Analysis Of Electromagnetic Scattering Characteristics Based On CBFM-MLFMA And GPU Parallel Technology

The research on analysis of electromagnetic scattering characteristics from targets is of great significance for radar target recognition and design of radar system.In particular,how to solve the problems of electrically large targets accurately and efficiently has become one of important research directions in the field of computational electromagnetics.With the rapid development of computer technology,GPU high-performance parallel computing has begun to receive enthusiastic attention from all walks of life,and has been widely applied in medical imaging,weather forecasting,engineering calculation and other fields.Of cource,many scholars in the field of computational electromagnetics are unwilling to lag behind,have carried out the research on electromagnetic computing methods based on GPU parallel acceleration,and have achieved many excellent results.To solve the electromagnetic scattering problems of electrically large targets,this paper has carried out the research on analysis methods for electromagnetic scattering characteristics from targets based on CBFM-MLFMA and GPU parallel technology.Firstly,the paper introduces the basic principle and solution process of the Method of Moment(Mo M)in detail.However,it is difficult to efficiently solve the problems of electromagnetic scattering characteristics of electrically large targets.Secondly,the paper has researched the characteristic basis function method based on singular value decomposition(CBFM-SVD)and the multilevel fast multipole algorithm(MLFMA).And the MLFMA is used to accelerate the calculation of impedance matrix elements of the CBFM-SVD.The paper completes the programming of the CBFM-MLFMA method.Then,the correctness and efficiency of the CBFM-MLFMA method are verified by calculating the radar cross section(RCS)of some typical targets.Finally,the heterogeneous parallel computing of CPU and GPU based on CUDA architecture is studied.The corresponding heterogeneous computing model is designed according to the parallelism of the CBFM-MLFMA method.The GPU parallel technology is used to optimize and accelerate the CBFM-MLFMA method.The paper stability simulates the RCS of electrically large targets,and solves the problems of the electromagnetic scattering characteristics of electrically large targets with high accuracy and high efficiency.