This paper introduces a 5G multi-frequency antenna design method based on multi-objective sequential domain patching. By etching helical metamaterials on radiation patches and loading asymmetric electric-inductive-capacitive metamaterials on the transmission line side, the antenna structure is designed and optimized using electromagnetic simulation software, HFSS, and MATLAB. The resulting multi-frequency antenna operates across multiple frequency bands: n1, n41, 3.5G, and 4.9G. To ensure antenna performance and radiation efficiency, the antenna multi-frequency bands are precisely controlled. The experimental results show that the error between each operating frequency band and target frequency bands of the antenna is within 7.2%. Compared with conventional antenna design methods, the use of metamaterials and intelligent algorithms to optimize the structural parameters and loading positions of metamaterials can improve antenna performance while shortening the design cycle. Overall, this research offers novel insights into the design of 5G high-performance antennas.

Orbital angular momentum (OAM) becomes a new resource for wireless communication due to the different modes being orthogonal. In OAM-based wireless communications, factors such as tilt and multipath distort the phase of the OAM beam, making the mode difficult to detect. We propose a multi-label class-specific lightweight neural network (MCSLNN) to measure tilt and detect mode from a single image. MCSLNN utilizes the MobileNetV2 network as the backbone feature extraction network, considering the terminal devices with limited computing resources. To improve the performance of multi-label classification, MCSLNN employs residual class-specific attention (CSRA) as the classification layer. Furthermore, MCSLNN employs the beam steering method to verify the correctness of the measured tilt. The network measures the tilt angle with an accuracy of 76% and an estimation error of ±1° in a validation experiment. Finally, we analyze the network's generalization from varying heights above the ground for reflection paths. The results indicate that MCSLNN is adaptable to diverse circumstances, thus making it suitable for 6G communication and radar applications.

Scattering of electromagnetic waves by a dielectric object can be described as an integral equation involving a Green function. These types of problems can be solved using a spatial spectral formulation, which requires sampling of the spectral Green function. To avoid sampling around the singularities on or near the real axis, the spectral Green function is represented on three separate complex paths. Using appropriate selection functions, these paths are recombined such that the original Fourier integrals are retrieved. This composite path method provides a general way to solve domain integral equations involving Green functions with simple singularities with minimal computational overhead.

An innovative ceramic filter is presented in this paper. The filter is composed of six tuning apertures, coupling channels, and a six-blind-hole coupling structure. The six blind holes are divided into two groups: one situated between the first and second cavities to induce inductive coupling, and the other positioned between the second and third cavities to facilitate capacitive coupling. Employing this structure facilitates the formation of a cascade quadruple (CQ) coupling unit among the 1, 2, 3, and 4 resonant cavities, thereby introducing two transmission zeros. Subsequently, an analysis of the influences of the depth and spacing of each blind hole on the coupling coefficients is presented. Finally, to further validate the theory, the filter tailored for base station applications was designed and implemented. The measurement results demonstrate a center frequency of 3.5 GHz with a bandwidth of 200 MHz. In the passband, the insertion loss was below 1.2 dB, and the return loss surpassed 19 dB. The test outcomes align closely with the simulation, confirming the reliability of the design.

A hybrid excitation variable-leakage-flux machine (HE-VLFM) for electric vehicles is presented in this paper, which places the field windings on the outer side of the stator and provides an axial magnetic field through the magnetic ring and the magnetic bridge. The HE-VLFM proposed in this paper has a rotor permanent magnet to provide the main magnetic field and an excitation winding to provide a variable auxiliary magnetic field, which can fulfill the requirements of the electric vehicle drive system with large torque at low-speed and wide speed regulation range. The electromagnetic performance of the proposed HE-VLFM is evaluated using the equivalent magnetic circuit method and three-dimensional finite element analysis and the results show that the HE-VLFM has a well-developed regulating flux capability.

We apply the third-order susceptibility tensor generated by the Simplified Bond Hyperpolarizability Model (SBHM) to address the long standing challenges in distinguishing the bulk and surface quadrupolar second-harmonic-generation (SHG) contributions in diamond lattices, such as silicon, which exhibit bulk inversion symmetry. Assuming that the quadrupolar contribution originates from the interface gradient of the excited electric field, we demonstrate through symmetry considerations and numerical calculations for Si(001) and Si(111) facet orientations that it is not possible to separate the different quadrupolar contributions when the incoming light is incident normally. However, we show that such separation is achievable with oblique incidence. Furthermore, we propose a novel experimental design to measure the bulk and surface quadrupolar SHG contributions separately by introducing a semi-vicinal surface. Using numerical SBHM simulations, we show for the first time that this semi-vicinal setup can prove the existence of spatial dispersion, a nonlinear dipolar bulk effect recently proposed. This approach may lead to a better understanding of various nonlinear contributions in silicon and enable precise nonlinear surface monitoring.

