Computation of the magnetic field generated by permanent magnets is essential in the design and optimization of a wide range of applications. However, the existing methods to calculate the magnetic field can be time-consuming or ungeneralised. In this research, a deep learning-based fast-computed and generalised model of three-dimensional (3D) magnetic field is studied. The volumetric deep neural network model (V-Net) which consists of a contracting part to learn the geometrical context and an expanding part to enable the concise localization was applied. We synthetically generated the ground truth datasets from permanent magnets of different 3D shapes to train the V-Net. The accuracy and efficiency of this deep learning model are validated. Predicting on 50 random samples, the V-Net took 4.6 s with a GPU T4 and 23.2 s with the CPU whereas the others took a few hundreds to thousands of seconds. Therefore, the deep learning model can be potentially utilised to replace the other methods in the computation and study of the magnetic field for the design and optimization of magnetic devices (the codes used in this research are published openly in https://github.com/vantainguyen/3D_V-Net_MagneticField).
Wide bandwidth compact rectangular and equilateral triangular microstrip antennas employing slots cut bow-tie shape ground plane profile are proposed. Amongst all the designs, patch employing three rectangular slots cut bow-tie shape ground plane yields optimum results. Using the rectangular patch, against conventional ground plane design, increase in bandwidth by 20%, resonance frequency, substrate thickness, and patch area reduction by 32%, 0.034λg, and 61.12%, are respectively achieved. In equilateral triangular patch design, a three rectangular slots cut bow-tie shape ground plane configuration shows bandwidth increase by 30%, and substrate thickness, fundamental mode frequency, and patch size reduction by 0.027λg, 16.4%, and 36.28%, respectively. Proposed designs exhibit broadside radiation pattern with broadside gain of above 5 dBi.
This paper proposes a method of independent control over each passband in a high performance triple bandpass filter, which is an essential requirement in the field of microwave communication systems. Individual techniques are presented here to control the excited modes that are responsible for the generation of triple passbands based on substrate integrated waveguide loaded with semi-circular mushroom resonators. Initially, a circular substrate integrated waveguide (CSIW) loaded with two cascaded semi-circular mushroom resonators (ScMRs) with distinct modifications and orientations in the schematic is employed to generate three passbands. The fundamental mode and next higher order mode of the entire resonator structure are utilized to generate three passbands, and distinct techniques of mode perturbations and variation in coupling strength are introduced to independently control the excited modes. Subsequently, the methods established to control the excited modes are employed to independently control the center frequencies (CFs) of three passbands. All those methods established to control the CFs of the three passbands are verified with experimental results which show good agreement with the simulated ones.
Powerline communication (PLC) noise is the main cause of reduced performance and reliability of the communication channel. The major source of these noise bursts, which distort and degrade the communication signal, is the arbitrary plugging in and unplugging of electric devices from the electrical network. It is therefore important to perform statistical modelling of the PLC noise characteristics to enable the development and optimisation of reliable PLC systems. This paper presents the Variational Bayesian (VB) Gaussian Mixture (GM) modelling of the amplitude distribution of the indoor broadband PLC noise. In the proposed model, a fully Bayesian treatment is employed where the parameters of the GM model are assumed to be random variables. Consequently, prior distributions over the parameters are introduced. The VB criterion is used to determine the optimal number of components where the Bayesian information criterion emerges as a limiting case. To find the parameters of the GM components, the variational-expectation maximisation algorithm is employed. Measurements from different indoor PLC environments are then used to validate the model. Thereafter, performance analysis is carried out, and the VB framework is compared to the Maximum Likelihood (ML) estimate method. It is observed that while the ML model performs better when the amplitude distribution contains multiple peaks, the VB framework offers high accuracy and good generalization to the measured data and is thus effective in modelling the amplitude distribution of the PLC noise.
The backscattering of electromagnetic waves incident on a rotating metallic wind turbine (WT) is analyzed by using the Physical Optics method. The model developed is general and allows the computation of the spectral Doppler shift of the backscattered waves. All the parameters involved are taken into account, relative to incident wave direction, wind horizontal direction, WT geometric and electromagnetic properties. Numerical computations are carried out for various cases and presented relative to a search radar.
