Compressing bistatic SAR target signatures with sparse-limited persistence scattering models

Here, the authors study the modelling of bistatic synthetic aperture radar (SAR) target signatures. They focus on building concise models inspired by physical phenomenology that explain target reflectivity over a wide range of bistatic angles and target pose. A regularised inversion method is proposed that decomposes the target into a set of scattering centres with limited persistence by utilising the model-based sparsity of scattering coefficients. The resulting sparse model can be used for online estimation of target signatures at much faster speeds than simulation or database approaches. Based on the prior work on parametric models derived for canonical reflectors, the authors hypothesise that the scattering coefficient as a function of the viewing angle is embedded in a low-dimensional subspace spanned by a set of functions or dictionary atoms that have a support, which is concentrated in the viewing angle domain. Exploiting the differentiability of the observation model in the scattering centre locations and the parameters associated with the basic functions, the authors propose a method to estimate the scattering centre locations in the continuum avoiding gridding errors. Using numerical simulations, they verify that the proposed method obtains a succinct representation of these objects using a sufficiently small set of scattering centres.