Applied Computational Electromagnetics Society Journal (ACES)

Multi-Objective Optimization of a Rectangular Structured Linear, Actuator Based on Taguchi Method

Muhammad Jawad — 2026-01-30

This paper introduces an innovative rectangular structured Linear Oscillating Machine (LOM) featuring an innovative arrangement of permanent magnets (PMs) which are positioned on the mover. A sinusoidal magnet arrangement (SMA) is chosen to minimize PM materials while ensuring the performance parameter, in terms of electromagnetic force, remains unaffected. The focus of the design is to reduce the cogging force and enhance the electromagnetic force to the PM mass ratio. The optimization is accomplished by utilizing design-of-experiment (DOE) Taguchi method optimization for a linear actuator, along with the finite element method (FEM). A multi-objective optimization approach is employed to refine the parameters of the linear actuator, focusing on the maximum force, the force-to-PM-weight ratio, and minimizing cogging force. By identifying the most effective parameters, an appropriate case with high sensitivity and accuracy is selected. The output parameters, like electromagnetic force and stroke of the proposed optimization approach, score very well, while the design is simply structured due to the utilization of a rectangular structured core part. The proposed design significantly reduces the cogging force to an exceptionally low level, enhancing performance and reliability.

Synthesis of Thinned Linear and Planar Antenna Arrays, Using a Taguchi-Enhanced Binary Gold Rush Optimizer

Weibin Kong — 2026-01-30

In this paper, a Taguchi-enhanced binary gold rush optimizer (TEBGRO) is proposed for designing thinned antenna arrays with a low peak sidelobe level (PSLL). The method integrates the Taguchi orthogonal experimental design into the population initialization phase, generating high-quality initial populations to improve convergence speed and stability. By combining a differential mutation interference factor and a time-varying transfer function, the algorithm further balances global exploration and local exploitation capabilities. Experimental results show that TEBGRO outperforms other binary optimization algorithms for both 100-element linear arrays and 20 x 10 planar arrays.

Enhanced CPML Based on the Autoformer Network for 2D, WCS-FDTD Method

Yumeng Wu — 2026-01-30

This paper proposes a novel convolutional perfectly matched layer (CPML) for the weakly conditionally stable finite-difference time-domain (WCSFDTD) method. The Autoformer neural network is introduced to replace the conventional multi-layer CPML. Employing only a single-layer structure, the Auto-former-driven CPML considerably reduces both the computational domain scale and algorithmic complexity. By leveraging sequence decomposition and sparse attention mechanisms, the wave-absorption performance of this method is significantly improved. Integrated into the 2D WCS-FDTD framework, the proposed method overcomes Courant-Friedrichs-Lewy (CFL) stability constraints for FDTD intelligent absorbing boundaries, with its time step size independent of fine grid sizes in any direction. Numerical results demonstrate that the proposed method can achieve excellent wave-absorption performance with high computational efficiency, while maintaining satisfactory robustness in complex scenarios.

A Hybrid Optimization Strategy of Random Forest and Differential, Evolution Algorithm for Wideband Antennas

Gengtao Huang — 2026-01-30

This paper presents a hybrid optimization strategy for wideband antenna design that leverages the strengths of both Random Forest (RF) and Differential Evolution (DE) algorithms. The strategy employs DE for iteratively updating antenna parameters and RF for feature selection in the process of antenna performance optimization. Initially, DE is applied to update antenna parameters for a predetermined number of iterations, generating a dataset of antenna performance metrics. This dataset is then used to train an RF model, which identifies the importance of each design variable. Feature selection, guided by the RF-derived importance, is applied to reduce the dimensionality of the search space. DE subsequently continues the optimization process within this reduced parameter space. Validation of this hybrid approach is performed through the design of a wideband slot antenna and compared against standalone DE, Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Simulated Annealing (SA). Results demonstrate that the proposed strategy significantly accelerates convergence, achieving the target reflection coefficient and gain with substantially fewer iterations than the other methods (reductions of 75.56%, 45%, 42.11%, and 50% compared to DE, GA, SA, and PSO, respectively). Furthermore, the hybrid strategy consistently finds superior solutions exhibiting lower loss values compared to the benchmark algorithms. The method offers a computationally efficient, interpretable, and effective approach to antenna optimization.

A Semi-Supervised Electromagnetic Imaging Algorithm Based, on Generative Adversarial Networks

Chun Xia Yang — 2026-01-30

In electromagnetic imaging applications, acquiring labeled data for supervised learning poses a significant challenge due to the high cost and time-consuming annotation processes. To address this limitation, we propose a semi-supervised electromagnetic imaging algorithm leveraging generative adversarial networks (GANs), which effectively integrates limited labeled data with abundant unlabeled measurements. Unlike conventional approaches that directly learn from raw scattered data, our method employs diffraction tomography (DT)-generated images as network inputs, thereby embedding spatial prior knowledge of scatterers to mitigate inherent artifacts such as boundary blurring and speckle noise. The framework features a modified U-Net architecture augmented with convolutional block attention modules (CBAMs) and residual blocks, enhancing feature extraction and segmentation robustness. Furthermore, adversarial training is introduced to refine the segmentation network using pseudo-labels generated from unlabeled DT images, enabling the discriminator to enforce physical consistency between labeled and unlabeled domains. Extensive simulations demonstrate the superiority of our method: when trained with only 100 labeled samples and 1,000 unlabeled samples, the proposed algorithm achieves a 23.0% reduction in mean squared error (MSE) compared to purely supervised counterparts. Additional validation on the handwritten digits and the “Austria” profile highlights its strong generalization capability for reconstructing unseen targets. This work bridges the gap between data-driven deep learning and physical priors, offering a practical solution for high-precision electromagnetic imaging under limited supervision.

