The integral equation fast Fourier transform (IE-FFT) is a fast algorithm for 3D electromagnetic scattering and radiation problems based on the interpolation of the Green's function. In this paper, a novel floating interpolation stencil topology is used to improve the IE-FFT algorithm. Compared to the traditional interpolation stencil topology, it can further reduce the storage and CPU time for the IE-FFT algorithm. The reduction is especially significant for volume integral equations. Furthermore, the accuracy of the algorithm is still good though the near-interaction element numbers are reduced. Finally, some numerical results including perfectly electric conductors, dielectric objects, composite conducting and dielectric objects are given to demonstrate the performance of the present method.
Wideband analysis of frequency dispersive geometries is a challenge in inverse scattering problems. Waveguide duct is an important case in aerial targets with dominant returns. Its dispersive behavior affects the range profile analysis due to occurrence of unwanted range extension. A new high frequency analysis using model based parameter estimation (MBPE) approach is presented. A group delay criteria derived from the nonlinear scattering phase response represents the duct length. Wideband sparse measured frequency domain samples of various waveguides are used as inputs to the model. Comparison is made with joint time-frequency analysis (JTFA) and inverse fast Fourier transform (IFFT) results.
In this paper, we propose a computational method for computing RCS of 3D conductor, by using piecewise surface impedance boundary conditions and forward backward iterative scheme. In our previous work, we have reported a numerical method combining Rytov's perturbation method and level set technique to construct a piecewise surface impedance, we showed that by using level set technique, we could model an arbitrarily shaped conductor by a piecewise distribution of low- and high-order SIBCs. The method proposed in this article postulates the use of local "buffer regions" to suppress spurious edge effects introduced by the abrupt termination of each SIBC and ensure stability of RCS computing.
Combining a two-year measurement and numerical approach, a semi-empirical model has been developed for prediction of rain attenuation at Ka-band in northern Taiwan. This was done using the drop size distribution (DSD) measurement and the extinction coefficient calculated by T-matrix, followed by regressing with rain attenuation measurements in all seasons. The attenuation due to rain can be estimated by calculating the extinction coefficient over all of the rain drops within the antenna beam volume. Comparing with the measured data demonstrates that the proposed model proves sufficiently accurate for Ka-band signal attenuation in site specific. For purpose of cross reference, we also compared the proposed model with Crane and ITU-R-P838 rain attenuation models. The RMS error and chi-square test shows that the proposed semi-empirical model has better performance to predicted rain attenuation than Crane and ITU-RP383 models, implied that both model predictions may not be quite reliable in some specific areas. Analysis suggests that seasonal effects are strong in signal attenuation due to rain types. It means that rain rate itself is not a quite reliable enough to be the single parameter in the rain attenuation model.
The purpose of this paper is to extend to spinor electromagnetism the differential forms, based on the Cartan exterior derivative and originally developed for tensor fields, in a very compact way. To this end, differential electromagnetic forms are first compared to conventional tensors. Then, using the local isomorphism between the O (3,C) and SL (2,C) groups supplying the well known connection between complex vectors and traceless second rank spinors, they are generalized to spinor electromagnetism and to Proca fields. These differential forms are finally expressed in terms of Hertz potentials.
This research presents the implementation of the Finite-Difference Time-Domain (FDTD) method for the solution of 3-dimensional electromagnetic problems in dispersive media using Graphics Processor Units (GPUs). By using the newly introduced CUDA technology, we illustrate the efficacy of GPUs in accelerating the FDTD computations by achieving appreciable speedup factors with great ease and at no extra hardware/software cost. We validate our approach by comparing the results with their corresponding simulated results obtained from Remcom's XFDTD software.
Exact formulas for internal impedance per unit length of tubular cylindrical conductors energized by time-harmonic current involve Bessel functions. These functions are defined by infinite series, which yield unstable and often erroneous results for complex arguments of large magnitudes. Although it is well known how to evaluate Bessel functions numerically and many routines are now available to perform the actual computation, the available software routines often fail when computing equations that consist of a product and a quotient of Bessel functions under large complex or real arguments. For such cases, different approximate formulas can be used. In this paper, three types of approximate formulas for internal impedance of tubular cylindrical conductors are compared with respect to numerical stability and accuracy.
