The paper proposes a microwave absorber in the form of half-spheres placed on a base layer as an alternative to the conventional pyramidal shape. The performance of the absorber is investigated, focusing on the influence of the diameter of the half-sphere, permittivity, loss tangent, and angle of the incident waves. The study also evaluates the case when the absorber is backed by a conducting plate that is required for shielded anechoic chambers. Simulation results using the CST Microwave suite indicate that increasing loss tangent enhances absorption while decreasing permittivity reduces reflectivity. The proposed absorber offers low sensitivity of reflection concerning the angle of incidence due to the symmetry of the spherical surface. The results show that the proposed absorber has a comparative performance to the pyramidal absorber of the same total thickness.

In recent years, uncertainty analysis methods have become a research hotspot in the field of Electromagnetic Compatibility (EMC), and non-intrusive uncertainty analysis methods are widely used in the field of EMC due to their advantages such as easy solver generalization and easy programming. The proposal of Bayesian optimization-based uncertainty analysis method further enhances the competitiveness of non-intrusive uncertainty analysis methods in solving complex EMC simulation problems. However, in traditional Bayesian optimization-based uncertainty analysis methods, Latin hypercube sampling strategy is used to construct the initial Gaussian process model, which lacks adaptive adjustment capability, and the quality of the initial Gaussian process model has a significant impact on the efficiency of subsequent calculations and the accuracy of the final results. This defect limits the computational efficiency and accuracy of Bayesian optimization methods in uncertainty analysis applications. In response to this issue, this paper proposes an active sampling strategy based on the Stochastic Reduced Order Model (SROM) method. This strategy improves the fitness function used by the SROM method in clustering to enhance the representativeness of the training set to the sampling space. By using this active sampling strategy instead of Latin hypercube sampling strategy, a higher quality initial Gaussian process model can be constructed, and the accuracy of Bayesian optimization method uncertainty analysis calculation is improved in the example, verifying the effectiveness of the proposed initial sampling point selection improvement strategy.

The transfer matrix method (TMM) with scattering matrices has been a valuable tool, facilitating the rapid characterization of multilayer devices in a very fast, stable, and memory-efficient manner. This paper presents a generalization of TMM with improved scattering matrices capable of simulating devices with full nine-element material tensors for the layers and any combination of signs for the real and imaginary parts of the isotropic external regions. The formulation of the bianisotropic transfer matrix method (BTMM) algorithm is covered in detail, and notes on implementation are provided. Example devices found in literature were used to benchmark the accuracy of the algorithm. The simulation of the bianisotropic device was corroborated with a bianisotropic finite-difference frequency-domain (FDFD) algorithm and a finite-element method (FEM).

This paper presents a dual-band antenna design for X/Ku bands, featuring asymmetrical feed ports, slot-loaded radiating patches, and a reconfigured defect ground structure (DGS) integrated with a back air cavity to achieve 32.11% reduced patch area and directional radiation. By coplanar waveguide (CPW) feeding and vertical interconnect accesses (VIAs) connecting all metal layers, the antenna miniaturizes via extended surface current paths and broadens bandwidth via DGS. Simulation and measurement show operating bands of 8.30-12.29 GHz (38.75% fractional bandwidth) and 12.91-14.21 GHz (9.58%), with measured gains aligning well with simulations. Compared to traditional dual-band designs, this work reduces physical size by over 30% while maintaining high gain, making it suitable for compact satellite communication, radar, and microwave energy systems.

This paper presents the design and performance evaluation of a flexible, multilayer wearable antenna optimized for sub-6 GHz 5G applications at 3.5 GHz. The proposed antenna introduces a fabrication-ready stackable design using textile-compatible materials, including Felt, a 2 mm EVA foam layer, and Shieldit Super. A key innovation lies in the use of low-permittivity EVA foam as an intermediate spacer, which enhances gain and impedance matching without requiring additional structural elements, thus maintaining a compact and mechanically flexible profile. The antenna achieves a peak realized gain of 7.81 dBi and a wide impedance bandwidth of approximately 15.7%, within a total thickness of just 4.34 mm. The design remains robust under bending and close-body scenarios, with specific absorption rate (SAR) analysis confirming compliance with international safety standards. Experimental and simulated results validate the antenna's consistent performance, underscoring its suitability for wearable and Wireless Body Area Network (WBAN) applications in future 5G systems.