A Novel Approach for Early-Stage Breast Cancer Detection

Mohamed Behih — 2026-01-30

This paper presents a novel approach for early-stage breast cancer detection using only a single radiofrequency 3D antenna sensor operating in several frequency bands below Ultra-Wide Band (UWB) frequencies. To this end, an innovative Inverted-F Antenna with Short Circuit-Like (IFA-SCL) is proposed, and the breast to be examined is fully placed inside this antenna between the radiating element and the ground plane. The designed and simulated antenna operates in the two frequency bands (902.8–928.0 MHz and 2.400–2.4835 GHz) of the Industrial, Scientific, and Medical (ISM) bands. After examination of the two patient breasts and by comparison of the antenna’s performances considering the return loss (S₁₁<−10 dB), tumor presence is detected when the resonance frequencies that cover the operating frequency bands corresponding to an unhealthy breast, are shifted to higher frequencies and the corresponding magnitudes are changed. A spherical shape model of the female breast tissues is created for designing and simulating antenna and tumor detection performances. Also, to test the practicality of the proposed method and detail the tumor detection performances with breast variability, three tests are performed using side-set and teardrop shape breasts. The present approach demonstrates great potential to become a new way for early-stage breast cancer detection, both quickly and with high efficiency. The proposed method, and thus the designed multi-band 3D antenna sensor, exhibits the capability to detect a tumor of spherical shape, of radius only 1 mm and embedded deeply in the breast. Furthermore, it is able to sense tumor presence even with breasts of various sizes and shapes. The patient’s safety is ensured by adhering to Specific Absorption Rate (SAR) limits.

Absorbed Power in Human Head Skin Due to Near-Field, Exposure up to 100 GHz

Fatih Kaburcuk — 2026-01-30

To prevent excessive skin temperature rise from overexposure due to near-field sources for frequencies from 6 to 300 GHz, international safety guidelines and standards for limiting exposure to electromagnetic (EM) waves introduce an incident power density (IPD) as an exposure reference limit and an absorbed power density (APD) as a basic restriction. At frequencies above 6 GHz, the penetration depth of EM waves in the human body model is particularly less, since EM wave penetration is more superficial in tissues. Therefore, the thickness of outermost tissues such as skin, which has different thicknesses in different realistic regions of a three-dimensional (3D) realistic human body, is a critical factor for the accuracy of EM dosimetry analysis. In this paper, the effect of skin thickness in a 3D planar head model on the spatially averaged APD over 1 cm2 and 4 cm2 areas due to near-field sources are investigated for the frequency range from 10 to 100 GHz. These investigations are performed using the finite-difference time-domain (FDTD) method considering different separation distances from the near-field source to the model of the skin surface. Numerical results show that skin thickness is the primary parameter in evaluating EM field exposure and the accuracy of EM dosimetry analysis.

Numerical Analysis of Large-Scale Phased Array Calibration, Using a Kronecker Product Formulation

Jake W. Liu — 2026-01-30

This paper proposes a calibration method for large-scale phased arrays based on a Kronecker product formulation, termed the Kronecker Product Method (KPM). Unlike the conventional Phase Toggling Method (PTM), which excites one element per measurement and suffers from limited measurement diversity, KPM employs simultaneous binary-phase excitation across all elements. This structure enhances robustness against noise and error accumulation, particularly in large arrays or high Signal-to-Noise Ratio (SNR) conditions. KPM retains PTM’s simplicity and 1-bit phase control compatibility while offering superior theoretical properties and practical performance.

Electromagnetic Shielding Effectiveness Prediction of Stacked, Chassis Based on Snow Ablation Optimizer Algorithm

Sen Wang — 2026-01-30

Stacked chassis has wide application in aerospace integrated electronic systems. However, with irregular enclosure structure and multi-module integrated into small space, its electromagnetic shielding effectiveness (SE) prediction is a challenge problem. In this paper, a novel method of SE prediction for stacked chassis is proposed based on snow-ablation optimizer (SAO) algorithm. First, a dedicated model of stacked chassis is selected and the SE at five internal sampling points is obtained via full-wave numerical simulation. Then, building on the Robinson equivalent model combined with the generalized Baum-Liu-Tesche (BLT) equations, and exploiting the initial simulation data as priors, the characteristic parameters of the stacked chassis are achieved via the SAO algorithm. On this basis and under vertically polarized and normally incident plane wave, the SE is predicted at all the positions across the central axis normal to the incident face of the chassis. The prediction results enable identification of the optimal SE distribution along frequency axis inside the chassis, thereby informing the internal layout and placement of sensitive devices. The prediction curves agree well with the simulation results, which can overcome the large-error limitations of conventional analytical approaches for complex cavities.

Optimization Design of Active and Passive Hybrid Shielding, for Electric Vehicle’s Wireless Power Transfer System

Yangyun Wu — 2026-01-30

Targeting the issues of electromagnetic exposure safety in the application of an electric vehicle’s wireless power transmission (WPT), this study proposes a surrounding active shield coils structure, laying on the four sides of the WPT system, which effectively reduces the lateral magnetic leakage field while supplementing the magnetic field inside the transmission channel. At the same time, this study proposes a ferrite groove structure as the passive shielding, achieving reduction of the vertical magnetic leakage field. On this basis, the paper takes system transfer efficiency and surrounding magnetic leakage field density as the optimization objectives, combining the extreme learning machine (ELM) surrogate model with multi-objective optimization algorithm for hybrid shielding structural design, realizing the further improvement of power transfer and electromagnetic shielding capability. A numerical simulation test is carried out and the results show that the proposed shielding scheme can ensure the system transfer efficiency, meanwhile reducing the magnetic leakage from all directions, and providing effective electromagnetic exposure safety protection for the human body.