A highly efficient parallel approach of multi-level fast multipole algorithm

Based on the different characteristics of memory requirement and CPU time at different levels in the multi-level fast multipole algorithm (MLFMA), a highly efficient parallel approach is proposed, which employs different techniques to parallelize the plane waves and translation matrixes at different levels. The formulae for efficiently implementing this proposed approach are presented by theoretical analysis and numerical experiments. The radar cross-section (RCS) of a conducting sphere with a diameter of 144 (wavelength) simulated by over 10 millions unknowns, is successfully computed in the center for electromagnetic simulation (CEMS) in Beijing Institute of Technology (BIT). The comparison of numerical performance between center for computation electromagnetics (CCEM) in University of Illinois at Urbana-Champaign (UIUC) and our CEMS is presented, demonstrating the strong computation power of our proposed approach. The RCS of a plane model are also calculated to show our capacities to solve large complex electromagnetic problems.