To address the convergence inefficiency of the conventional CS-Krylov-block method in solving electromagnetic scattering problems, this paper presents an adaptive Krylov subspace basis function method (AKSBFM) based on spectral energy thresholds. In this method, Krylov subspace basis functions (KSBFs) are first generated within each extended subdomain using localized self-impedance matrices. Singular value decomposition (SVD) is performed on the candidate basis set to evaluate energy contributions, and only the dominant components exceeding a predefined energy threshold are retained. As a result, the number of basis functions per subdomain is automatically adjusted, and a compact, well-conditioned reduced matrix system is constructed. This energy-guided truncation significantly eliminates redundant modes, yielding improved numerical stability and reducing the condition number by up to two orders of magnitude. Numerical experiments demonstrate that, compared with the traditional CS-Krylov-block method, AKSBFM improves computational efficiency while ensuring computational accuracy.

In the past few years, deep learning has emerged as a transformative force in tackling challenges within the realm of electromagnetic inverse scattering, driving remarkable advances and reshaping conventional approaches. Among them, the physics-assisted learning method that combines traditional inverse scattering algorithms with deep neural networks has demonstrated excellent real-time inversion capability and lower computational complexity. For two-dimensional inverse scattering problems, an approximate solution of the target is first obtained using a linear approximation algorithm, followed by mapping learning from low to high precision with a neural network. To enhance both precision and generalizability, this study integrates a Transformer module into a CBAM U-Net framework, giving rise to a refined architecture aptly named TransAtten U-Net. By retaining certain positional information while enhancing the correlations between features, the overall feature extraction effect is improved. Through simulation experiments, the paper compares the performance of the proposed TransAtten U-Net two-step method, TransAtten U-Net direct method, and CBAM U-Net two-step method. Experimental results demonstrate that the proposed TransAtten U-Net two-step method not only achieves higher accuracy than the other two approaches, but also exhibits a stronger generalization capability across diverse scenarios, along with enabling real-time imaging.

This paper proposes a single-port microwave sensor for solid material characterization, based on defected ground structures (DGSs) with complementary split-ring resonators (CSRRs). Fabricated on an RO5880 substrate, the sensor was analyzed through both simulation and experimental measurement. Electromagnetic simulation and optimization were conducted using CST Studio Suite within the 1.5-3.0 GHz frequency range. The sensor's performance was evaluated with three materials of known permittivity: RO5880, RO4350, and FR-4. Results show that the two proposed configurations, one with a DGS CSRR (Design A) and the other with an added slot on the DGS CSRR (Design B) yielding Q-factors of 332 and 357, respectively. The higher Q-factor in Design B indicates increased sensitivity across all tested materials compared to Design A. For example, Design B achieved the highest sensitivity of 4.71% for RO5880 material compared to Design A. Thus, the added slot enhanced field coupling, improving measurement sensitivity and confirming the sensor's suitability for microwave-based solid material characterization.

An analysis of electromagnetic scattering by multi-story, multi-window buildings is presented by utilizing the Kirchhoff Approximation method. This investigation specifically analyzes scattering characteristics for vertically polarized incident plane waves. Scattering fields are calculated via radiation integrals associated with equivalent current sources induced on the building's exterior and across virtually closed window apertures by the incident wave. Fields within window regions are represented using rectangular waveguide modes, enabling the conversion of reflected fields from window glass into equivalent currents. The formulation's validity is established through comparisons with the physical optics method and scale model measurements. Discussions address the influence of window glass and polarization on wave propagation in wireless communication scenarios like 4G LTE operating at 700 MHz.

Use of a small radiating loop (12 cm diameter) is recommended in EMC standards (Mil-Std-461G:2015(RS101) and IEC 61000-4-39:2017) for immunity testing with low-frequency magnetic fields. We investigated the limitations of this method using finite-element simulations. We studied the effects of fields from radiating loops with different radii and their induced voltages in different diameter receiver loops that represented wiring of equipment under test (EUT). We also studied the windowing-method recommended in those standards. It involves positioning the loop successively over all locations on each face of the EUT. Our results show that this radiating loop can only simulate exposure to larger real-world EM fields when the EUT's wiring area is smaller than the radiating loop. Another limitation is that the magnetic field from the radiating loop drops significantly with distance perpendicular to the loop surface. Therefore, the windowing-method with a small radiating loop is only suitable for simulating exposures to real-world sources with fields that do not extend a large distance from the loop. In addition, the field distribution (width and depth) of the real-world EM source must be accounted for before deciding to use a small radiating loop for immunity testing of an EUT.

