A decoupled design strategy for a single-layer dual-band dual-circular-polarization (dual-CP) reflectarray (RA) is proposed. Two types of dual-band unit cells are developed by integrating either a K-band metallic resonator (MR) or a complementary slot structure (CSS) with a Ka-band metallic resonator, enabling reduced inter-band coupling and improved phase-tuning flexibility in the K band. This decoupling alleviates the need for collaborative optimization between the two frequency bands, thereby simplifying the RA design procedure. An 18-cm RA prototype is designed, fabricated, and measured. Experimental results demonstrate that two beam directions are achieved, each supporting a pair of transmit and receive (Tx/Rx) beams with gain differences within 0.2 dB, validating the feasibility of the proposed decoupling strategy.

This article presents an efficient half-space characteristic mode analysis (HSCMA) framework for the accurate analysis of large-scale finite periodic structures above a half-space. Finite periodic structures above a half-space are commonly analyzed using either the half-space periodic Green’s function method or the half-space integral equation method. The former models a finite array through the infinite array approximation and provides acceptable accuracy only for sufficiently large arrays with uniformly spaced elements but generally fails to capture edge effects accurately. The latter provides a rigorous full-wave solution, but is computationally prohibitive for electrically large structures because of the large number of unknowns and the associated dense impedance matrix. To overcome this limitation and improve modeling accuracy, characteristic modes (CMs) of an element in the half-space finite periodic structure are extracted. The extracted CMs are then employed as entire-domain basis functions to reduce the order of the discretized impedance matrix. To further accelerate matrix filling, matrix reuse and list interpolation techniques are applied. Moreover, the multilevel fast multipole algorithm (MLFMA) is integrated into the HSCMA framework to expedite matrix-vector multiplication during the solution of electromagnetic scattering problems involving electrically large finite periodic structures above a half-space. The improved accuracy and efficiency of the method are validated through numerical examples. Compared with commercial software, the proposed approach achieves a 94% reduction in computational time and a 69% decrease in memory usage.

This letter proposes an efficient optimization strategy for designing ultra-wideband, wide-angle connected conformal arrays. An irregular element arrangement is introduced, wherein elements are randomly positioned to enhance layout flexibility and achieve superior spatial distribution. However, the irregular arrangement of connected conformal arrays leads to element heterogeneity, complicating array optimization and beamforming. To address these challenges, the space mapping (SM) technique is employed to design irregular array configurations for ultra wideband connected conformal arrays with strong mutual coupling (MC). By integrating the covariance matrix adaptation evolution strategy (CMA-ES) with the active element pattern (AEP), the optimization accuracy of the coarse model in the SM design process is further improved. Additionally, the effects of curvature radius and phase variation on the irregular array configuration of conformal arrays are considered. The proposed method achieves sidelobe level reductions of 1.43–7.15 dB, while the gain loss remains below 1.43dB. it has demonstrating strong potential for ultra-wideband, wide-angle array applications.

In this article, a local time-stepping (LTS)-based arbitrary high-order finite-difference time-domain (LTS-ADER-FDTD) method is proposed for the efficient simulation of transient electromagnetic problems involving highly refined local structures. By adopting the arbitrary high-order (ADER) approach, the method converts temporal derivatives into spatial derivative operations, thereby achieving ADER accuracy in time discretization. Through nonuniform grid partitioning, the LTS strategy enables different regions to adaptively select time steps according to their spatial step sizes, significantly reducing the computational burden. The proposed approach ensures numerical stability and computational consistency during multitime-step advancement while maintaining high spatiotemporal synchronization. Both theoretical analysis and numerical experiments demonstrate that, compared to the conventional finite-difference time-domain (FDTD) method, the proposed algorithm considerably reduces the number of time iterations while preserving comparable computational accuracy, leading to a significant improvement in overall efficiency. This provides a robust and effective solution for the modeling and simulation of complex multiscale electromagnetic problems.