In this paper, we present a numerical study to assess the robustness of an all-optical photonic limiter based on a two-dimensional (2D PC) TiO2 photonic crystal with a single ZnO nonlinear two-photon absorption (TPA) defect to manufacturing disturbances. These disturbances studied here concern diameters and positions. It is revealed that our limiter configuration is very robust to manufacturing errors.

In a permanent magnet synchronous machine (PMSM) system, the voltage mode-based rotor flux observer suffers from DC drift, primarily due to measurement errors, parameter variations, and non-zero initial states. To address this issue, the second-order flux observer (SOFO) is utilized, equipped with filtering capability aimed at reducing harmonic components. However, the DC offset induced by external disturbances cannot be completely eliminated by the second-order transfer function alone. Traditional magnetic flux correctors typically update correction values only at zero-crossing points of the magnetic flux. In this study, we propose an improved orthogonal flux corrector (IOFC) that combines a generalized integrator to effectively filter out the DC offset. In comparison with traditional OFC methods, our approach involves reconstructing two magnetic linkage functions, thereby doubling the correction frequency within a single cycle. Consequently, the frequency of correction term updates is threefold compared to conventional OFC methods. Finally, IOFC is implemented and tested on a PMSM platform for experimental verification.

We propose and numerically analyze a new type of notch filter oper-ating at terahertz frequencies, using a cylindrical periodic structure. This study takes place in the context of increasing demand for precise filtering devices in the terahertz frequency range, crucial for various applications in telecommuni-cations, sensing, and the medical field. This research focuses on the numerical analysis of the proposed structure, using the transfer matrix method to examine how changes in geometric parameters influence wave transmission. Particular attention is given to the effects introduced by the radii of the cylinders making up the structure. The principal results show that perfect symmetry (radii R₁ = R₂) produces no significant transmission dip, indicating the absence of reso-nance in the frequency band studied. This configuration allows the device to function as a passive filter. The introduction of asymmetry (R₁ = R₂) leads to the appearance of transmission dips, meaning that the device functions as a notch filter, capable of blocking specific frequencies. This phenomenon offers a method of selective filtering, by ``activating'' or ``deactivating'' the filter's behav-ior. Our research demonstrates the potential of the proposed cylindrical periodic structure as an innovative solution for the design of notch filters in the THz range. The ability to precisely control wave transmission through geometrical adjustments opens up new ways to develop highly selective filter devices adapt-able to various technological applications.

In the wireless power transfer (WPT) system of electric vehicles, reducing the magnetic leakage and minimizing the use of magnetic shielding materials while maintaining transmission efficiency are difficult problems. To this end, a single-hole compensated passive magnetic shielding structure is proposed in this paper, with the system's magnetic leakage reduced and transmission efficiency improved through metal shielding and passive shielding. First, the magnetic shielding principles and design concepts of the magnetic core, aluminum plate, and passive shielding coils are analyzed. The single-hole compensated passive magnetic shielding structure is proposed, and then a mathematical model of the structure is derived. Second, an optimization method is proposed, using Matlab and Ansys Maxwell software to reduce the volume of metallic materials while keeping magnetic leakage within a safe range. Finally, a WPT device based on the proposed structure is constructed according to the optimized magnetic shielding and coil parameters, and the effectiveness of the structure is validated through simulation and experimentation. The results demonstrate that, when the system output power is 4 kW, leakage is reduced by 62.7% compared to the single-hole unshielded coil structure using the same materials with the proposed structure. Compared to the all-aluminum plate and all-magnetic core structure, not only is leakage reduced by 1.2%, but there is also a reduction of 40.4% in magnetic core usage and 30.1% in aluminum plate usage. Moreover, the transmission efficiency reaches 93.49%.

