Multilevel Fast Multipole Algorithm for Electromagnetic Scattering by Large Metasurfaces using Static Mode Representation

Emanuele Corsaro,Giovanni Miano,Antonello Tamburrino,Salvatore Ventre,Carlo Forestiere

2024-10-05

Abstract:Metasurfaces, consisting of large arrays of interacting subwavelength scatterers, pose significant challenges for general-purpose computational methods due to their large electric dimensions and multiscale nature. This paper introduces an efficient boundary element method specifically tailored for metasurfaces, leveraging the Poggio-Miller-Chang-Harrington-Wu-Tsai (PMCHWT) formulation. Our method combines the Multilevel Fast Multipole Algorithm (MLFMA) with a representation of the unknown equivalent surface current density by means of static modes, a set of entire domain basis functions dependent only on object shape but independent of the material and frequency. The compression of the number of unknowns enabled by the Static Mode Representation (SMR), combined with the \(\mathcal{O}(N \log N)\) complexity of MLFMA matrix-vector products, significantly reduces CPU time and memory requirements compared to classical MLFMA with RWG basis functions. We demonstrate the accuracy, time, and memory requirements of this method through several test cases including the full-wave simulation of a $100 \lambda \times 100 \lambda$ canonical metalens. The MLFMA-SMR method offers substantial benefits for the analysis and optimization of metasurfaces and metalenses.

Computational Physics,Optics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: in electromagnetic scattering simulation, for large - scale metasurfaces (arrays composed of a large number of sub - wavelength scatterers), the existing general - purpose computational methods face challenges brought by the large electrical size and multi - scale nature. These challenges lead to a significant increase in computing time and memory requirements, making the simulation very time - consuming and resource - intensive. Therefore, the paper proposes an efficient boundary element method, which is specifically optimized for metasurfaces to reduce computational complexity and memory consumption while maintaining high accuracy.

Specifically, the paper combines the Multilevel Fast Multipole Algorithm (MLFMA) with Static Mode Representation (SMR). By using the entire - domain basis functions (such as static modes) to represent the unknown equivalent surface current density, the number of unknowns is compressed. This method can not only significantly reduce CPU time and memory requirements but also handle large - scale metasurface structures in full - wave simulations, such as a standard metal lens of 100λ × 100λ.

The main contributions of the paper are:
1. **Proposing a new method**: Combining MLFMA and SMR and using the static mode as the entire - domain basis function effectively reduces the number of unknowns.
2. **Improving computational efficiency**: Through the O(N log N) complexity matrix - vector multiplication of MLFMA, the computing time and memory requirements are significantly reduced.
3. **Verifying the effectiveness of the method**: Through multiple test cases, including the full - wave simulation of a 100λ × 100λ standard metal lens, the accuracy and efficiency of this method are demonstrated.

Overall, this paper aims to provide an efficient and accurate computational method to address the computational challenges in electromagnetic scattering simulation of large - scale metasurfaces.