The Synthetic Aperture Interferometric Radiometer (SAIR) has demonstrated significant potential in Earth remote sensing and radio astronomy. However, most existing imaging methods rely on single-scale visibility function, while SAIR systems typically employ sparse arrays with insufficient sampling, which results in unsatisfactory imaging quality. In this paper, we propose a novel deep learning-based imaging method that addresses this limitation by leveraging multi-scale visibility function. The multi-scale visibility fusion network (MS-VFNet) introduces cross-attention mechanisms in the visibility domain for feature fusion across different scales, fully exploiting the implicit structural information, and subsequently reconstructs the brightness temperature images through a dedicated reconstruction module. The simulation results demonstrate that the proposed MS-VFNet achieves superior reconstruction accuracy and image quality compared to state-of-the-art methods, further validating the feasibility of multi-scale fusion in SAIR super-resolution imaging.

A compact four-element MIMO antenna with dimensions 40 × 32 × 1.2 mm³ is presented. The design employs stylized C-shaped radiators with slanted edges and a shared defected ground plane integrated with folded stubs to enhance impedance matching and bandwidth. Fabricated on an RT5880 substrate (ε_r = 2.2, tanδ = 0.0009), the antenna achieves an ultra-wide operating range of 4.24-19.70 GHz with inter-element isolation above 20 dB. Diversity metrics, including envelope correlation coefficient (< 0.02), diversity gain (≈10 dB), channel capacity loss (<0.0325 bits/s/Hz), and total active reflection coefficient (-12 to -24 dB), are verified through simulation and measurement, confirming robust MIMO characteristics. Owing to its wideband operation and high isolation, the antenna is suitable for sub-6 GHz 5G NR (n79: 4.4-5.0 GHz), WLAN/Wi-Fi (5.15-5.825 GHz), X-band navigation and radar (8-12 GHz), and Ku-band satellite communication (12-14.5 GHz).

To effectively suppress parameter perturbations, external disturbances, and ensure the stability of the PMSM system under uncertain conditions, this paper proposes a novel fast integral terminal sliding mode composite controller (NFITSMC) for speed-loop of PMSM based on super-twisting integral terminal sliding mode disturbance observer (STITSMO). Firstly, the mathematical model of the PMSM with parameter perturbations and external disturbances is analyzed. Then, the NFITSMC speed-loop controller is designed, where the NFIT sliding mode surface combines proportional, integral, and nonlinear terms, enabling effective suppression of parameter perturbations and external disturbances to ensure system stability under uncertainties. Meanwhile, the adaptive exponential switching reaching law adjusts the convergence speed according to the distance between the system state and the sliding surface, thereby mitigating system chattering. Next, the STITSMO disturbance observer is designed, in which the IT sliding mode surface is combined with a second-order super-twisting control law, allowing dynamic gain adjustment based on error magnitude to achieve global fast convergence of the adaptive nonlinear system. Finally, simulations and experiments validate that the NFITSMC-STITSMO composite controller demonstrates superior performance in finite-time convergence, robustness, chattering suppression, and disturbance rejection, making it suitable for high-performance and high-order PMSM control systems.

This paper investigates, by means of finite-element simulations, how the position of an aperture affects the electrostatic shielding effectiveness of a rectangular metallic enclosure. First, we compute the electric-field distribution on the surface of a completely closed enclosure placed in an external electric field. The results show that, for every wall, the field is weakest at the center, and that the field on walls parallel to the external field is far lower than that on walls perpendicular to it. Next, we determine the electric field that leaks into the enclosure after an aperture is introduced. We find that the field strength decreases with the distance from the aperture, that the field near the aperture is proportional to the surface field at the aperture's location when the aperture is covered. Also, its magnitude can be predicted by the classical model of the small aperture coupling. Finally, we investigate the coexistence effect and formulate guidelines for choosing the aperture position to achieve optimum shielding performance.