A gradient-indexed core photonic crystal fiber (PCF) is proposed to realize sub-wavelength field confinement in the terahertz (THz) regime. It is verified that the gradient index (GRIN) profile PCF confirms superior field localization compared to the standard PCF. The in-plane quality factor of the GRIN PCF is evaluated as 2.2849 × 10⁹ which is 10 times greater than the conventional case. Moreover, the power fraction is found to be 84.04% and 99.69% along with the confinement loss of 0.31 dB/cm and 0.341 × 10^-7 dB/cm for the standard and GRIN type PCF at 0.2 THz. It is significant that the designed PCF also produces radial and azimuthal polarizations with enhanced field propagation due to the implicated triangular GRIN profile. The proposed GRIN PCF is useful for sub-THz communication, sensing and imaging applications.

A tightly-coupled patch array antenna, capable of broadband operation and low profile, is proposed in this study. The antenna subarray comprises four closely-coupled square patches that rotate in sequence. An impedance matching strip is utilized to achieve broadband matching of the antenna. By configuring the subarray with various feeding phases, it is possible to achieve the switching of six polarization states. The simulated and measured results suggest that the proposed antenna demonstrates a broad impedance matching band ranging from 4.2 GHz to 5.25 GHz (22.2%) over a scanning angle span of ±45˚, while maintaining a low profile of 0.033λ₀. The antenna's inherent simple structure, low profile, wide bandwidth, and favorable radiation characteristics position it as a promising option for multifunction phased array radar systems.

Currently, excessive AC loss in the rectangular winding motor used for electric vehicles poses a significant challenge, necessitating effective measures to suppress the losses. This paper focuses on the Prius IV motor, employing a finite element two-dimensional model established using JMAG software. The influence of conductor material and the number of rectangular winding layers on motor AC loss under various operating conditions is thoroughly analyzed. Maintaining a constant number of rectangular winding layers, aluminum (Al) conductors replace copper (Cu) conductors in 2-layer, 4-layer, 6-layer, and 8-layer configurations, respectively. AC losses are compared among motors with 4-layer, 6-layer, 8-layer, and 10-layer Cu rectangular windings, all having identical slot dimensions. Subsequently, the 10-layer Al conductor scheme is chosen to optimize motor design. The results demonstrate an average reduction in AC loss up to 59.24% after motor optimization, further reducing motor manufacturing costs.

A triple band MIMO antenna is designed and analysed at sub-6GHz for 5G applications on an FR-4 substrate. This paper contains the transition of an antenna from a simple microstrip antenna to the proposed defected L-shaped microstrip patch antenna, which comprises single, 2-element MIMO, and 4-element MIMO antennas with permittivity of 4.3, and the dimensions of those antennas are 60 × 60 mm², 60 × 120 mm², and 120 × 120 mm² correspondingly. These antennas resonate at three resonant frequencies which are 3 GHz, 4.1 GHz and 5.2 GHz under sub-6 GHz. HFSS has been used to design these antennas and to obtain the parameters like S-parameters, gain, VSWR and MIMO parameters like ECC, DG, TARC, and MEG. At those resonant frequencies, single element antenna has S₁₁ of -26.83 dB, -20.06 dB, and -19.16 dB; two element MIMO antennas have S₁₁ of -22.7 dB, -40.09, and -20.54 dB; and quad element MIMO antennas have S₁₁ of -15 dB, -24.8 dB, and -22.7 dB. The overall antenna gains are 2.5061 dBi, 3.1903 dBi, and 4.2989 dBi for single, 2-port, and 4-port MIMO antennas. This antenna is well suited for a range of applications including FWA systems that utilize 3 GHz frequency, Smart Cities and connected vehicles that rely on 4.1 GHz, and high-bandwidth activities such as video streaming, cloud computing, and mission-critical communications that require 5.2 GHz. Additionally, it can support future developments in both 5G and Wi-Fi technologies.

The Method of Moments (MoM) is widely used in Electromagnetic Compatibility (EMC) uncertain simulation due to its advantages, such as non-embedded simulation, high computational efficiency, and immunity from dimensional disasters. The theoretical research of the MoM has been relatively complete, but many of its key practical issues have not been fully discussed, which will result in the calculation accuracy in practical engineering applications falling short of theoretical expectations. With the help of the Feature Selective Validation (FSV) method, this paper analyzes and discusses two aspects. One is how to reasonably select the perturbation, and the other is the relationship between the uncertainty input size and the accuracy. By solving key practical issues of the MoM, the aim is to further promote it in the EMC field.

