Super-resolution for SAR based on the MUSIC estimator

Abstract

In this thesis the performance of the MUSIC (MUltiple SIgnal Classification) algorithm for increasing the focusing of Synthetic Aperture Radar images was evaluated. This is achieved by exploiting the high separation capability of MUSIC. After a detailed discussion of the basic MUSIC approach, initially proposed for the estimation of the Direction–of-Arrival (DOA) of signals using an antenna receiver array, a new 2D post-processing algorithm (applied on already focused images) based on MUSIC is proposed and tested on simulated and real data (acquired by DLR’s E-SAR system). The aim is to separate targets closer than the Fourier resolution limit. The separation performance is quantized in terms of fractions of the resolution cell. Limitations and drawbacks are also discussed. Finally, an outlook of the application of the MUSIC approach in a wider SAR context is given.