Geometry Reconstruction Based on Attributes of Scattering Centers by Using Time-Frequency Representations

Experiments have shown that the inverse scattering problem of a radar target can potentially be resolved by the attributed scattering centers. The signatures of radar images are mainly determined by the aspect dependencies of the scattering amplitude and location of scattering centers. According to the aspect dependency characteristics, scattering centers are classified into three types. Each type provides different geometry information on the targets. This study presents the characteristics of these three types of scattering centers by using the parametric models and signatures of different radar images. The time-frequency representations (TFRs) of radar returns are found to have some advantages in geometry reconstruction compared with other radar images. Based on the parametric models of the scattering centers, a framework for recovering geometry by using TFR is proposed. Numerical experiments with an unmanned aerial vehicle (UAV) and a civil aircraft show the validity of this technique. The results indicate that the main geometries of the two aircrafts are reconstructed with high precision.