A spatial electromagnetic field analysis method is proposed by adding variable speed nodes to the circuit topology to estimate the optimal location of multiple mobile high-frequency power supplies at multiple nodes in this paper. In the process of continuous motion, the speed and position of motion affect the accumulated power and loss at the circuit node. At the same time, the transmission efficiency and delay characteristics of the high-frequency mobile power supply will also change with the precise positioning of the mobile power supply and the change of the spatially coupled electromagnetic field. The spatial electromagnetic field analysis method with variable speed nodes is used to divide the circuit topology of mobile high frequency power supply system according to the number of nodes. The continuous motion of variable speed nodes is used to simulate the real-time positioning of multiple mobile high-frequency power sources. By analyzing the real-time variation of the high-frequency electromagnetic field at variable speed nodes, the quantitative relationship between the electromagnetic characteristics of the node space and the speed and positioning of the mobile power supply is established. Finally, the fast optimal positioning of each mobile high-frequency power supply in the continuous moving process is obtained. Compared with the position estimation results obtained by the traditional relation calculation method, when the size is greater than 100, the proposed method can locate the position of multi-mobile high-frequency power supply faster and more accurately, and the circuit efficiency reaches 90%. The simulation results verify the correctness of the theoretical analysis.

For a conical log spiral antenna (CLSA), it is quite common to place the strip conductor conformally to the conical surface, and the antenna requires an extra impedance matching network. On the other hand, non-conformal orientation can solve the impedance matching issue, but fabrication is not as straightforward as conformal placement. This work considers the non-conformal placement of a strip conductor which facilitates self-matching while using smart additive manufacturing techniques for prototyping to ease the fabrication complexity. The impact of the additional dielectric support on the performance parameters of CLSA is investigated. Finally, the CLSA was prototyped using two different conductive elements (copper strip and conductive paint) on the 3D-printed support. Experimental and numerical results are shown to agree well for both copper strip and paint-based approaches. The self-matched CLSA provided a maximum impedance bandwidth of 128%, 3-dB axial ratio bandwidth (AR BW) of 63.56%, and gains of 10.32±1.94 dBi. The additive manufacturing techniques are shown to allow design flexibility and mitigate fabrication difficulties.

High altitude electromagnetic pulse (HEMP) couples to cables and introduces interference into the connected electronic equipment. Responses arising from the transient electromagnetic field typically follow an exponentially damped sinusoid behavior. Thus, damped sinusoids with different parameters are recommended in the International Electrotechnical Commission (IEC) standards as typical injected waveform for HEMP conducted immunity test. To guarantee the compliance of the injected pulse, accurate measurement of the injected pulse is needed. Wideband proportional current sensors are often applied to measure the injected damped sinusoid. However, bandwidth requirements of wideband proportional current sensor for damped sinusoid measurement are not specified. In this paper, two formulae are deduced to establish the relationships between the bandwidth requirements and the fundamental resonance frequency of the damped sinusoid to be measured. It is convenient and simple for the on-site engineers to check whether the bandwidth of the proportional current sensor is suitable by the formulae. Monte-Carlo simulation is conducted in support of the recommended formulae.

The issue of signal outages in sub-THz frequency communication for future 6G networks is addressed by this research. A machine learning method is proposed, employing Random Forest and K-Means algorithms to predict the optimal frequency band and outage probabilities for UAV relays. Both space and frequency diversity are explored to enhance signal strength, and metasurface-carrying UAVs are introduced with a 16 × 16 mm² design. This design significantly reduces the predicted outage probability from 0.1% to 0.0178%. Finally, triangular and hexagonal UAV swarm formations with metasurfaces are investigated, demonstrating improved performance through heatmap results.

A novel switchable dual-band bandpass filter (BPF) is proposed, where each passband can be independently controlled. The filter is composed of a tri-mode resonator, a dual-mode resonator, and feed lines coupling with the resonators. By controlling the PIN diodes loaded on the open end of the resonators, four operating states (i.e., dual bands, lower passband, upper passband, and all-stop band) are realized. A switchable dual-band BPF prototype is designed, fabricated, and measured with the center frequencies of 1.575 and 2.45 GHz having the bandwidths of 300 and 220 MHz, respectively. The prototype occupies an area of 0.128λ_g², where λ_g is the guided wavelength at the center frequency of the lower band (1.575 GHz). The measurement results indicate that the proposed switchable dual-band BPF has low insertion loss and high OFF-state suppression.