A quantum–classical hybrid framework is proposed for local induced-current reconstruction in characteristic mode analysis of perfectly electrically conducting (PEC) structures. Selected current-component values are reformulated as complex inner-product estimation problems between a modal sampling vector and a modal weighting vector. A modified Hadamard test combined with quantum phase estimation (QPE) is employed to evaluate the complex inner product. Ship and aircraft examples show that the local-query estimates are consistent with direct classical inner products. IBM Quantum execution further demonstrates proof-of-principle implementation of the inner-product estimation subroutine. The proposed framework provides a quantum-assisted local-query interface for characteristic-mode current post-processing and serves as a building block for future electromagnetic workflows with quantum-accessible modal data.

An efficient method, characteristic mode-multilevel fast multipole algorithm-clustering (CM-MLFMA-clustering) method, is proposed for analyzing electromagnetic scattering in multiobject systems. This approach integrates theory of characteristic modes (TCMs) with the fast multipole method (FMM) to accelerate the computation of total scattering contributions, including individual object scattering and mutual coupling effects. To accurately account for mutual coupling effects induced by scattering fields between objects, we introduce a novel coupling modal weighting coefficient (CMWC), extending the conventional modal weighting coefficient (MWC). The total scattering fields are obtained through a linear superposition of dominant modes weighted by both MWC and CMWC. Additionally, a dual-scheme FMM is proposed to enhance computational efficiency. The first scheme employs multilevel fast multipole algorithm (MLFMA) for pairwise electromagnetic coupling calculations, while the second scheme utilizes a K-means clustering algorithm to efficiently manage large multiobject systems. Numerical results confirm the method's accuracy, demonstrating a significant reduction in memory requirements and computational complexity, achieving approximately O(N) scaling observed in numerical simulations.

A novel strategy for accelerating electromagnetic scattering simulations of multiple objects is proposed, based on truncating coupling interactions by inter-object distance. The method utilizes pre-computed characteristic modes (CM) of isolated objects, which are reused to reconstruct induced currents via a newly defined coupling modal weight coefficient (CMWC). The multilevel fast multipole algorithm (MLFMA) is adopted to accelerate the solution of CMWC. To further improve computational efficiency, the coupling between far-separated objects is approximately neglected. Numerical examples demonstrate that the proposed method significantly improves simulation efficiency while maintaining calculation accuracy.

Classic theory of characteristic modes (TCM) has been further developed into various forms for efficient analysis and design of different electromagnetic structures. With the help of quantum computing concept, this paper tries to propose a quantum unified theory of characteristic modes to provide a unified framework. It is expected that the proposed quantum unified TCM framework can help better understand the TCM in terms of theoretical aspects, applications, and further potentials for effectively solving electrically large and complex electromagnetic problems.

The high cost of generating training datasets is a thorny problem for researchers exploring the applications of artificial intelligence (AI), especially for intelligent electromagnetic (EM) computations, because EM modelling datasets are expensive to produce. The method of active learning (AL) can reduce the cost by selectively labelling data. In this paper, an AL approach is developed to cut down the size of dataset for AI-based prediction of EM scattering problems. By using the electric field boundary conditions, more pertinent targets are screened to constitute a refined dataset. Numerical experiment results show that compared with full supervised learning method, the proposed AL scheme can achieve similar or higher prediction accuracy using a much smaller dataset.

An ultra-wideband antenna array employing a substrate integrated coaxial line (SICL) feeding network and a stacked metasurface configuration is presented for millimete-rwave (mm-wave) applications. The antenna element achieves an operating frequency range of 16.45–50.38 GHz, corresponding to a fractional bandwidth of 101.5%. The bandwidth enhancement mechanism is analyzed using characteristic mode analysis (CMA) in conjunction with an equivalent circuit model. Electric field and surface current distributions indicate that the stacked metasurface mitigates radiation pattern degradation at high frequencies associated with the enlarged electrical aperture of a single-layer metasurface. A 1 × 8 antenna array is then developed by integrating a transition structure with the SICL feeding network. The array exhibits an operating bandwidth of 15.84–49.15 GHz (102.5% fractional bandwidth) and a peak gain of 18.58 dBi. The proposed design demonstrates superior bandwidth performance compared with existing wideband metasurface-based antenna arrays.