This work is motivated by the recently domination of mobile and wireless communications technologies, in addition to the fast evolution of the new generation of mobile and wireless communication that leads to the successive mobile generations XG 3G, 4G, 5G, and in near future 6G. Currently, the fifth generation (5G) technologies still need more development for a compact and efficient device. In this manuscript, Meta cells units (metamaterial and metasurface) are employed for improving the main parameters of antenna performance like gain (G), bandwidth (BW), reflection coefficient ($S_{11}$), and radiation efficiency. The proposed antenna design shows multiple resonant frequencies which means that the design is able to operate at multiband of frequencies, including 6\,GHz band that considers the main targeted band of 4G and 5G mobile communication technologies. The simulation results, for the different models via adding meta cells to the proposed design model, show excellent improvement for the performance parameters that are improved excellently in comparison with the conventional circular patch (CCP) and previous literature. In addition, the use of meta cells reduced the resonant frequency which means that it can serve a lower frequency with small size of substrate, and it is half size of CCP, making it suitable for many applications that require a compact antenna design.

In this paper, a Quadband Octagon Patch Antenna is designed whose operating frequencies are 20.22 GHz, 23.64 GHz, 27.35 GHz, 28 GHz, 28.37 GHz which achieved return losses of -19.67 dB, -19.14 dB, -19.66 dB, -19.04 dB, -19.04 dB, -20.41 dB and gain of 6.1 dB, respectively. The substrate employed in this antenna is FR4, which features a dielectric constant value 4.4 and a loss tangent value of 0.002. With a radiation efficiency of 90.85%. This antenna is small, measuring only 15 × 25 × 1.6 mm³ dimensions. Multiple slots of different lengths are inserted in an antenna to form a 5G Quadband Octagon Patch Antenna in order to accommodate more operating frequencies. Later, a 2*2 MIMO octagon patch antenna having a 3.6 mm radius, 50 × 30 × 1.6 mm³ of dimensions, 4.4 dielectric constant, and 1.6 mm thickness with defective ground structure is designed. Here this single quad band antenna was turned to a broadband MIMO antenna by means of a defective ground structure.

This paper presents a compact multiple-input multiple-output (MIMO) microstrip antenna system covering the 2.4 GHz and 5 GHz wireless local area network (WLAN) bands. By etching rectangular slots on the microstrip patch and adjusting the dimensions of both the antenna and the rectangular slots, the antenna system can radiate at the operating frequencies of 2.44 GHz and 5.3 GHz simultaneously. To achieve high port isolation across the two distinct WLAN bands, a ``WM''-shaped defected ground structure (DGS) is etched on the ground plane to reduce mutual coupling in the 2.44/5.3 GHz frequency bands. Simulation results demonstrate that within the frequency ranges of 2.41-2.49 GHz and 5.22-5.39 GHz, the isolation of the two dual-band antenna systems achieves maximum coupling suppression of 26.7 dB and 14 dB, respectively. This DGS can serve as a potential solution for decoupling in WLAN MIMO antennas.

This research offers a 12-element antenna array optimized for MIMO utilization in fifth-generation (5G) mobile phones. The antennas operate in the sub-6 GHz long-term evolution (LTE) frequency range, specifically between 3.4 and 3.6 GHz. To fulfil the growing demand for faster data speeds and reliable connection in 5G networks, the presented MIMO antenna setup offers a balance between compact size and high performance, making it well suited for integration into smartphones. Every radiating element in the array is tuned to approximately 3.5 GHz and features an open-slot structure, which effectively reduces mutual coupling and enhances isolation. Antenna arrangement has been constructed on an FR-4 substrate of dimensions 150 mm × 80 mm × 0.8 mm, corresponding to the layout restrictions of standard 6-inches smartphones. A prototype was developed to validate the design through measurements. The results demonstrate excellent impedance matching (return loss > 10 dB), high isolation (> 20 dB), strong radiation efficiency (exceeding 66%), and a low envelope correlation coefficient (< 0.03) covering the target frequency range.

In this paper, machine learning supports a compact electromagnetic band gap structure (EBG) based dual band microwave sensor which is proposed for dielectric characterization of liquids with high sensitivity. Two edges, located via metalized holes, are electrically coupled with a suspended microstrip line. Two channels are placed in the electric field region of each EBG patch. Therefore, the change in frequency shift and quality factor are observed, which will help to describe the dielectric characterization of Liquid Under Test (LUT). A matrix-based mathematical model, and machine learning based prediction model are developed for the calibration and validation of the sensor. The results are experimentally verified through fabricated prototype for the binary mixture of water and ethanol. The proposed sensor achieved a compactness with size of 0.164λ_2.47GHz × 0.164λ_2.47GHz, an average sensitivity of 0.931, 0.243, and a quality factor of 170, 230 for band-1 and band-2, respectively. The calculated dielectric constant of different samples shows good agreement with the values reported in the literature. The machine learning based model is developed using the Support Vector Regression algorithm and achieves the high value of coefficient determination (R²) which is 99.01, and the less root mean square error (RMSE) value is 0.009.

