approaches to remove spatially correlated artifacts and extract the common underlying topography. Elevation models generated from a DFDB data set acquired with the airborne F-SAR sensor over tidal flats in northern Germany are presented, and comparisons with an airborne laser scanner reference show errors with a standard deviation of around 14 cm and a mean absolute deviation of less than 10 cm. Index Terms-Digital elevation model (DEM), dual frequency, repeat-pass interferometry, SAR interferometry (InSAR). 0196-2892

Dual-Frequency Airborne SAR for Large Scale Mapping of Tidal Flats

“…where σ 0 represents the assumed backscatter coefficient of the scene. According to values reported in the literature [3,10], and from our own observation of data of a previous campaign over the Jade bight reported in [11], the backscatter over the tidal flats can be expected to range from −5 dB to −25 dB, depending on water content, incidence angle and frequency of acquisition. Accordingly, for the repeat-pass baseline selection, backscatter values ranging from −5 to −25 dB were considered.…”

“…Finally, note that even for case 4 and considering no additional error sources or baseline deviations, the use of repeat-pass S-band alone is not enough to fulfill the accuracy requirement. Nevertheless, given the negligible penetration expected over the tidal flats [10,11], the S-band height models can be used to generate a final merged XS DEM. In this case, the relative accuracy of the resulting DEM should be at least as good as the X-band one, provided that the merging process accounts for the individual DEM statistics.…”

“…The work presented in this paper is a follow-up of our initial developments on dual-frequency dual-baseline (DFDB) SAR interferometry presented in [11]. In contrast to the references cited above, our strategy for the generation of high resolution DEMs over tidal flats uses a combination of single-pass and repeat-pass airborne interferometry.…”

“…Moreover, our strategy considers not a single frequency of acquisition, but rather a combination of data acquired simultaneously in the X and S frequency bands. While a full description of the DFDB processing can be found in [11], this paper addresses modifications in data calibration, the mapping of wide areas by means of a mosaic consisting of several swaths and also the derivation of supplementary information for the indication of water covered areas or presence of vegetation, which might bias the derived elevation model. The work is conducted within the German governmental funded GeoWAM project, with partners from federal institutions, a small private company, and universities (see funding acknowledgement).…”

On the Frequency Sweep Rate Estimation in Airborne FMCW SAR Systems

“…In the last years, an increasing interest toward Synthetic Aperture Radar (SAR) systems [1] mounted onboard aerial platforms, such as airplanes [2][3][4][5][6], helicopters [7], unmanned aerial vehicles (UAVs) [8,9] and drones [10,11], is registered [12]. This is mainly due to the high operative flexibility and relatively low operative costs of these platforms.…”