The multi-objective optimization of the six-pole outer rotor hybrid magnetic bearing (OSHMB) not only solves the nonlinear and strong coupling problems of the three-pole magnetic bearing (THMB), but also makes the magnetic bearing structure more compact and improves the maximum bearing capacity. Firstly, the structure and working principle of the OSHMB are introduced, and the mathematical models of suspension forces are established by the Maxwell tensor method. Secondly, the key parameters of the OSHMB are multi-objective optimized, and an optimal set of parameters is obtained through the sensitivity analysis, constructing the response surface model, and the multi-objective optimization based on the genetic algorithm. Based on the optimal parameters, the force current characteristics and maximum carrying capacity of the OSHMB are analyzed. Finally, the experimental platform is built. The suspension experiments, anti-interference experiments and load loading experiments are performed. It can be seen that the maximum bearing capacity of the OSHMB is about 9.6% higher than that of the SHMB.

For full duplex communication, a signal parallel transfer method based on partial power transmission couplers is proposed in this paper. The power transfer uses a serial LC compensation structure topology, and the data transmission channel adopts a double coupling resonant circuit. In terms of power transmission, some power coupling inductors and power compensation capacitors form a power resonance network with a high frequency trap function, which can isolate the influence of signal transmission. Therefore, there is no need for an additional trap, which reduces power loss and the space occupied by the structure. In terms of signal transmission, the partial coupling coil method can increase the coupling frequency and the data transfer rate. In addition, the signal transmission circuit has the characteristics of dual resonance frequencies. The forward and reverse signals modulate the carrier at different resonance frequencies to realize full duplex communication. Finally, the simulation results prove that the scheme is practicable for full duplex communication and parallel transmission of power, achieving anoutput power of 1.4 KW, and the highest transmission rate can reach 1 Mbps.

A dual-polarized wide-angle scanning array antenna is proposed in this paper. The proposed antenna array consists of sixteen elements with the working band from 9.5 to 10.5 GHz. A microstrip patch fed from two orthogonal directions is applied to achieve dual-polarization. In order to obtain good impedance matching and wide bandwidth of the antenna, capacitive coupling feeding is adopted. The measured results show that the proposed array can cover a wide scanning range of ±58°. The polarization isolations of antenna are higher than 17 dB. The isolations between receiving sub-array and transmitting sub-array are higher than 22.3 dB. The proposed array antenna is suitable for Van Atta applications.

In this paper, a novel tunable LC bandpass filter (BPF) based on LC magnetic-dominant mixed coupling is proposed. The design equations for the coupling coefficient and resonating frequency are given. The magnetic dominant coupling region and electric dominant coupling region are studied. The magnetic-dominant mixed coupling is used to compensate the bandwidth of the tunable filter, so that the tunable filter with constant absolute bandwidth can be obtained. The filter is designed, simulated and measured, and the measurement matches the simulation very well. The measurement shows that the central frequency tuning range is from 72 MHz to 222 MHz with -3dB bandwidth of 16.5±3.5 MHz.

This paper proposes a novel miniaturized interdigital capacitor loaded interdigital filter, which is applied in C-band (3.2 GHz~4.2 GHz). By loading an interdigital capacitor on the open end of the resonator of the interdigital filter, the length of the resonator is shortened by 28%. The resonant frequency offset caused by tap introduction is adjusted by using the method of impedance compensation at the open end of resonator 1 and resonator 5, which further reduces the size of the filter. The miniaturized filter is fabricated on a 0.254 mm-thickness alumina substrate with relative dielectric constant of 9.8 by thin film process. Measured results are as follows: the passband of the filter is 3.2 GHz~4.2 GHz; the insertion at center frequency is -1 dB; the return loss is less than -18.3 dB. The size of the filter is 4.98 mm*6.45 mm (0.15λ_g*0.20λ_g), which is 37.8% smaller than that of the traditional interdigital filter.

This paper explores a loaded conductor backed coplanar waveguide (CB-CPW) split ring resonator (SRR) fed U-slot planar antenna used for healthcare monitoring via the wireless scientific industrial medical (ISM) band and medical service band at fifth generation (5G-MSB). The antenna has been designed with bio-tissue layers, muscle layers, skin, and fat. The parameters of the designed antennas, such as miniaturization, increased gain, and enhanced bandwidth, are presented. The proposed prototype results in the total size of 640 mm³. Such designed antenna has been operated at (3.4-3.6) GHz - fifth-generation medical service band and at (2.38-2.48) GHz - industrial scientific band and can realize proximately omnidirectional radiation pattern over the operating bands.

