SAR image simulation for object's feature analysis

Abstract

Synthetic Aperture Radar (SAR) image simulation is one of the potential tools to investigate the features of man-made objects in high resolution SAR image. In this paper, we present a SAR imaging simulator based on ray tracing for high frequency electromagnetic scattering prediction of complex objects. Input of the simulator is three dimensional (3D) model of an object, such as a building, vessel etc. The physical optics (PO) and geometrical optics (GO) are adopted to calculate the scattering of the object. To improve the computation efficiency, binary space partitioning (BSP) tree is constructed to describe the object model. After the radar cross section (RCS) is obtained, the image can be simulated by radar imaging algorithm. The frequency and angle method and the image domain integration method are performed to obtain the simulation SAR image. Considering the computation efficiency, image domain integration method is recommended for simulated image generation. The simulations of a sphere and a trihedral corner reflector are performed to validate the results of the simulator. In addition, a flat roof building is simulated. The change of double bounce effect of the building with different aspect angle in the simulation SAR images is investigated. The results show good agreement between the results of our modeling method and theory or measured data.