This study addresses multi-physics coupling challenges in high-speed sliding electrical contacts for Electromagnetic Launch Systems. A three-dimensional transient finite element model integrating electromagnetic-thermal interactions is established. By combining modified adhesion theory with Holm's contact resistance theory, we derive an analytical expression for dynamic contact resistance (DCR) that incorporates electromagnetic contact pressure, tangential friction, and temperature-dependent conductivity. The proposed localized modeling strategy with rail reverse motion technique efficiently resolves armature motion through coordinate transformation. Governing equations are discretized via the Galerkin method, with interfacial current continuity constraints and thermal partition coefficients enabling precise separation of Joule heating and frictional heat. Numerical results demonstrate robust current continuity compliance and reveal that DCR reaches its minimum during current peaks, exhibiting strong negative correlation with electromagnetic thrust. Tangential friction suppresses resistance rise by expanding the actual contact area, while electromagnetically dominated skin effects generate localized hotspots at contact trailing edges, armature limbs, and throat regions. This work elucidates the coupled frictional-Joule heating mechanisms governing multi-physics interactions, providing critical foundations for thermal management optimization in Electromagnetic Launch Systems.

In this paper, a wideband design of a conformal metasurface for RCS reduction in the range of 6.3~8.3 GHz is introduced. The proposed unit cell has a reflection amplitude less than -0.5 dB. The methodology for reducing the radar cross section using metamaterial is introduced along with the evolution of the design of the proposed unit cell. The conformal array is modeled using this unit, and the performance of a metasurface when being attached to a conformal metallic object is investigated. When the proposed metasurface is attached to a 3D object, it can achieve more diffused scattering patterns and wide scattering angles, regardless of the polarization of the incident waves and across a wide range of incident angles. It can achieve more than 10 dB reduction RCS. The sheet operates well for incident wave angles up to 60°. Both simulated and measured results demonstrate that the conformal metasurface effectively achieves diffuse reflection and RCS reduction which holds significant potential for applications in the field of advanced stealth technology, and the sheet size is scalable to larger sizes.

This paper presents an adaptive Direct Torque Control with Space Vector Modulation (DTC-SVM) strategy for a five-phase induction machine fed by a two-level inverter, designed to tolerate single open-phase faults. Under such fault conditions, the five-phase inverter generates only 16 voltage vectors distributed across 8 sectors, requiring a reconfiguration of the DTC-SVM control scheme. The proposed method introduces a modified vector selection and sector allocation approach tailored to faulted operation, enabling reliable performance without extensive controller redesign. Simulation studies in MATLAB/Simulink and experimental validation on a 3.5 kW five-phase induction motor confirm the effectiveness of the proposed approach. The results show that the adaptive DTC-SVM reduces torque ripple, maintains stable flux and current waveforms, and preserves fast dynamic response during sudden load changes. In addition, the method remains straightforward to implement, combining the simplicity of DTC with the improved voltage utilization of SVM.

A novel compact four-element MIMO antenna system is developed, featuring a rectangular Microstrip patch structure integrated with dual vertical grooves. The projected antenna design incorporate slits, triangular slot, and DGS to realize dual-band resonance with simultaneously multiband response and mitigating mutual coupling effects. The antenna operates effectively at 0.9 GHz and 2.4-2.6 GHz bands, supporting multiple IoT and wireless communication standards such as GSM (0.9 GHz), Lora (2.4 to 2.5 GHz), Wi-Fi (2.4 GHz), Bluetooth (2.4 GHz), LTE (Band 8 , Band 7, Band 41 and Band 53), and 5G (n8, n41). The antenna demonstrates satisfactory impedance matching, achieving return losses of -16.16 dB at 0.94 GHz and -31.19 dB at 2.496 GHz, with isolation levels exceeding -15 dB. The antenna delivers stable far-field radiation with 2.8-3.6 dBi gain, and prototype validation demonstrates strong agreement with simulations, confirming its suitability for compact, high-performance IoT (2.4 GHz) and 5G Sub-6 GHz devices.