This paper presents the application of machine learning-based approach toward prediction of path loss for the large intelligent surface-assisted wireless communication in smart radio environment. Two bagging ensemble methods, namely K-nearest neighbor and random forest, are exploited to build the path loss prediction models by using the training dataset. To generate the data samples without having to run measurement campaign, a path loss model is developed owning to the similarity between the large intelligent surface-assisted wireless communication and the reflector antenna system. Simple path loss expression is deduced from the system gain of the reflector antenna system, and it is used to generate the data samples. Simulation results are presented to verify the prediction accuracy of the path loss predictions models. The prediction performances of the trained path loss models are assessed based on the complexity and accuracy metrics, including R² score, mean absolute error, and root mean square error. It is demonstrated that the machine learning-based models can provide high prediction accuracy and acceptable complexity. The K-nearest neighbor algorithm outperforms random forest algorithm, and it has smaller prediction errors.

Based on the principle of bionics, this paper combines the design of flexible bionic antenna with Chinese culture, and proposes a dual-material bionic antenna with Electromagnetic Band Gap(EBG) structure. The antenna uses a polyimide flexible substrate. Radiation patch of this antenna is shaped like a ``pear flower'', and the ``CHINA'' shaped slot is etched on the ground to form a Logo mark. In order to reduce the impact of antenna radiation on human body, the introduction of an EBG structure made of Polydimethylsiloxane (PDMS) material makes the front-to-back ratio of the antenna radiation significantly increased. The antenna was bent in different ways and was placed on human body model for simulation and testing. The results showed that the antenna achieved an impedance bandwidth of 18.8% (2.22-2.46 GHz), the peak gain was 4.02 dBi, and the antenna was low sensitive to deformation, which makes it suitable for modern flexible electronic equipment.

A Compact wideband operating from 3 to 18 GHz MIMO antenna with quadruple notches is presented in this paper. The elements in MIMO configuration are arranged in orthogonal fashion with each other to minimize the coupling effects. The antenna consists of circular rings and a modified microstrip feed. By engraving a crescent shaped slot, split ring-shaped slot, circle shaped slot in the circular monopole, rectangular spiral shaped slot engraved along the feed line quadruple notches are attained. The antenna operates from 3 GHz to 18 GHz with notches in the range of 3.2 GHz-4.2 GHz centered at 3.5 GHz, 4.5 GHz-5.5 GHz centered at 4.9 GHz, 6.2 GHz-7.3 GHz centered at 6.6 GHz, and 8.1 GHz-8.8 GHz centered at 8.5 GHz. The element has a very compact size of 0.28λx0.22λx0.016λ at 3 GHz and is hence suitable for portable devices.

Beam Collection Efficiency (BCE), sidelobe level outside the receiving area (CSL), and cost are need to be considered in optimizing the transmitting array of a Microwave Wireless Power Transmission (MWPT) system. To solve the problem of too low BCE caused by dividing a small number of subarrays, this paper proposes a novel one-step subarray partition algorithm named Multi-Particle Multi-Parameter Dynamic Weight Particle Swarm Optimization Subarray Partition (MPMP-DWPSO-SP). The algorithm optimizes the position and structure of each element at the same time, and the number of the subarrays is no more than 4. It is verified by simulation that the BCE obtained by using this algorithm to optimize the Sparse Quadrant Symmetrical Rectangular Array (SQSRA) with an aperture of 4.5λ×4.5λ and the array element number of 8×8 can reach more than 90%. In addition, a new intelligent optimization model is designed for dividing the 8×8 array into 2 subarrays, and BCE and CSL can reach 91.69% and -17.61 dB.

We introduce a theory of optical responses of bianisotropic layers with arbitrary effective medium parameters, which results in generalized Fresnel-Airy equations for reflection and transmission coefficients at all incidence directions and polarizations. The poles of these equations provide explicit expressions for the dispersion of Fabry-Perot resonances and surface electromatic waves in bianisotropic layers and interfaces. The existence conditions of these resonances are topologically related to the zeros of the high-k characteristic function h(k)=0 of bulk bianisotropic materials and taxonomy of bianisotropic media according to the hyperbolic topological classes [32, 33].

