Efficient Analysis of Electromagnetic Scattering from Targets based on UDS and SVR

Electromagnetic scattering characteristics of complex targets have always been the research focus in radar recognition. In this paper, we propose an efficient method that combines sampling design with machine learning to predict the radar cross section (RCS) of complex targets. In particular, uniform design sampling (UDS) is adopted to acquire the target's backscattering data of high representativeness, support vector regression (SVR) is trained as an efficient and accurate surrogate model for regression. Afterward, several experiments were carried out based on a SLICY model, and the results fully verify the effectiveness of the proposed method.