We propose a multi-level low complexity fast direct solution of electric field integral equations (EFIEs). Far coupling submatrices are compressed by a kernel-independent method with nested equivalence source approximation (NESA). In this work, equivalence sources are constructed at upper levels while skeletons are selected via the adaptive cross approximation (ACA) at the bottom level to further accelerate the computation. To factorize and inverse the system matrix, an elimination matrix is introduced to each group to reduce the far blocks dimension. Then, an LU factorization is performed to reduce the near blocks dimension. This process is repeated recursively from the bottom level to the top. The inverse of the system matrix is then represented as a multiplication of a series of matrices arising from the factorization process. Numerical results verify the accuracy and the linear complexity of our proposed method.

Low Complexity Fast Direct Solution for Multiscale Problems with Nested Equivalence Source Approximation / Zuo, Yuhan; Li, Mengmeng; Vipiana, Francesca; Ding, Dazhi. - (2025), pp. 1-3. ( 19th European Conference on Antennas and Propagation, EuCAP 2025 Stockholm (Swe) 30 March 2025 - 04 April 2025) [10.23919/eucap63536.2025.10999958].

Low Complexity Fast Direct Solution for Multiscale Problems with Nested Equivalence Source Approximation

Zuo, Yuhan;Vipiana, Francesca;
2025

Abstract

We propose a multi-level low complexity fast direct solution of electric field integral equations (EFIEs). Far coupling submatrices are compressed by a kernel-independent method with nested equivalence source approximation (NESA). In this work, equivalence sources are constructed at upper levels while skeletons are selected via the adaptive cross approximation (ACA) at the bottom level to further accelerate the computation. To factorize and inverse the system matrix, an elimination matrix is introduced to each group to reduce the far blocks dimension. Then, an LU factorization is performed to reduce the near blocks dimension. This process is repeated recursively from the bottom level to the top. The inverse of the system matrix is then represented as a multiplication of a series of matrices arising from the factorization process. Numerical results verify the accuracy and the linear complexity of our proposed method.
2025
978-88-31299-10-7
File in questo prodotto:
File Dimensione Formato  
Low_Complexity_Fast_Direct_Solution_for_Multiscale_Problems_with_Nested_Equivalence_Source_Approximation.pdf

accesso riservato

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 496.05 kB
Formato Adobe PDF
496.05 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
EuCAP2025zyh3page_FV_revPDFexpress.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: Pubblico - Tutti i diritti riservati
Dimensione 498.54 kB
Formato Adobe PDF
498.54 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/3011176