The near and far field EM responses over layered media have long been exploited in diversified applications such as remote sensing, monitoring and communication. In this work, we utilize the near field dependence of the EM fields of a three layered structure resembling air-sea ice-sea water to estimate the thickness and dispersion characteristics of sea ice using deep learning technique. We explore two key methods of field measurement termed as the fixed and scaled sweep methods. In the fixed radial sweep method, the receiver distance and height from the source are kept constant, and in the scaled sweep method both the receiver distance and height are set as a scaled function of the operating wavelength. A synthetic training dataset has been generated (using analytical computation and FEM simulation) in the low MHz band, which is used to train a deep learning model. The model is tested on different test datasets with frequencies inside, below and above the training limits. Even though the fixed sweep method is simpler to implement, the scaled sweep appears to perform better across the wide range of test frequency, both in and outside the training range. When the test frequency is inside the training range, the percentage errors for thickness, dielectric constant, and loss tangent were found to be <2%, <10%, and <5%, respectively, for the fixed radial sweep, whereas for the scaled sweep the percentage error is < 1% for all three measurement parameters. When the test frequency deviates further from the training range, the percentage error gradually increases. Later, we investigate the problem of determining sea ice thickness assuming a priori knowledge of sea ice dielectric parameters, and results show that the model estimates the thickness of the sea ice bulk with error as low as 0.1%.
This paper presents a novel theoretical and numerical approach for an infinitesimal current source (ICS) located in a planar isotropic multilayer medium. Using the mixed-potential integral equation (MPIE) formulation for depicting the electromagnetic disturbance created by the ICS, a detailed definition of Green's functions of Lorenz potentials and fields is provided in this paper. The proposed Green's functions are valid for the considered multilayer isotropic medium, which can have arbitrary layer parameters. This paper also analyzes two commonly observed special cases of the multilayer medium - the multilayer soil including air and the multilayer lossless dielectric - and the proposed equations are modified to meet the requirements of the medium. Green's functions can be obtained from the systems of linear equations proposed in this study. In comparison to other approaches, the advantage of the proposed procedure is that the solutions of the equations are immediately obtained in any field layer of the multilayer medium. In addition, the proposed system of linear equations can be solved easily using well-known numerical computation methods. Furthermore, this paper offers an alternative way of obtaining Green's functions, which are closed-form expressions for the kernels of spectral-domain Green's functions.
The equation of motion for a test particle moving in given fixed external fields is analyzed and compared to the corresponding equation of motion derived from the Darwin Lagrangian for a system of interacting charged particles. The two approaches agree as long as the part of the electric field that arises from the partial time derivative of the vector potential is taken into account. It is, however, only via the Darwin approach that the origin of this field can be understood as arising from a breakdown of the test particle approximation. Applying the formalism to an electron moving outside a long solenoid results in a classical analog of the Aharonov-Bohm effect.
A 3D printing and printed circuit board (PCB) hybrid fabricated modularized dual-stopband artificial magnetic conductor (AMC)-loaded filtering antenna is proposed for an X-band high-power radar system.By loading low-cost microstrip AMCs of different frequency responses into a waveguide slot array, we achieve a modularized filtering antenna whose frequency response can be simply controlled by replacing different AMCs. The waveguide slot array only works as a fixture to host different AMCs to achieve various filtering antenna frequency responses. The interchangeable modularized design helps to reduce the difficulty and cost of component fabrication by eliminating the need for complex resonant cavities inside the waveguide filtering antenna, which is time-efficient at the stage of product prototyping when numerous iterations are needed on a trial-and-error base. A dual-stopband filtering antenna is designed and fabricated in the X-band to verify the design concept. The passband covers 9.25-10.6 GHz with the passband gain greater than 10 dBi. The antenna radiates frequency-dependent scanning beams in the passband. The stopbands are 8.1-9 GHz and 10.75-11.5 GHz, and the out-of-band rejection is larger than 35 dB. The proposed design concept provides a different thought to achieve a low-cost filtering antenna by using interchangeable modularized components. The fabricated antenna prototype is a capable candidate for high-power airborne radar applications.
A dual frequency dual circularly polarized cross-slot waveguide array working at 4.9 GHz and 5.8 GHz is proposed for wireless communication/airborne weather radar applications. Different from the traditional cross-slotted waveguide antenna, to improve space utilization, two sets of cross-slots are slit on both sides of the longitudinal axis of the waveguide's E-plane to realize dual-frequency operation. When the antenna operates in the TE10 mode, the cross-slots on each side radiate left-handed and right-handed circularly polarized electromagnetic waves at two different frequencies, respectively. To suppress grating lobes, phase perturbation structures are periodically loaded in the waveguide to tune the propagation phase constant, thereby changing the effective electric spacing between radiating elements while keeping the antenna a compact physical aperture. The proposed grating lobe suppression method avoids the dielectric loss caused by dielectric loading, eliminates the need for complex array arrangement, and achieves the grating lobe suppression at dual frequencies simultaneously. The metallic 3D printing technology, selective laser melting (SLM), is used to fabricate the antenna in one piece in one run using aluminum alloy. The proposed antenna has gains of 10 dBic and 14.5 dBic with 47% and 69% aperture efficiencies at 4.9 GHz and 5.8 GHz, respectively. It is a capable candidate for air-to-ground (ATG) communication applications.