RCS prediction of polygonal metal plate based on machine learning

In this paper, a method based on machine learning and physical optics (PO) is proposed to accurately predict the monostatic radar cross section (RCS) of a polygonal metal plate. We use the PO method to calculate the RCS of the target, and then use the neural network to correct the result. By comparing the RCS calculated by the method of moments (MOM), our algorithm can accurately obtain the RCS of the target at different frequencies and different angles.