To address the problem that the control performance of permanent magnet synchronous motor (PMSM) in model predictive torque control (MPTC) is highly sensitive to motor parameters, an improved model predictive torque control scheme for PMSM based on anti-stagnation particle swarm online parameter identification (ASPSO-IMPTC) is proposed. First, an improved MPTC strategy based on inductance and magnetic chain parameter compensation is proposed. Compared with conventional MPTC, the proposed method can acquire accurate motor parameters in real-time, thereby enhancing both the control performance and parameter robustness of PMSM. Second, a review mechanism is proposed to enhance traditional PSO parameter identification. This method prevents particle swarm stagnation, enhances the parameter identification ability of the traditional method, and improves the real-time accuracy of the motor parameters. The parameter robustness of the motor is further enhanced. Finally, the experimental results show that the proposed ASPSO-IMPTC strategy can effectively improve the control performance and parameter robustness of PMSM when parameters mismatch occurs in PMSM.

In this paper, a racket-shaped ultra-wideband (UWB) multiple-input multiple-output (MIMO) antenna is analytically designed using characteristic mode analysis. The antenna has an overall size of 60 × 60 × 1.6 mm³ and consists of four racket-type radiating elements, four ground planes shaped like the number 6, and a cross-shaped decoupling structure between the radiating units. In the single antenna configuration, the feed position is determined by analyzing the current and electric field distributions of its characteristic modes. The bandwidth and current distribution are optimized by integrating seven small rings, L-shaped branches, and etched slots to ensure the simultaneous excitation of six characteristic modes, thereby enabling its UWB performance. In the MIMO setup, four elements are orthogonally arranged, and a cross-shaped decoupling structure along with a defected ground structure is employed to reduce mutual coupling, achieving over 20 dB isolation between any two elements. Simulated and measured results confirm that the antenna operates over the 3-21 GHz range, fully encompassing the UWB range of 3.1-10.6 GHz. Furthermore, the antenna achieves up to 77% radiation efficiency, a peak gain of 5.75 dBi, and a low envelope correlation coefficient (ECC).

In this paper, a compact multiple-input multiple-output (MIMO) antenna is proposed for unmanned aerial vehicles (UAVs). By simultaneously exciting common mode (CM) and differential mode (DM) from a T-shaped slot, wideband coverage is achieved. Four such slot antennas are used to form a four-antenna module operating in the N79 band (4.4-5.0 GHz). The size of the four-antenna module is merely 43 × 8 mm², demonstrating excellent miniaturization and integration. The dominant coupling between adjacent elements occurs through currents of the same mode. When CM and DM currents coexist, partial cancellation of coupled currents at the feed point enables high isolation without ex-ternal decoupling structures. Two modules are symmetrically positioned along the longer edges of the frame, forming an 8-element MIMO antenna. The antenna achieves isolation greater than 11 dB and an envelope correlation coefficient (ECC) below 0.04. The measured total efficiency is better than 52%, with an average of 56%. Featuring compact footprint, zero-clearance constraint and high isola-tion, the proposed antenna is a promising candidate for 5G UAVs.

Aiming at the issues of large current ripple and significant torque pulsation in switched reluctance motor (SRM) model predictive current control (MPCC) under varying operating conditions, this paper innovatively proposes a novel SRM model predictive current control method integrating an Extended State Observer (ESO) and the Lehuy model. By constructing a nonlinear current prediction framework based on the Lehuy model, the data dependency on traditional Look-Up Table (LUT) methods is significantly reduced. Meanwhile, the real-time compensation of system disturbances is achieved by introducing the ESO, resolving parameter mismatch issues under dynamic operating conditions. Simulated and experimental results demonstrate that this method, implemented on a 12/8-pole SRM prototype, achieves a current ripple reduction of 41.5% and torque pulsation suppression of 32.7% compared to traditional LUT-MPCC. This research provides new insights into the robust control of SRMs in high-precision servo scenarios.

This paper proposes a multistep reflector backing for compact spiral antennas to provide a consistent unidirectional pattern over a wide frequency band with improved gain, Axial Ratio (AR), and efficiency for 2-18 GHz applications. The effect of the variation of step number and step sizes of the reflector on antenna parameters has also been studied in the proposed work. The fabricated prototype antenna provides good impedance matching and circular polarization over the entire frequency range and is compact in size with a height of 0.078 wavelengths (λ_m) at lowest frequency. The antenna exhibits a rotating radiation pattern with frequency in azimuth direction providing almost a constant beamwidth of 117° with the variation in gain limited to only 0.75 dBic above 8 GHz, yielding a flat gain response at higher frequencies. The compact size and improved parameters of the designed wideband antenna with a stepped reflector makes it a suitable candidate for electronic warfare applications.