This letter presents the design and experimental validation of an active frequency-selective rasorber (FSR) for dynamic wavefront manipulation and broadband scattering suppression. The proposed structure integrates a resistive ring loaded with spiral inductors, a polarization-conversion layer, and a reconfigurable gradient trapezoidal frequency-selective surface (FSS) embedded with PIN diodes. By controlling the diode states, the FSR exhibits distinct electromagnetic responses over three frequency bands. In the OFF state, the design achieves a wide absorption bandwidth of 112% from 0.94 to 3.35 GHz, a transmission window of 40.5% from 3.35 to 5.05 GHz, and a polarization-conversion band from 5.3 to 6.5 GHz. In the ON state, the transmission window is switched to a reflection dominant band, while absorption is maintained within 1.05–2.9 GHz and 6.65–8 GHz. Benefiting from the combined absorption and polarization-conversion mechanisms, an RCS reduction of more than 10 dB is achieved over a broad frequency range of 1.5 8 GHz. A 10 × 10 prototype is fabricated and measured, showing good agreement with simulations and validating the effectiveness of the proposed design for electromagnetic shielding and low scattering aperture applications.

Classic theory of characteristic modes (TCM) has been extended to effectively handle the analysis of large finite arrays through the implementation of periodic characteristic mode analysis (PCMA) and quasi-PCMA (QPCMA) methods. Inspired by the concept of quantum computing, this paper tries to propose a quantum unified TCM framework and set up a general way to understand the PCMA and QPCAM methods in a unified manner. It is expected that the proposed quantum unified framework can provide a clear guide to develop novel TCM-based algorithms for more general array structures.

In the last decades, integral equations (IEs) have been commonly used for the analysis of time-harmonic electromagnetic scattering and radiation by perfect electrically conducting (PEC) objects. The classical discretization scheme of IEs is the method of moments (MoM), which uses basis functions to represent the unknown current and applies Galerkin testing with testing functions. After the MoM discretization, matrix equation system ZI = V is obtained, where Z denotes the MoM matrix, I contains the unknown current expansion coefficients, and V represents the excitation source contribution.

An efficient modeling method for analyzing the electromagnetic scattering from large-scale quasi-periodic arrays with varying element sizes is proposed through a smart combination of characteristic mode basis functions (CMBFs), empirical interpolation method (EIM), and adaptively enriching greedy algorithm (AEGA). This method takes a small set of CMBFs as the entire-domain basis functions and extends it to the whole array in the process of the method of moments (MoMs). To speed up the assembly time of the reduced MoM impedance matrix of the whole array resulting from CMBFs, the EIM is used for Green's function interpolation to obtain a set of interpolation basis functions independent of geometric parameters. With this set of interpolation basis functions, the reduced MoM impedance matrix can be quickly assembled under any geometric parameters. The AEGA is employed to address the inefficiency of the EIM in high-dimensional parameter spaces induced by mutual coupling. As a result, the dimension of the final MoM matrix equation of modeling the whole array is significantly reduced, and its solution process is greatly accelerated with shorter runtime. Finally, numerical results validate the efficiency and accuracy of the proposed method.

In this work, we investigate the Harrow-Hassidim-Lloyd (HHL) quantum algorithm for the solution of electric field integral equation (EFIE)-based matrix equation system for analyzing electromagnetic scattering from three-dimensional (3D) perfect electrically conducting (PEC) objects. The HHL algorithm is one of the representative quantum algorithms for the solution of linear equation systems. Numerical experiments are presented to investigate the computational cost and accuracy of the HHL algorithm for solving the EFIE matrix equation system executed on the virtual simulator on the classical computers.

