Iterative Music for Single-Look Sar Tomography

This paper proposes an iterative adaptation of the Multiple Signal Classification (MUSIC) algorithm for single-look tomographic Synthetic Aperture Radar reconstruction. Our primary objective is to adapt MUSIC for single-look scenarios to deal with the spatial resolution degradation associated with multi-look processing. Compared to the spatial smoothing approach proposed in the literature, a key advantage of our method is its independence from requiring a uniform baseline. Consequently, there is no need for sector interpolation to make the baseline uniform. The proposed method begins with an initialization using the pseudo-spectrum derived from matched spatial filtering. The latter is enhanced through a series of iterative steps, resulting in substantial resolution improvements. Numerical simulations confirm the superior performances of our method compared to state-of-the-art iterative approaches.