基于双层ACA算法快速求解单站RCS问题

利用积分方程法计算三维目标单站RCS时,需要逐个角度地进行矩阵方程的求解。为了提高计算效率,本文采用自适应交叉近似算法(ACA)对多角度照射时生成的激励矩阵进行低秩压缩,减少了矩阵方程的求解次数;进一步基于单站角度上的分组方式提出了双层ACA算法,该算法对内存占用极小,提高了算法的并行性,而且更有效地实现了激励矩阵的降秩;最后结合多层快速多极子算法(MLFMA)实现电大尺寸目标的快速求解。数值计算结果表明,该算法能大幅减少大宽角条件下的单站RCS计算时间,具有较高的计算精度和计算效率。

Fast Computation of Monostatic RCS Based on Two-Level Adaptive Cross Approximation Algorithm

For backscattering problems of three-dimensional objects in integral equation method, the matrix equations corresponding to all the excitation vectors need to be solved. An adaptive cross approximation(ACA) algorithm is applied to accelerate the computation of monostatic radar cross section(RCS). The low-rank compression of the excitation matrix can be done via ACA algorithm and the method can greatly cut down the number of right-hand sides to be solved. A two-level ACA algorithm based on grouping of the incident directions is proposed to improve the computational and parallel efficiency. To facilitate the analysis of electrically-large problems, the multilevel fast multipole algorithm(MLFMA) is applied to accelerate the matrix-vector product. Numerical examples demonstrate that the proposed technique is highly efficient and accurate for monostatic scattering problems over a wide angular range.