Since the first working multilevel fast multipole algorithm (MLFMA) for electromagnetic simulations was proposed by Chew's group in 1995, this algorithm has been recognized as one of the most powerful tools for numerical solutions of extremely large electromagnetic problems with complex geometries. It has been parallelized with different strategies to explore the computing power of supercomputers, increasing the size of solvable problems from millions to tens of billions of unknowns, thereby addressing the crucial demand arising from practical applications in a sense. This paper provides a comprehensive review of state-of-the-art parallel approaches of the MLFMA, especially on a newly proposed ternary parallelization scheme and its acceleration on graphics processing unit (GPU) clusters. We discuss and numerically study the advantages of the ternary parallelization scheme and demonstrate its flexibility and efficiency.

Radio-frequency electromagnetic waves can be harnessed to produce an alternative source of energy to replace batteries in many low-power device applications. An efficient radio frequency (RF) energy harvesting circuit was designed and constructed using a dynamic Pi-matching network in order to convert frequency-modulated electromagnetic waves in the range of 88-108 MHz to direct current through a 3-step process. The circuit consists of a 50 Ω copper plate dipole antenna, a Pi impedance matching network, and a five-stage voltage doubler circuit. These three modules are connected through SubMiniature version A (SMA) connectors for convenient assembly. The dynamic Pi matching technique for RF energy harvesting is theoretically explained and simulated in the Advance Design System software environment. The experimental values obtained in this proposed work are in good agreement with the simulations. The harvesting system is capable of producing up to 14.3 V direct current voltage across a 100 kΩ load in field tests carried out at a displacement of 760 m from a transmission tower. At 6.7 km from the tower, a DC value of 61.5 mV was still obtainable at the ground level. The direct-current power that was generated through the energy harvesting was applied for the demonstration of three tasks with satisfactory results: illuminating a light-emitting diode, energy storage in a Panasonic VL2020 rechargeable battery, and activation of a TMP20AIDCKT temperature sensor in an urban area which enabled low power device activation and energy storage.

To restore power feeding as soon as possible and reduce repair costs and labor, a precise and robust fault location method for transmission lines is proposed. This method is based on the current and voltage synchronously collected by the phasor measurement units (PMUs) at two terminals of the line and does not require line parameters to calculate the fault distance. The line parameter is not approximately constant, but is affected by power load, temperature, and humidity, which affects the accuracy of most fault location algorithms that rely on line parameters. Therefore, the method proposed in this paper is robust and accurate. The method is based on the sequence fault component network and synchronous measurement technology, which is not affected by the system's pre-fault state, fault type, fault inception angle, and fault phase. Then, the method is verified in PSCAD/EMTDC by choosing different path resistances, fault types, fault inception angles, load currents, and line transpositions. A large number of simulation results show that the proposed method has high accuracy and robustness.

We propose a compact microstrip patch antenna that uses a negative permittivity substrate to achieve an end-fire radiation pattern. The antenna is designed to operate at X-band frequencies with a patch footprint of 0.9λ × 0.05λ and a thickness of λ/20. We show that loading a narrow patch with a negative permittivity substrate introduces an effective shunt inductance that resonates with the strong fringing capacitance of the patch. At resonance, the electric field is vertically polarized and approximately uniform across the patch, producing transverse nulls that improve the directivity of the antenna. The negative permittivity substrate is implemented using a thin-wire effective medium with four vias spread across the patch. The antenna is matched to 50 Ω using a quarter-wavelength transformer. The fabricated antenna operates at 10.8 GHz with a peak return loss of 30 dB and a bi-directional directivity of 10.7 dBi. The antenna has a 10-dB impedance bandwidth of 3.8% and radiates with a simulated efficiency of 93%.

Although the design of multiband band-pass filters (MBPFs) has been thoroughly studied in the literature, the synthesis of high-order and multiple pass-band filters with controllable transmission zeros (TZs) and high band-to-band isolation is hardly feasible. In this paper, we present a novel design strategy to cope with this issue. Adopting a star-like topology, the proposed design method is based on the parallel association of N-1 band-stop stepped-impedance stubs to form an N pass-bands resonator. We show that such a simple design principle allows the accurate control of TZs positions. The principle and theory of these associated band-stop resonators (ABSRs) based filter are exposed, and their efficiency is shown through the synthesis, design, simulation, and measurement of quad-band and quint-band band-pass filters. Very good in-band filter performance and very high band-to-band isolation are achieved for both filters without the need for complex optimization process. These results make the ABSRs an attractive solution to achieve multiple band responses with advanced specifications.