Solution to mixed potential integral equation (MPIE) for layered media problems is very important for microstrip structures. Starting from the MPIE for layered media problems, through following the concept of theory of characteristic modes (TCM) for conducting objects, TCM for modelling layered media structures is set up first. The transmission and reflection coefficients of periodic structures with layered media substrate environment are then connected to the characteristic mode parameters of periodic structures. These characteristic mode parameters can be easily obtained using substructure TCM of small finite arrays. With these approximate characteristic mode parameters, we can conveniently analysis and design periodic structures in layered media environment. Combining the steps mentioned above yields a unified design concept for electromagnetic periodic structures. This newly developed design concept is expected to have more applications to the design of novel electromagnetic structures.

Microwave devices are susceptible to high-power microwave breakdown effect, which seriously impacts the reliability of their integrated microwave systems, including electromagnetic communication stations and other electronic systems. To evaluate the high-power microwave breakdown characteristics of microwave devices, this article proposes a modified method for predicting the high-power microwave breakdown threshold. The proposed method considers the influence of more operating factors such as temperature, gas, and pulse shape by adjusting physical quantities like diffusion coefficient, adhesion rate, and pressure. Additionally, the newly modified equation can be efficiently solved using the discontinuous Galerkin approach, which has been already proved significant advantages in handling complex multiscale structures. Numerical examples demonstrate the remarkable accuracy and efficiency of the proposed method by comparing the calculated threshold with results from commercial software. We believe that this method provides an effective tool for predicting the breakdown threshold of microwave devices under various conditions, theoretically guiding the design of protection mechanisms for high-performance microwave components.

A physics-informed deep learning-based scheme is introduced for computing partial inductances of interconnects. This scheme takes a physics-based skin depth map and a geometry identifier of the interconnects as inputs and provides the current density distribution on the interconnects as the output. The predicted currents are then used to compute the partial self-resistances, self-inductances, and mutual-inductances of the interconnects. The proposed method leverages an Attention U-net, a U-shaped convolutional neural network with attention modules. During the training of Attention U-net, a specifically designed loss function is used to ensure the accurate modeling of the currents on the structure as well as ports. The accuracy, efficiency, and generalization ability of this physics-informed deep learning method are demonstrated via inductance extraction of the interconnects with and without a ground plane, including straight single interconnects, interconnects with sharp bends, parallel interconnects, and multiple conductor crossover buses. Numerical results show that the proposed scheme can predict the current density distribution of one interconnect scenario in 15.63 ms on GPU, 1157x faster than the physics-based solver, while providing self-inductances, mutual-inductances, and self-resistances of interconnects with around 1%, 3%, and 4% l2 -norm error, respectively.

This paper proposes a method, utilizing the characteristic modes (CMs) of the scatterers, to achieve rapid and accurate prediction of radar cross section (RCS) sequence for multiple scatterers. The linear reconstruction of the scatterers' RCS sequence by using CMs requires information about the scatterers' flight position and attitude angles. Therefore, predicting the RCS sequence over a specific period of time can be transformed into forecasting the scatterers' position and attitude parameters during the period. By leveraging neural networks, the position and attitude angle variations of the multiple scatterers can be predicted rapidly and accurately, enabling rapid and accurate prediction of the RCS sequence throughout the scatterers' flight. Numerical example confirms the accuracy and efficiency of the proposed method.

An efficient solution method that combines the theory of characteristic modes (TCM) with the fast multipole method (FMM) for angular glint characteristics analysis of multiple objects is proposed. The proposed method enables rapid reconstruction of objects' scattering fields under external excitation through TCM, while employing FMM to accelerate the computation of inter-object electromagnetic coupling effects. The angular glint characteristics of multiple objects are obtained via the poynting vector method. Compared with conventional methods, the proposed method significantly reduces computational overhead while maintaining accuracy, providing a novel solution for rapid assessment of radar tracking errors.