Perfect electromagnetic conductor (PEMC) is a novel concept in electromagnetic fields of interesting properties and many potential applications. This paper introduces a new technique to design an artificial surface that has equivalent PEMC properties. The proposed PEMC boundary is based on a periodic structure composed of two conducting patches on a grounded dielectric slab. One of them is embedded inside the substrate and the other lies on the surface of the substrate. A conducting via is used to connect the two patches. In the resulting PEMC boundary, the polarization of the reflected wave is controlled by the tilting angle between the two patches.
In this work we propose a new class of optical pressure sensors suitable for robot tactile sensing. The sensors are based on a tapered optical fiber, where optical signals travel, embedded into a PDMS-gold nanocomposite material. By applying different pressure forces onto the PDMS-based nanocomposite we measure in real time the change of the optical transmittivity due to the coupling between the gold nanocomposite material and the tapered fiber region. The intensity reduction of the transmitted light intensity is correlated with the pressure force magnitude.
This work presents a two-dimensional (2D) subgrid technology for the geodesic finite different time-domain (FDTD) algorithm, which is applied to solve global extremely low frequency (ELF) electromagnetic EM wave propagation problems in the Earth-ionosphere system. The new technology provides arbitrarily locale resolution to study finer structure without disturb the global grid structure. Combined with the subgrid technique, the new geodesic FDTD algorithm can solve EM propagation problems in specific locale regions without extra computational burden. Based on the original geodesic FDTD formulations, the 2D subgrid technique is developed, and its computational stable relation is derived and analyzed. Then, possible three-dimensional (3D) subgrid structure is proposed. Finally, potential applications for the subgrid technique are suggested.
A composite medium containing perfectly conducting short needles can have a range of frequency for which the real part of the effective permittivity of the composite is negative. Such a range of frequency can be taken as negative bandwidth. This negative bandwidth for a composite medium is dependent upon parameters like positioning, orientation, length and needle density of short needles. Effects of random errors in positioning and orientation of short needles upon the ensemble averaged effective permittivity are analyzed. It is studied theoretically that increasing error in positioning and orientation of short needles reduces negative bandwidth.
In this paper, the authors propose a method based on the combination of inverse fast Fourier transform (IFFT) and modified particle swarm optimization for side lobe reduction of a thinned mutually coupled linear array of parallel half-wave length dipole antennas with specified maximum return loss. The generated pattern is broadside (φ=90 degree) in the horizontal plane. Mutual coupling between the half-wave length parallel dipole antennas has been taken care of by induced emf method considering the current distribution on each dipole to be sinusoidal. Directivity, first null beamwidth (FNBW), return loss of the thinned array is also calculated and compared with a fully populated array. Two cases have been considered, one with symmetric excitation voltage distribution and the other with asymmetric one. The method uses the property that for a linear array with uniform element spacing, an inverse Fourier transform relationship exists between the array factor and the element excitations. Inverse Fast Fourier Transform is used to calculate the array factor, which in turn reduces the computation time significantly. The element pattern of half-wave length dipole antenna has been assumed omnidirectional in the horizontal plane. Two examples are presented to show the flexibility and effectiveness of the proposed approach.
A methodology for electromagnetic simulation of initially charged structure with a discharge source (ICSWDS) has been investigated. The ICSWDS can be applied to a lot of areas such as high power electromagnetic (HPEM) radiators. As a method of electromagnetically simulating the ICSWDS, converting initially charged structures into equivalent transient structures and modeling discharge sources by using step voltage sources have been found. A Blumlein pulse forming line (PFL) has been simulated, manufactured and tested to validate this approach. A measured waveform from the test has a good agreement with a simulated waveform.