A portable 4D snapshot hyperspectral imager (P4DS imager) with compact size, fast imaging time, low cost, and simple design is proposed and demonstrated. The key components of the system are a projector, a liquid crystal tunable filter (LCTF), and a camera. It has two operating modes dependent on the set state of the LCTF: a 3D light measurement mode that produces a 3D point cloud reconstruction of the object, and a hyperspectral imaging mode yielding spectral data. The camera imaging plane is the same for both operating modes allowing the collected spatial and spectral data to be directly fused into a 4D data set without post-processing. The P4DS imager has excellent performance with a spectral resolution of 10 nm, a spatial depth accuracy of 55.7 um, and total 4D imaging time of 0.8 s. 4D imaging experiments of three different samples, colored doll statue, green broccoli, and a human face, are presented to demonstrate the efficiency and applicability of the system. Due to being cost-effective, portable, and good imaging performance, the proposed system is suitable for commercialization and mass production.

Birds and bats are at risk when they are flying near wind turbines (WT). Hence, a protection of bats and birds is postulated to reduce their mortality e.g. due to collisions with the rotor-blades. The use of radar technology for monitoring wind energy installations is becoming increasingly attractive for WT operators, as it offers many advantages over other sensor systems. Timely localization and classification of the approaching animal species is very crucial about the reaction measures for collision avoidance. In this work, a localization, classification and flight path prediction technique has been developed and tested based on simulated radar signals. This allowed us to classify three different birds and one bat species with an accuracy of 90.18%. For accurate localization and target tracking, five frequency modulated continuous wave (FMCW) radars operating in Ka-Band were placed on the tower of the WT for 360˚ monitoring of the WT.

Small footprint of the multi-input-multi-output (MIMO) antenna is extremely desirable for space-constrained ultra-wideband (UWB) communication systems. Compact MIMO antennas with improved isolation and wide operating bandwidth are the significant subject of the work. Therefore, this paper presents a miniaturized four-port polarization diversity UWB-MIMO antenna operating in the frequency range of 3.1-12 GHz with band-notched characteristics. Four octagon-shaped radiating elements with a common ground are placed orthogonal to each other for good isolation. Band rejection features between 4.5 and 5.5 GHz were achieved by including an open-ended slot at the upper edge of the octagon-shaped antenna. The MIMO antenna was etched on a low-cost 32.3 x 32.3 x 0.8 mm³ FR-4 dielectric substrate. The antenna radiates in a quasi-omnidirectional pattern on the H-plane throughout the operational bandwidth, with higher than 15 dB isolation, low envelope correlation, and high antenna gain. As a result, this antenna is well suited for diverse applications and portable devices.

Compact low-profile four and eight elements Multi-Input Multi-Output (MIMO) antenna arrays are presented for 5G smartphone devices. The proposed antenna systems can operate at two dual-wideband with triple resonance frequencies that cover the extended Personal Communication Purposes (PCS) n25 band and other related applications, the mobile china's band, and the LTE Band-46. The proposed antenna element is designed based on modified Minkowski and Peanocurves fractal geometries. Desirable antenna miniaturization with multi-band capability is obtained by utilizing the space-filling and self-similarity properties of the proposed hybrid fractal geometries where the overall antenna size is (11.47 mm × 7.19 mm). All antennas are printed on the surface layer of the main mobile board. Based on the self-isolated property, good isolation is attained without employing additional decoupling structures and/or isolation techniques, increasing system complexity and reducing antenna efficiency. For evaluating the performance of the proposed antenna systems, the scattering parameters, antenna efficiencies, antenna gains, antenna radiation characteristics, envelope correlation coefficients (ECCs) and mean effective gains (MEGs) are investigated. The performances are evaluated to confirm the suitability of the proposed MIMO antenna systems for 5G mobile terminals. The proposed eight elements MIMO system has been fabricated and tested. The measured and simulated results are in good agreement.