Publications (76)42.24 Total impact
 [Show abstract] [Hide abstract]
ABSTRACT: A polymer electrolyte fuel cell (PEFC) is one of desired clean energy converters. In order to raise the power generation efficiency of PEFC, it is necessary to monitor the generation of current inside MEA (membrane electrode assembly) in PEFC. In this paper, a noncontacting technique for detecting the distribution of power generation current inside the MEA using the static magnetic field around the PEFC is examined. The estimating method of the generation current distribution using the extended evolution strategy by 3D FEM (finite element method) is proposed, and the effectiveness of the method is investigated by verification experiment using an actual fuel cell. 
 [Show abstract] [Hide abstract]
ABSTRACT: In this paper, we propose a novel method for estimating the position and posture of a smartphone which has a magnetic dipole inside a speaker by using the weighted integral method with a linear magnetic sensor. We derive a system of linear equations relating the dipole position and the weighted integrals of the magnetic flux density on the linear sensor. We prove that, although the coefficient matrix for a single linear sensor is possible to be singular, use of two linear sensors that do not exist on the same plane enables us to uniquely localize the dipole. Based on the simulations, we develop a noncoplanar, cruciform array of magnetic impedance sensors. We demonstrate that the method is applicable to position and posture estimation of a neodymium magnet and a smartphone. 

Article: Orthogonal polynomial approach to estimation of poles of rational functions from data on open curves
[Show abstract] [Hide abstract]
ABSTRACT: This paper deals with the problem of finding poles of rational functions from function values on open curves in the complex plane. For this problem, Nara and Ando recently proposed an algorithm that reduces the problem to a system of linear equations through contour integration. The main aim of this paper is to analyze and improve this algorithm by giving a new interpretation to the algorithm in terms of orthogonal polynomials. It is demonstrated that the system of linear equations is not always uniquely solvable and that this difficulty can be remedied by doubling the number of the linear equations. Moreover, to cope with discretization errors caused by numerical integration, we introduce new polynomials similar, in spirit, to discrete orthogonal polynomials, which yield an algorithm free from discretization errors.  [Show abstract] [Hide abstract]
ABSTRACT: This paper presents a novel method for the localization of a passive radio frequency identification (RFID) tag in a 2D space. First, we show that the position of the tag can be estimated as the intersection of two orthogonal lines, which are determined by the weighted integrals of the magnetic flux density created by the tag. To measure them, we developed a square magnetic sensor composed of six coils: four rectangular coils, a set of four bowtieshaped coils connected in series, and a square coil. Using this sensor with side lengths of 200 mm, an RFID tag was localized in a 180 mm square domain with an average error of 5.1 mm. Using the sensor with side lengths of 400 mm, the RFID tag was also localized in a 300 mm square domain with an average error of 1.8 mm.  [Show abstract] [Hide abstract]
ABSTRACT: Spatial filtering velocimetry was proposed in 1963 by Ator as a velocitysensing technique for aerial cameracontrol systems. The total intensity of a moving surface is observed through a set of parallelslit reticles, resulting in a narrowband temporal signal whose frequency is directly proportional to the image velocity. However, even despite its historical importance and inherent technical advantages, the mathematical formulation of this technique is only valid when infinitelength observation in both space and time is possible, which causes significant errors in most applications where a small receptive window and high resolution in both axes are desired. In this study, we apply a novel mathematical technique, the weighted integral method, to solve this problem, and obtain exact sensing schemes and algorithms for finite (arbitrarily small but nonzero) size reticles and shorttime estimation. Practical considerations for utilizing these schemes are also explored both theoretically and experimentally.  [Show abstract] [Hide abstract]
ABSTRACT: For nondestructive inspection of ferromagnetic pipes using magnetic flux leakage (MFL), a sensor consisting of two coils to measure the discrete Fourier sine and cosine coefficients of MFL is developed. The position of a crack in a pipe can be estimated from the squared sum and the ratio of the voltages induced in the two coils. Each coil has sixteen elemental coils with 78 permalloy cores which are placed along a circle with a radius of 6.3 mm and connected in series. The number of turns comprising an elemental coil at an angle ϕ is proportional to sinϕ in one coil and proportional to cosϕ in the other. It was experimentally shown that a circular crack 4 mm in diameter in a ferromagnetic pipe with diameter 21.6 mm was localized. The proposed structure is suitable for narrow pipe inspection where the sensitivity of our previous sensor composed of two coaxial crescentshaped coils decreases.  [Show abstract] [Hide abstract]
ABSTRACT: Euler's equation provides us with a system of linear equations for localizing a magnetic dipole from measurements of the magnetic field and its gradients. However, so far, the condition for the coefficient matrix of the linear equations to be singular has not been shown. In this paper, we show that the matrix is singular if and only if the dipole moment is perpendicular to the dipole position vector, where the observation point is set at the origin. Moreover, we show that, even in this case, the true position can be uniquely reconstructed by using the MoorePenrose generalized inverse of the gradient tensor.  [Show abstract] [Hide abstract]
ABSTRACT: This paper presents an algebraic reconstruction method for dipolequadrupole sources using magnetoencephalography data. Compared to the conventional methods with the equivalent current dipoles source model, our method can more accurately reconstruct two close, oppositely directed sources. Numerical simulations show that two sources on both sides of the longitudinal fissure of cerebrum are stably estimated. The method is verified using a quadrupolar source phantom, which is composed of two isoscelestrianglecoils with parallel bases. 
Article: Partial differential equationbased localization of a monopole source from a circular array
[Show abstract] [Hide abstract]
ABSTRACT: Wave source localization from a sensor array has long been the most active research topics in both theory and application. In this paper, an explicit and timedomain inversion method for the direction and distance of a monopole source from a circular array is proposed. The approach is based on a mathematical technique, the weighted integral method, for signal/source parameter estimation. It begins with an exact form of the sourceconstraint partial differential equation that describes the unilateral propagation of wideband waves from a single source, and leads to exact algebraic equations that include circular Fourier coefficients (phase mode measurements) as their coefficients. From them, nearly closedform, singleshot and multishot algorithms are obtained that is suitable for use with bandpass/differential filter banks. Numerical evaluation and several experimental results obtained using a 16element circular microphone array are presented to verify the validity of the proposed method.  [Show abstract] [Hide abstract]
ABSTRACT: This paper presents an algebraic method for an inverse source problem for the Poisson equation where the source consists of dipoles and quadrupoles. This source model is significant in the magnetoencephalography inverse problem. The proposed method identifies the source parameters directly and algebraically using data without requiring an initial parameter estimate or iterative computation of the forward solution. The obtained parameters could be used for the initial solution in an optimizationbased algorithm for further refinement.  [Show abstract] [Hide abstract]
ABSTRACT: Euler's equation provides us with a system of linear equations for localizing a magnetic dipole from measurements of the magnetic field and its gradients. However so far, the condition for its coefficient matrix to be singular has not been shown. In this paper, we show that it is singular if and only if the dipole moment is perpendicular to the dipole position vector viewing from the observation point. Moreover we show that even in this case the true position can be reconstructed by the MoorePenrose generalized inverse.  [Show abstract] [Hide abstract]
ABSTRACT: In this paper, we investigate an inverse source problem of the time harmonic Maxwell equations at a fixed frequency, where the source consists of multiple point dipoles. An algebraic algorithm is proposed to identify the number, locations and moments of the dipoles from boundary measurements of tangential components of the electric and magnetic fields. Also, a Hölder stability result is shown. The proposed algorithm is numerically verified.  [Show abstract] [Hide abstract]
ABSTRACT: This paper presents a novel algorithm and sensor for estimating the position of a magnetic dipole. By transforming Euler's equation of degree 3 into an integral form, we have linear equations relating the dipole position to the surface integrals of the magnetic flux densities on a cube. To measure all the quantities required in the linear equations, we develop a cubic sensor with a side length of 50 mm which consists of 18 coils. We show that the coefficient matrix of the linear equations is symmetric and traceless, which can be used to improve localization accuracy. By performing the nonlinear least squares method with the initial solution given by the proposed method, the average and maximum error are 8.3 and 17.5 mm, respectively, in the range of 500 mm.  [Show abstract] [Hide abstract]
ABSTRACT: This paper presents a method for identifying the generalorder poles of a meromorphic function algebraically from its values on the unit circle, which has various applications in inverse source problems in potential analysis. First, we derive a system of Dthdegree equations for N distinct poles zn of order Dn, where n = 1, 2, ..., N and D = max 1 ≤ n ≤ N{Dn}. Then, we transform these equations into linear equations for the coefficients of the Nthdegree equation whose roots are zn so that the poles are obtained algebraically from data. The obtained poles can be used as an initial solution for iterative algorithms. A method for estimating the order Dn of each pole is also proposed and is numerically verified.  [Show abstract] [Hide abstract]
ABSTRACT: This paper presents a method for estimating the position of a radiofrequency identification (RFID) tag in twodimensional (2D) space. A lowfrequency (135 kHz), passive RFID tag with a loop antenna transmits its ID number to the tag reader by electromagnetic induction. This study reveals that the radial distance and the azimuth of the RFID tag in a 2D plane can be determined from the absolute value and phase of the firstorder Fourier coefficient of the magnetic flux density generated by the tag. The authors develop a sensor unit that consists of four coils: two special coils that generate electromotive forces that are proportional to the firstorder Fourier cosine and sine coefficients and two conventional loop coils. Using the developed sensor, a 29mmdiameter diskshaped RFID tag can be localized with a maximum (average) error of 18 mm (5 mm) within a circular domain of radius 140 mm. 