This paper presents a broadband MIMO antenna for miniaturized satellites and a novel metamaterial for decoupling. The proposed metamaterial exhibits single-negative characteristics (with only permittivity ε being negative) in the 4.66-6 GHz and 6.5-9 GHz frequency bands, and double-negative characteristics (with both permittivity ε and permeability μ being negative) in the 6-6.5 GHz and 9-15.97 GHz frequency bands. In MIMO antenna applications, it significantly improves isolation between antenna elements. Finally, we develop a 4-port ultra-wideband (UWB) MIMO array antenna operating from 4.66 to 15.97 GHz, covering C-band, X-band, and Ku-band. It achieves S₁₁ < -10 dB with an operating bandwidth of 11.31 GHz. With the integrated metamaterial suppressing inter-unit coupling, the antenna demonstrates low-coupling performance (S₁₂, S₁₃, S₁₄ < −20 dB and ECC < 0.035) across the entire operating band. The maximum gain reaches 7.83 dBi, providing both high-gain performance and ultra-wideband decoupling capabilities. This MIMO antenna measures 55 mm × 55 mm × 1.6 mm and uses an FR4 substrate.

This paper proposes a composite control strategy integrating Adaptive Non-singular Terminal Sliding Mode Control (ANTSMC) with a Non-singular Fast Terminal Double-Power Sliding Mode Observer (NFTDPSMO) to achieve high-precision control of PMSM system. The strategy combines an adaptive non-singular terminal sliding mode controller with a novel sliding mode disturbance observer. The ANTSMC adaptively adjusts the convergence speed according to the distance between the system state and the sliding surface to suppress chattering, while the NFTDPSMO employs a triple-composite term with denominator modification to achieve singularity-free operation and global fast convergence. Simulated and experimental results demonstrate that under complex operating conditions including parameter perturbations, load variations, and external disturbances, the proposed composite controller achieves faster dynamic response, reduced current and torque pulsations, lower harmonic distortion (THD of only 7.1%), and significantly enhanced robustness and steady-state performance.

This paper proposes a novel microstrip patch antenna element and MIMO design based on quarter-mode substrate integrated waveguide (QMSIW). This design not only achieves antenna miniaturization but also effectively reduces the mutual coupling between antenna elements. The antenna element employs a triangular patch as the main radiator, with its long side grounded via two metal cavities. For bandwidth enhancement, a T-shaped strip is positioned at the center of the triangular patch's long side, and a new mode is introduced. A pair of slots is etched at the junction between the strip and the patch; adjusting the slot size enables dual-mode operation and control coupling. Building on this element, a 2 × 2 MIMO system is developed, featuring a compact size and requiring only one dielectric substrate, thereby achieving high integration and low cost. The patch occupies an area of (0.22 × 0.22λ₀²)/2, while the strip occupies 0.066 × 0.068λ₀² with high integration. The antenna achieves N78 band coverage with a total area of 0.0287λ₀². Experimental results demonstrate an 8.9% -10 dB impedance bandwidth (3.30-3.61 GHz) and -16 dB isolation, ensuring excellent overall performance. The antenna offers an effective solution for future 5G wireless communication systems.

Although Transformer models have made significant progress in the field of hyperspectral image denoising, their original architecture still has limitations in processing the spatial and spectral correlations of images. It often results in the loss of details in spatial features and insufficient exploration of the uniqueness of different spectral bands. To overcome these challenges, this paper proposes a Transformer based spatial spectral attention network aimed at enhancing the utilization efficiency of spatial spectral correlations. In response to the common problem of over smoothing in spatial feature processing, a dual channel spatial feature fusion module is introduced, which effectively enhances the capture of spatial details and ensures clear reproduction of image textures and edges. Meanwhile, in the spectral dimension, a multi-scale spectral feature extraction with self-attention mechanism is applied, which can sensitively identify and utilize the differences between spectral bands, thereby achieving more accurate feature extraction at the spectral level. By integrating residual connections in the spatial spectral feature extraction layer, the model can efficiently fuse spatial and spectral information, ultimately achieving high-quality denoising. The experimental results have verified the excellent performance of this method on both the ICVL dataset and the Urban real dataset, achieving good denoising results and demonstrating significant advantages in maintaining image details and spectral fidelity.

