We consider the problems of numerical solution of volume integral equations with the kernels depending on the difference of arguments, based on discretization and iteration methods. Many important classes of problems of mathematical physics can be described by equations of this form; for example, this is the case for problems of electromagnetic and acoustic scattering on three-dimensional transparent bodies.

View full abstract

The problem of diffraction of an external electromagnetic field by a locally nonhomogeneous body in a perfectly conducting waveguide is examined. This problem is reduced to solving a volume singular integral equation (VSIE). The examination of this equation is based on the analysis of the corresponding boundary value problem (BVP) for the system of Maxwell's equations and the equivalence of this BVP and VSIE. The existence and uniqueness for VSIE in L₂ are proved. A numerical Galerkin method for the solution of VSIE is proposed and its convergence is proved.
Inverse boundary value problem for determination of permittivity of dielectric body in a waveguide is reduced to the nonlinear volume singular integral equation.

View full abstract

This paper proposes a design method of arbitrarily-shaped bandpass filters with multiple resonators in the limited space. Although the proposed method is based on the genetic-algorithm (GA) optimization, we newly introduce the fitness function which estimates not only the magnitude of S-parameters but also the phase characteristics in a passband. Conventional GAs without phase evaluation are difficult to design a filter having the specified number of resonators efficiently, whereas the present method can easily construct such a filter. As an example, arbitrarily-shaped planar-circuit filters with 3 and 4 resonators are optimized in the limited space and fabricated. The effectiveness of the present technique is verified by comparison of frequency responses between the calculated and the measured results.

View full abstract

Obtaining meaningful numerical data for circular disks having radii larger than 3.0 wavelengths using double precision or quadruple precision arithmetic has proven difficult. The present study investigates the reason for this difficulty and obtains numerical data for diffracted fields of disks of larger radii. In addition, the characteristics of the diffracted field are discussed. Calculation of the current distribution using the GNU Multiple Precision Arithmetic Library and the Multiple Precision Floating-Point Reliable Library are demonstrated to provide meaningful data for disks having radii larger than 3.5 wavelengths. Finally, the calculation precision of the coefficients of simultaneous linear equations is revealed to be the source of the difficulty in obtaining meaningful numerical data for circular disks of larger radii.

View full abstract

Medical image diagnosis and computer aided diagnosis are modern important medical techniques developed with computer technology. Particularly, medical image diagnosis using optical waves of lasers is very important technical tools for physiological examination of human body. Transmission properties mainly depend on optical absorption effects due to biological characteristics consisting of atomic and molecular structures. Image responses of optical transmitted projection include optical scattering characteristics that disturb transmission properties through biological structures. Computer numerical simulations of spatial filtering for optical scattering waves superposed on transmitted and attenuated waves are discussed to improve image diagnosis. By numerical simulation of FDTD method, statistical characteristics of optical waves are discussed for optical propagation, attenuation and scattering in random inhomogeneous biological media that consist of random particle models given by random numbers programming. Spatial filtering characteristics of grid structure are shown for exact image optical projection excluding scattering effects through physiological media. Spatial filtering characteristics for off-axial scattering optical waves are graphically shown by propagation properties with large attenuation in lossy waveguide-type grids by FDTD method.

View full abstract

Over set grid generation method coupled with FDTD method is proposed for the analysis of the moving boundary problems while considering the Doppler Effect. As a bench mark, the received wave at the observation point is calculated when the observation point moves and compared with the theoretical results. Good agreements were obtained.

View full abstract

In this paper, we propose a new polarization splitter constructed from a circular core fiber (fiber1) and an elliptical core fiber (fiber2). The polarization splitters are designed so that the x-polarization and the y-polarization are derived from the fiber2 and the fiber1, respectively. The device length L_m and the bandwidth BW_-15 at R≤-15dB (R: the extinction ratio) are discussed by adjusting mainly the ellipticity e₂ of the core in the fiber2 and the relative index difference Δ₂ of the fiber2. When e₂ and Δ₂ are varied in the range 0.60≤e₂≤0.85 and 0.5%≤Δ₂≤1.0%, respectively, it is found that: (1) L_m takes its minimum value 18.96mm at Δ₂=1.0% and e₂=0.79; (2) BW_-15 takes its maximum value 4.98nm at Δ₂=1.0% and e₂=0.85.

View full abstract

The peak value of transmitted pulse in printed circuit boards (PCB) is important for a pulse peak detection devices. When an input line and an output line are connected to each pad with the direction of right angle, the propagating pulses with the narrow time duration separate into some parts and decrease the peak value of pulse response. This paper presents an improved line-pad connected structure. The microstrip line is in contact with a pad from outside by considering the pulse propagation time passing through the via structure. We obtained the large peak value of the pulse response for which the time duration is larger than 0.2ps.

View full abstract

In this paper, we propose an improved method to analyze the scattering problem by dielectric gratings with dielectric rectangular cylinders sandwiched between two multilayers, by combining an improved Fourier series expansion method and multilayer method for both TM and TE waves.
Numerical results are given for the influence of incident angle and frequency of the transmitted power in terms of the parameters ε ₃/ε ₀ of rectangular cylinders in the middle layer, yielding the basic characteristics of resonance peak for switching or frequency selective devices for both TM and TE cases.

View full abstract

The present study investigated whole-body specific absorption rate of a human body in a vehicle cabin for plane-wave exposure. The rationale for this investigation is that fields in the vehicle without human have been enhanced in particular frequency region due to standing waves, and thus power absorption in the human body is of interest. For our computational results, the whole-body average specific absorption rate of the human in the vehicle was found to be 60% smaller than that in free space. The reason for this upset is that the standing wave over the vehicle cabin was suppressed due to power absorption by the human.

View full abstract

The point matching method can perform very precise electromagnetic simulation, however, error analysis still remains to be a difficult task. We propose a novel approach to predict the computational error and discuss the way to select proper truncation mode numbers.

View full abstract

An alternative definition of the wavelet transform matrix used for the sparsification of the moment matrix in the method of moments (MoM) is developed. In scattering problems, if the scatterer is a finite periodic structure, the moment matrix has a periodic structure. By using the periodicity, we establish a new approach to construct the wavelet transform matrix that yields more efficient sparsification. The scattering of plane waves from finite periodic structures, e.g., linear conductor array, is considered to demonstrate the effectiveness of the proposed method. Numerical results indicate that the proposed method yields a sparser coefficient matrix and achieves more significant saving in CPU time compared to the conventional approach, with sufficiently accurate solutions.

View full abstract