Although the deep learning (DL) technique has been introduced into computational physics in recent years, the feasibility of applying it to solve electromagnetic (EM) scattering field from arbitrary scatters remains open. In this paper, the convolutional neural network (CNN) has been employed to predict the EM field scattered by complex geometries under plane wave illumination. The 2D finite difference frequency domain (FDFD) algorithm, wrapped by a module to randomly generate complex scatters from basic geometries, is employed to produce training data for the network. The multi-channel end-to-end CNN U-net is modified and combined with a Resnet, which can prevent the vanishing gradient problem in the training process, to form the EM-net. The well-trained EM-net has good performance in this problem since it is compatible with various of complex shapes, multiple kinds of materials and different propagation directions of the incident waves. The effectiveness of the proposed EM-net has been validated by numerical experiments, and the average numerical error can be as small as 1.02%. Meanwhile, its speedup ratio over the FDFD method is as large as 2000.

Existing numerical electromagnetic (EM) solvers are usually computationally expensive, time consuming, and memory demanding. Recent advances in deep learning (DL) techniques have demonstrated superior efficiency and provide an alternative pathway for speeding up simulations by serving as effective computational tools. In this paper, we propose a DL framework for real-time predictions of the scattering from an isolated nano-structure in the near-field regime. We find that, to achieve precise approximation of the optical response obtained from numerical simulations, the proposed DL framework only requires a small training data set. The fully trained framework can be three orders of magnitude faster than a conventional EM solver based on the finite difference frequency domain method (FDFD). Furthermore, the proposed DL framework has demonstrated robustness to changes in design variables which govern the nano-structure geometry and material selection as well as properties of the incident wave, shedding light on universal scattering predictions at the nano scale via deep learning techniques. This framework increases the viability of the design and analysis of complex nanostructures, offering great potential for applications pertaining to complex light-matter interaction between electromagnetic fields and nanomaterials.

This paper presents a stabilized nodal discontinuous Galerkin pseudospectral time domain (DG-PSTD) algorithm for fully anisotropic electromagnetic waves. This solver permits arbitrary high-order basis functions and adaptive hexahedral elements, thus very efficient for large-scale wave propagation in complex media. Maxwell’s equations are reformulated in a unified hyperbolic form, where a localized anisotropic Riemann solver is derived to serve as a numerical flux to exchange information across adjacent elements in the DG-PSTD scheme. This local analysis method also helps impose the time-domain anisotropic plane wave incidence in the total/scattering field framework. Numerical validations and applications demonstrate the efficiency, accuracy, and capability of this new high-order solver for 3-D large-scale generally anisotropic electromagnetic media.

Hybrid plasmonic waveguides leveraging the coupling between dielectric modes and plasmon polaritons have emerged as a major focus of research attention during the past decade. A feasible way for constructing practical hybrid plasmonic structures is to integrate metallic configurations with silicon-on-insulator waveguiding platforms. Here we report a transformative high-performance silicon-based hybrid plasmonic waveguide that consists of a silicon nano-rib loaded with a metallic nanowire. A deep-subwavelength mode area (${\lambda }^2/4.5\times {10}^5-{\lambda }^2/7\times {10}^3$), in conjunction with a reasonable propagation distance (2.2–60.2 μm), is achievable at a telecommunication wavelength of 1.55 μm. Such a nano-rib-based waveguide outperforms its conventional hybrid and plasmonic waveguiding counterparts, demonstrating tighter optical confinement for similar propagation distances and a significantly enhanced figure of merit. The guiding properties of the fundamental mode are also quite robust against possible fabrication imperfections. Due to the strong confinement capability, our proposed hybrid configuration features ultralow waveguide cross talk and enables submicron bends with moderate attenuation as well. The outstanding optical performance renders such waveguides as promising building blocks for ultracompact passive and active silicon-based integrated photonic components.

A novel frequency reconfigurable antenna based on substrate integrated waveguide (SIW) for S-band and C-band applications is proposed. The antenna employs a SIW resonant cavity structure with a rectangular-ring slot and a long slit etched on the top surface. An extra via post is introduced to adjust the resonance frequency of the low-frequency mode independently without affecting the impedance matching of the high-frequency mode or decreasing the gain. Several PIN diodes are symmetrically placed along the rectangular-ring slot to attach the patch to the top surface. The working state of PIN diodes is controlled by a simplified bias structure. A prototype was fabricated and measurements show the proposed antenna has a bandwidth of 18 MHz in S-band low-frequency mode and 322 MHz in C-band high-frequency mode. Good agreements between the simulated and measured return losses, as well as radiation patterns, are observed.

Promising microwave ultra-broadband water-based metasurface absorbers have attracted increasing attention due to their novel applications in EM radiation prevention, stealth technology, and energy harvesting. In this paper, the proposed water-based metasurface absorber, composed of mushroom-shaped and layer-shaped water with resin shell, can reach over 90% polarization-insensitive and wide-angle absorption from 7.2 to 100 GHz with a relative bandwidth of 173%. Moreover, the proposed absorber is reconfigurable to control the water content, salinity, and type of inside liquid for tunable absorption and infrared radiation with thermally tunable absorption. Since the experimental and simulated ultra-broadband absorption originates from multiple resonances concentrated on the top, the ground-free all-dielectric design is achieved. Furthermore, the proposed curved absorber is fabricated and demonstrated for omnidirectional absorption. The water-based metasurface absorber shows great potential in achieving both microwave and infrared radiation camouflage for practical shaped structures in both civil and military domains.