Analysis Of RACA And Its Improved Algorithm Based On Electromagnetic Scattering Characteristics From Perfect Electric Conducting

In the art of radar target recognition,stealth warship,etc.,analysis of electromagnetic scattering characteristics of targets has big significance.With the development of computer technology and computational electromagnetic,the method of moments(MOM)has become an important approach for solving integral equations(IE)relevant to electromagnetic scattering of electrically large targets.In this dissertation,in order to reduce the calculation time and memory requirements,two efficient numerical algorithms based on RACA are deeply studied to calculate electromagnetic scattering characteristicsof perfect electric conducting targets.The main contents and contributions of this dissertation include the following aspects:1.Giving an intensive study on adaptive cross approximation(ACA)and recompression adaptive cross approximation(RACA)algorithm.ACA can effectively perform the compression and decomposition on the far field low rank matrix,but the decomposed matrix is not orthogonal.By introducing QR decomposition and singular value decomposition(SVD),RACA algorithm can further compress the matrix.Numerical examples show that the RACA algorithm has small memory requirement.2.The barycentric subdivision method is a fast algorithm to fill the impedance matrix,which can replace the double integral in the integral expression by the sum of 9 points to reduce the calculation time of each impedance matrix element.However,the barycentric subdivision method has the drawback of big memory requirement in the process of filling impedance matrix.This paper proposed the combination of the barycentric subdivision and RACA algorithm to reduce memory requirement.In the far field area this paper uses the RACA algorithm to extract the row and column elements adaptively,and use the concentric segmentation algorithm to accelerate the calculation speed.In the near field region,this paper uses barycentric subdivision algorithm to fill the matrix effectively,and combines the near field preconditioning technique to accelerate the process of solving matrix equation.Numerical examples demonstrate the correctness and effectiveness of the improved method.3.The equivalent dipole moment(EDM)method can further accelerate the speed of the filling of the impedance matrix without memory reduce.To solve this problem,this paper proposed a combination of RACA algorithm,barycentric subdivision method and EDM algorithm.The RACA plays the role of extracting the row and column elements in the far field region,while the EDM method is used to accelerate the filling process.In the near field,this paper uses barycentric subdivision algorithm to fill the matrix effectively,at the same time combines the near field preconditioning technique to accelerate the matrix calculation.Numerical examples show that the improved algorithm arises the computational efficiency of the MOM when analyzing electromagnetic problems..The proposed algorithm provides a new idea for the calculation of electromagnetic scattering characteristics of electrically large objects on general computer.