Ultrafast farfield simulation of non-paraxial computer generated holograms

Article Materials Science, Multidisciplinary

Deep Convolutional Neural Networks to Predict Mutual Coupling Effects in Metasurfaces

Sensong An et al.

Summary: The paper introduces a deep learning approach to predict the actual electromagnetic responses of metasurface elements with near-field coupling effects, leading to significant improvements in efficiency for beam deflector and metalens designs. This methodology can be readily applied to explore mutual coupling effects and enhance various metasurface designs' performance.

ADVANCED OPTICAL MATERIALS (2022)

Add to Collection

Article Optics

Macroscopic wave-optical simulation of dielectric metasurfaces

Sebastian Linss et al.

Summary: The novel method proposed in this study uses a specially adapted finite-difference beam propagation method to semi-rigorously simulate diffractive optical elements, allowing for linear scaling with the number of grid points within a reasonable runtime. Its applicability is demonstrated through simulations of a metalens and a polarization-dependent beamsplitter based on effective-medium metasurfaces, showing high conformity to rigorous simulations.

OPTICS EXPRESS (2021)

Add to Collection

Article Nanoscience & Nanotechnology

Deep Learning to Accelerate Scatterer-to-Field Mapping for Inverse Design of Dielectric Metasurfaces

Maksym Zhelyeznyakov et al.

Summary: The inverse design of optical metasurfaces is an emerging field with great potential in miniaturizing conventional optics and developing new optical functionalities. A data-driven forward simulation framework has been developed for inverse design, which is more accurate, faster, and more memory efficient than traditional methods.

ACS PHOTONICS (2021)

Add to Collection

Article Optics