Comparison of parametric sparse recovery methods for ISAR image formation

A parametric sparse representation model of the inverse synthetic aperture radar (ISAR) signal has been proposed recently, and the ISAR signal is decomposed as a summation of many basis-signals determined by the target rotation rate. Based on the parametric sparse representation model, several sparsity-driven algorithms are proposed to retrieve both the target rotation rate and the ISAR image. In this paper, four parametric sparse recovery algorithms are compared mainly in three aspects: the accuracy of the rotation rate estimation, the ISAR image quality and the computational load. Numerical examples are presented to show the advantages and disadvantages for each method.