Efficient RCS Extrapolation Method of Target Scaling Based on Physical Optics

A surrogate modeling technique based on physical optics (PO) amplification model is proposed for extrapolation of radar cross sections (RCS) for targets of different sizes. Inspired by physical optics, the amplification factor is introduced to establish the relationship between model size and scattered field. Dividing the RCS of the target into general trend and local fluctuations. The overall trend is predicted by the nonlinear least square (NLS) method and local fluctuations are automatically extrapolated by Gaussian process regression (GPR). To demonstrate the accuracy of this approach, several examples are applied to verify the performance of the RCS extrapolation method.