RCS Modelling of Extended Targets Using Supervised Learning

Knowing the Radar Cross Section (RCS) of specific targets is of primary importance in target detection and recognition. The RCS may significantly vary with radar operating frequency and target's size, shape, material, and orientation with respect to radar illumination direction. Thus, a time-efficient way to model target RCS is of paramount importance in simulating close-to-reality scenarios where the position and orientation of target frequently vary. In this paper, an efficient estimation technique using machine learning algorithm is presented that can predict the RCS of targets of any shape and size. The proposed method is compared with the RCS obtained from the MATLAB tool POfacets. Computational time and mean square error (MSE) of estimated RCS with respect to the actual one are used as performance metrics.