In this work, a multistatic antenna configuration with Compact Textile Wearable Antennas (CTWAs) is proposed for microwave imaging. The setup includes eight CTWA antennas printed on a denim jeans fabric wrapped around the phantom with a foam layer in between. The phantom is made of Delrin and is intended to replicate the breast tissue. It contains three PVC inclusions that mimic the tumors with sizes of 6 mm, 10 mm, and 3 mm. These inclusions are located at the center, near the outer surface, and between the center and the outer surface. The CTWA antennas surrounding the phantom collects the scattered data for further processing. Distorted Born Iterative Method (DBIM), along with the Reweighted Basis Pursuit (RWBP), a convex optimization algorithm, is used to reconstruct the object profile. RWBP is an iterative reconstruction algorithm in the sparse domain capable of extracting the size, shape, location, and dielectric distribution of the inclusions. The experiment is conducted using two different approaches. In one approach, multistatic without rotation, the receiver antenna is fixed at positions 0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315°. In the other approach, multistatic with rotation, the receiver antenna is rotated in steps of 9° about the center of the phantom to achieve angles in between. In the multistatic configuration without rotation, the inclusions of 10 mm, 6 mm, and 3 mm were detected with errors of 0.125, 0.35, and 0.433, respectively, While in the case of multistatic configuration with rotation, the inclusions were detected with errors of 0.070, 0.066 and 0.03 respectively. The results obtained are compared with the previous studies available in the literature. These results demonstrate that the proposed wearable multistatic antenna configuration is suitable for medical imaging applications, enabling better target detection, localization, comfort, and flexibility.

For assessing Geomagnetically Induced Currents (GICs) risks in mid-latitude coastal power grids, this study developed a three-dimensional Earth conductivity model that incorporates coastal effects. The model was constructed using geological cross-sectional data from the China Earthquake Administration and measured terrestrial conductivity data. Focusing on the strong geomagnetic storm of February 27, 2023, the spatiotemporal characteristics of GICs in Shandong Peninsula across 34 substations and 46 transmission lines were accurately computed. The GICs peak of the substation appears at Muping , and the GICs peak of the transmission line appears at Zouxian-Luzhou Line. Notably, two coastal substations exceeded the safety limit for GICs. Areas of highest risk are concentrated in the southeastern coastal region of Shandong, underscoring the significant impact of coastal effects and changes in Geological Structure. This method and its findings provide a global reference for predicting and issuing early warnings regarding GICs in long-distance mid-latitude coastal power grids.

Coarse discretization introduces significant errors in the solution of scattering problems, in part due to discretization errors in the contrast operator. We present a procedure for the automatic construction of a modified contrast operator for electromagnetic scattering problems by using trainable neural networks to represent a modified contrast operator. We achieve a higher accuracy on a coarse discretization while still keeping computation time down compared to a fine discretization. By using synthetic data from a full-wave Maxwell solver to train the network for one-dimensional slab scatterers and two-dimensional polygonal scatterers, we are able to use the techniques found in deep learning to improve accuracy in coarse-grid forward scattering problems.

An original computational framework is developed to simulate the propagation of ultrashort laser pulses with arbitrary temporal and spectral profiles through uniform linear dielectric materials. The study investigates how spectral phase sampling during propagation affects computational efficiency and accuracy. The proposed approach enables a comprehensive analysis of ultrashort pulse evolution in both the time and frequency domains. To demonstrate its effectiveness, the algorithm is applied to various propagation phenomena, such as temporal and spectral shifts, pulse broadening, asymmetric distortions, and chirping in dispersive media, using a wavelet-based time-frequency decomposition.

