Title :
MLFMA optimal decomposing strategy
Author_Institution :
Dept. of Inf. Technol. & Mechatron. Eng., Shang Hai XingJian Coll., Shanghai, China
Abstract :
To decompose the far-neighborhood and near-neighborhood data domains of Multi-Level Fast Multi-Pole Algorithm (MLFMA), an optimal parallel implementation is discussed in this paper. To solve this problem, the characteristics of the two fields are respectively considered. Especially in the near-neighborhood decomposing, the clustering property of data is applied. The effectiveness of the optimized strategy is demonstrated by experimental results. Comparing with the former strategy, a higher parallel efficiency is obtained.
Keywords :
electrical engineering computing; electromagnetic wave scattering; parallel processing; pattern clustering; MLFMA optimal decomposing strategy; data clustering property; electromagnetic scattering; far-neighborhood data decomposition; multilevel fast multipole algorithm; near-neighborhood data decomposition; optimal parallel implementation; Computers; Electromagnetic scattering; Geometry; Impedance; MLFMA; Mathematical model; Matrix decomposition; MLFMA; clustering property; far-neighborhood; near-neighborhood; optimal; parallel implementation;
Conference_Titel :
Mechatronic Sciences, Electric Engineering and Computer (MEC), Proceedings 2013 International Conference on
Conference_Location :
Shengyang
Print_ISBN :
978-1-4799-2564-3
DOI :
10.1109/MEC.2013.6885461
