Distributed SAR Interferometry for Digital Elevation Model Generation Using Microsatellites: Processing and Demonstration

Abstract

Digital elevation models (DEMs) are important remote sensing products obtained through synthetic aperture radar (SAR) interferometry. The DEM height accuracy improves as the baseline increases, but the interferograms become more susceptible to phase unwrapping errors. Resolving phase unwrapping errors requires either additional satellites or additional passes of the same satellites with different baselines, limiting the system’s capacity of monitoring fast dynamic phenomena. Single-pass generation of DEMs with high accuracy and free of unwrapping errors therefore still remains an open challenge. This work presents for the first time a concept of distributed SAR interferometry for single-pass DEM generation based on a cluster of small-aperture microsatellites, separated in both the along- and across-track directions and operating well below the Nyquist sampling frequency. The conventional SAR interferometric processing approach cannot be applied as the data from each small aperture in azimuth cannot produce an unambiguous SAR image. This work presents a novel optimal processing scheme based on the maximization of the generalized likelihood that jointly exploits the data from all antenna apertures to simultaneously resolve the phase and the azimuth ambiguities and produce a highly accurate DEM free of unwrapping errors. An airborne demonstration is furthermore presented using multi-baseline data from a dedicated flight campaign of the F-SAR system of the German Aerospace Center, which demonstrates feasibility of this concept. This work expands the understanding of joint processing of multiple SAR images and unlocks attractive options for future distributed Earth observation missions.