In this paper, a broadband circularly polarized (CP) magneto-electric (ME) dipole antenna based on microstrip line aperture-coupled feeding structure is proposed. The antenna is fed by the microstrip line, which is utilized to couple the energy to the antenna through the slot etched on the ground plane. Based on a linearly polarized (LP) ME dipole antenna, two centrosymmetric L-shaped strips are loaded to the patches located at the +45° diagonal position and the patches located at the -45° diagonal position are truncated. As a result, the current direction is changed to be parallel to the equivalent magnetic current to radiate CP wave. To improve the axial-ratio (AR) bandwidth of the antenna, the rectangular slot initially etched on the ground plane is modified to an asymmetric cross slot, which can generate a minimum AR point at high frequency. In this way, the AR bandwidth is increased from 32.4% to 51.7%. On this basis, to further extend the AR bandwidth, the metal columns are introduced at both ends of the antenna, and another additional AR minimum point is generated at the low frequency. The measured results indicate that the impedance bandwidth (|S₁₁| ≤ -10 dB) is 57% (2.40-4.33 GHz) and the 3 dB AR bandwidth is 63% (2.39-4.57 GHz). Moreover, the proposed antenna exhibits flat gain and stable unidirectional radiation pattern across the operational frequency region.

This paper introduces a compact, circularly polarized (CP) truncated-corners patch antenna with a highly stable phase center for Global Positioning System (GPS) L1 applications. The antenna is excited by a central coaxial feed and features two corner truncations and a square-shaped slot with an integrated tab to generate CP. Designed for the GPS L1 band (1.563-1.587 GHz), the antenna achieves a -10 dB impedance matching bandwidth of 100 MHz (1.52-1.62 GHz), covering the target frequency range. Key performance metrics include a wide 3-dB axial ratio (AR) beamwidth of 190° and a fractional AR bandwidth of 1.5%. A prototype was fabricated and tested, and the measured results show reasonable agreement with simulations, confirming the design's effectiveness.

This paper presents a single-layered dual-band microstrip patch antenna with periodic sinusoidal slots. The antenna operates within 860-930 MHz and 2400-2500 MHz. The first band encompasses both the European as well as North American UHF-RFID bands while the second band covers the 2.45 GHz Microwave-RFID/WLAN band. Moreover, the presence of the sinusoidal slots helps the antenna exhibit frequency-dependent beam scanning within the 2.45 GHz band. Four such antenna units are placed close to each other with different orientations to create a 2 × 2 antenna system. This system can detect UHF-RFID tags using both horizontal and vertical polarizations. It can also perform frequency-dependent beam scanning with horizontal polarization in the yz-plane and vertical polarization in the xz-plane. The beam maximum scans from -300 to 310 in the yz-plane and -320 to 290 in the xz-plane.

The article presents the geometry and optimization of a small-sized dual-band F-shaped monopole antenna for medical applications in the 2.45/5.8 GHz ISM bands. The antenna is printed on a partial ground FR4 substrate, and its geometrical parameters are optimized by a hybrid procedure based on Artificial Neural Networks (ANNs) and Particle Swarm Optimization (PSO).Two ANN architectures, Feedforward Backpropagation Neural Network (FFBPN) and Generalized Regression Neural Network (GRNN), are trained on a dataset of 121 samples generated using varying patch sizes. The FFBPN model is better with a mean squared error of 3.9506 × 10^-5 at epoch 21. The optimized antenna has resonances at 2.448 GHz and 5.8 GHz with S₁₁ < -10 dB, bandwidths of 610 MHz (2.21-2.82 GHz) and 400 MHz (5.61-6.01 GHz), and peak gains of 2.262 dBi and 3.861 dBi, respectively. Measurements on a prototype are in agreement with simulations to confirm the appropriateness of the design for wireless medical devices, such as wearable sensors and telemedicine systems.