Propagation of electromagnetic fields and power in a circular waveguide containing chiral nihility metamaterial is studied. Space inside the waveguide is divided into two circular regions. One region contains chiral nihility metamaterial while other region is of free space. Two cases of the waveguide, in this regard, are considered for analysis. For the case of perfect electric conductor (PEC) waveguide, there is no net electric field and power propagation in chiral nihility region of the guide whereas both fields and power exist in non-nihility region (which is free space in our cases) of the guide. For perfect electromagnetic conductor (PEMC) waveguide, both electric and magnetic fields exist in the chiral nihility and non-nihility regions.
Our previous work has proved that the Monostatic Radar Cross Section (MRCS) of array antennas can be decomposed into the multiplication of array MRCS factor and element MRCS factor. The principle was derived in a special case that the array only had dipole antenna elements. However, many array antennas have more general antenna elements whose current is aperture distributed along the antenna structure. Obviously it encounters limited application problem when the principle is used to analyze more general array antennas other than dipole arrays. Therefore, the principle is extended into the more general array with arbitrary aperture antenna elements in this paper. In deriving the principle, the devices in the feed are assumed to have identical transmission and reflection coefficients. In order to validate the principle the scattering pattern of a waveguide slot array and an array with helix antenna elements are synthesized utilizing the array RCS factor. The simulation and calculation results prove that the principle is correct for the RCS pattern synthesis of general arrays with aperture antenna elements.
In this paper we propose a computational method for constructing variable surface impedance, based on combining Rytov's perturbation method and level set technique. It is well-known that the choice of the most appropriate order of Rytov's expansion is important both for accuracy and implementation. By using level set method, we constructed a piecewise distribution of low- and high-order surface impedance boundary conditions on the surface of an arbitrarily shaped conductor. It is found that the proposed method is able to give good results both in terms of accuracy and implementation cost.
To describe propagation of polarized electromagnetic wave within a disperse random medium a new Monte Carlo based technique with an adopted vector formalism has been developed. The technique has been applied for simulation of coherent backscattering of circularly polarized optical radiation from a random scattering medium. It has been found that the sign of helicity of circular polarized light does not change for a medium of point-like scatterers and can change significantly for the scatterers with the higher anisotropy. We conclude that the helicity flip of the circular polarized light can be observed in the tissue-like media. We find that this phenomenon manifests itself in case of limited number of scattering events and, apparently, can be attributed to the pulse character of incident radiation rather than to the specific form of scattering particles.
In this study, transmission characteristics of a novel THz wire waveguide --- conical metal wire with dielectric coating at 0.1-1 THz are studied. The investigation results show that the coated conical wire with virtually low attenuation and high energy concentration is a promising candidate as THz transmission medium. The calculation results agree well with that of simulation such as high frequency structure simulation (HFSS), which is based on the finite element method. In this paper, a novel transition from a coaxial line to the coated conical metal wire is designed. Although coaxial probe excitation has been used in microstrip lines and rectangular waveguides in microwave, millimeter-wave frequency domains, the present study shows that it is also an effective method to excite conical wire at THz frequency. As shown in the investigation results, the return loss of coax-conical wire transition is better than 20 dB from 0.1-0.5 THz, and the insertion loss is as low as 1 dB (the total length is 15 mm). It is a promising THz transition structure.
The paper describes pulse-based ultra-wideband (UWB) microwave imaging experiments for breast cancer detection using heterogeneous breast phantoms with dielectric properties mimicking the human breast. Three homogeneous and seven heterogeneous breast phantoms are used in tumor detection experiments. The phantoms have dielectric permittivity and conductivity higher than previously reported experiments, as well as clutters to represent the glandular tissue in human breast. The experiments are conducted in time-domain with pulse generator and real-time oscilloscope.
Low frequency imaging in radar domain can have applications for stealthy or buried targets. The transient scattering response from a ramp waveform is related to the profile function of the target, namely its transverse cross-sectional area along the line-of-sight, and thus provides information about the target size, orientation and geometrical shape. Such ramp responses can be used to generate a 3-dimensional image of the global shape of the target. Former imaging algorithm uses approximate limiting surfaces and is therefore limited to single convex objects. Here is proposed a new algorithm able to reconstruct non-convex as well as separated targets, from their ramp response signatures.