Skeletonized Nested Iterative Solver for Characteristic Modes of Multiscale Metallic-Dielectric Objects

IF 4.6 2区计算机科学 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

IEEE Antennas and Wireless Propagation Letters Pub Date : 2026-10-01 Epub Date: 2026-08-18 DOI:10.1109/LAWP.2026.3724999

Di Wu;Chenghong Gao;Qi Wu

引用次数: 0

Abstract

Skeletonization is integrated into the nested iterative framework to accelerate substructure-based characteristic mode analysis of multiscale metallic-dielectric composite structures. In the proposed method, skeletonization is combined with the multilevel fast multipole algorithm to accelerate the matrix–vector multiplications in the implicitly restarted Arnoldi method (IRAM), enabling efficient modal extraction for multiscale structures. Meanwhile, skeletonization is further employed to speed up the construction of the dual sparse approximate inverse (SAI) preconditioners. Numerical results demonstrate that the proposed method maintains the accuracy of characteristic modes while reducing the matrix-filling time by 41.7%, the SAI-construction time by 53.3%, and the memory consumption to about 30% of that required by Nested-IRAM.

查看原文本刊更多论文

多尺度金属介质物体特征模态的骨架嵌套迭代求解器

将骨架化技术集成到嵌套迭代框架中，加快了基于子结构的多尺度金属-介电复合材料结构特征模态分析。该方法将骨架化与多层快速多极算法相结合，提高了隐式重新启动Arnoldi方法（IRAM）中矩阵向量乘法的速度，实现了多尺度结构的高效模态提取。同时，进一步采用骨架化来加快对偶稀疏近似逆（SAI）预条件的构造。数值结果表明，该方法在保持特征模态精度的同时，将矩阵填充时间缩短41.7%，将sae构建时间缩短53.3%，将内存消耗降低到Nested-IRAM的30%左右。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

IEEE Antennas and Wireless Propagation Letters 工程技术-电信学

CiteScore

8.00

自引率

9.50%

发文量

529

审稿时长

1.0 months

期刊介绍： IEEE Antennas and Wireless Propagation Letters (AWP Letters) is devoted to the rapid electronic publication of short manuscripts in the technical areas of Antennas and Wireless Propagation. These are areas of competence for the IEEE Antennas and Propagation Society (AP-S). AWPL aims to be one of the "fastest" journals among IEEE publications. This means that for papers that are eventually accepted, it is intended that an author may expect his or her paper to appear in IEEE Xplore, on average, around two months after submission.