Article: An inverse source problem for Helmholtz's equation from the Cauchy data with a single wave number
[Show abstract] [Hide abstract]
ABSTRACT: The inverse source problem of the Helmholtz equation in an interior domain is investigated. We show the uniqueness and local stability, where the source consists of multiple point sources. An algebraic algorithm is proposed to identify the number, locations and intensities of the point sources from boundary measurements. Uniqueness and nonuniqueness results for some distributed sources are also established. The proposed method is verified numerically.  [Show abstract] [Hide abstract]
ABSTRACT: A simple sensor for magnetic flux leakage methods of detecting cracks on the surfaces of ferromagnetic pipes is proposed. We show that the crack position can be determined by the Fourier coefficients of the leakage magnetic flux density on a circle inside the pipe in the pipe crosssectional plane. Coils that directly output the Fourier cosine and sine coefficients were made. The experimental results showed that, using only these two coils rather than dozens of magnetic sensors, the center position of a crack on the inside/outside surface of the pipe could be localized.
Publication Stats
509  Citations  
42.24  Total Impact Points  
Top Journals
Institutions

20082013

Osaka ElectroCommunication University
Edo, Tōkyō, Japan


20002013

The University of Tokyo
 • Department of Information Physics and Computing
 • Department of Interdisciplinary Information Studies
白山, Tōkyō, Japan


20072012

The University of ElectroCommunications
 Department of Mechanical Engineering and Intelligent Systems
Edo, Tōkyō, Japan


20062009

Kyoto Prefectural University of Medicine
 Department of Dermatology
Kioto, Kyōto, Japan


20022006

Saitama University
 Department of Mechanical Engineering
Saitama, Saitamaken, Japan 
National Institute of Informatics
Edo, Tōkyō, Japan
