Abstract
In this work, we focus on sparse representation of two-dimensional (2-D) radar signatures for man-made targets. Based on the damped exponential (DE) model, a 2-D augmented state-space approach (ASSA) is proposed to estimate the parameters of scattering centers on complex man-made targets, i.e., the complex amplitudes and the poles in down-range and aspect dimensions. An augmented state-space approach is developed for pole estimation of down-range dimension. Multiple-range search strategy, which applies one-dimensional (1-D) state-space approach (SSA) to the 1-D data for each down-range cell, is used to alleviate the pole-pairing problem occurring in previous algorithms. Effectiveness of the proposed approach is verified by the numerical and measured inverse synthetic aperture radar (ISAR) data.