摘要:

AbstractIn a previous paper we presented procedures of analyzing causality between a time series based on information theory and information flow between time series based on a concept of directed transinformation, which is an information quantity with a direction. In principle, the directed transinformation is defined based on joint probability density between time series. As the length of the time series becomes larger, however, the volume of computation needed also increases exponentially. In this paper, we will discuss efficient computation algorithms. First, we assume Gaussian properties with the time series to present the “correlation function method” as a means of computing of the directed transinformation from correlation function. Second, we present a “linear production model method” in which the information quantity is computed by means of impulse response of a linear production model of the time series. Also, we proved that the linear production model can easily be composed by auto‐regression if the time series is steady. Finally, we discussed the validity of time‐series analysis based on the directed transinformation through analysis of an artificially produced simulated series. We have concluded that the proposed computation method can provide correct analysis even if the causality of the time series can not be analyzed correctly based on the correlati

ISSN:8756-6621    

DOI:10.1002/ecja.4400670602

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1984

数据来源: WILEY

摘要:

AbstractIn the discussion of the two‐class feature extraction problem in pattern recognition, Malina [7] proposed recently a new criterion for overcoming the defects of Fisher's criterion. the discussion in this paper focuses on the effectiveness of Malina's criterion and at the same time points out the problem in his feature extraction method, indicating how to avoid the difficulties. First, no clear definition is made in Malina's method for the coordinate system in the feature space, and the coordinate system used is not optimal. Second, the eigenvalue problem must be solved including the solution of inverse matrix in order to determine the feature axis. For the first problem, this paper points out tha the coordinate axis diand djin the coordinate system in Malina's method satisfies diTdj= 0, where is the sum of covariance matrices in the class. Then the optimal coordinate system from that viewpoint is given. For the second problem, the inverse matrix is eliminated by performing a certain transformation (A‐1transformation) to the coordinate system of the original pattern space. the validity of A‐1transformation is shown by the invariance of the distances among classes, and it is shown that the above optimal coordinate system under A‐1transformation is the optimal orthogonal coordinate system satisfying diTdj= 0, i j. Finally, Malina's criterion is extended to the multiclass case and the problems are di

ISSN:8756-6621    

DOI:10.1002/ecja.4400670603

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1984

数据来源: WILEY

摘要:

AbstractThis paper assumes a situation where the state of a plant cannot be measured directly, and discusses the problem to construct the internally stable model‐following linear system without using a state observer. the problem consists in constructing a compensator (dynamic compensator plus feedback) for the plant so that the transfer function matrix of the obtained closed‐loop coincided with that of the model, and the closed‐loop system (the model‐following system) is internally stable. This paper shows the necessary and sufficient condition for the problem to be solvable, using a geometric approach, and a construction method is presented for the stable model‐following system without using an observer. In contrast to the existing methods, the construction method proposed by this paper does not require any additional condition, and can construct the stable model‐following system with the given parameters without repeating the trial‐and‐e

ISSN:8756-6621    

DOI:10.1002/ecja.4400670604

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1984

数据来源: WILEY

摘要:

AbstractIn the usual switched‐capacitor filter (SCF), all the switches connected to a node are simultaneously on or off. Thus, every node voltage in the SC circuit changes in synchronization only to in‐phase or out‐phase clock, and has a very simple representation by RC equivalent circuit. This paper derives first the RC equivalent circuits for the basic SC circuits, and presents a method of simulation for active or passive RC circuit using an SC circuit, based on the obtained result. Finally, it is verified by several experiments that the SC circuit constructed by the proposed method operates as the theory pre

ISSN:8756-6621    

DOI:10.1002/ecja.4400670605

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1984

数据来源: WILEY

摘要:

AbstractAfter examining the conventional quasistereophonic system, this paper presents the bandwidth division system and the phase shift system as two new types of quasistereophonic systems. These two new systems are based on the understanding that “if the correlation coefficient of a two‐channel acoustic signal is zero (and independent), the sensation of acoustic spreading is a maximum.” It is shown that the correlation coefficient of the two‐channel acoustic signals obtained by means of these systems approaches 0. In addition, using a computer simulation of the quasi‐stereophonic system of frequency‐division type, psychological tests were conducted. the tests showed that differences in sound image quality were principally due to sound image broadening sensations and distance sensations, and that those characteristics corresponded physically to the absolute value of the correlation coefficient and to the correlation coefficient itself. Further, if the correlation coefficient is zero, the sound image is perceived to be broadened to about the extent of the speaker separation, thus verifying the effectiveness of the new broadening sensation con

ISSN:8756-6621    

DOI:10.1002/ecja.4400670606

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1984

数据来源: WILEY

摘要:

AbstractA shaping filter for a stationary stochastic process can be viewed as the open‐circuit voltage transfer function of a resistance‐terminated reactance 2‐port. Further, Levinson's theory regarding spectral analysis (AR analysis) based on the autoregression model of a stationary stochastic process is equivalent to Ricca's theorem in network theory. Using these facts, from the digital lattice filter which is developed from the AR analysis, we can determine the impulse response of the inverse of the reflection coefficient at the output of the resistance‐terminated reactance 2‐port corresponding to the stationary stochastic process which is analyzed. In this paper we present a technique in which an AR analysis is performed on an autoregressive, moving average process (ARMA process), the impulse response is approximated by a rational function as described above and, by determining the open‐circuit voltage transfer function, the ARMA coefficients are derived. Also, we present two different regression methods for determining the rational function approximations to the impulse response. In general, there is no guarantee that the rational function approximations determined by these methods will be physically realizable transfer functions, but it has been verified by modeling tests that, for an ARMA process of finite degree, by this technique we can determine ARMA coefficients which are better than those obtained by AR analysis to approximate the pow

ISSN:8756-6621    

DOI:10.1002/ecja.4400670607

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1984

数据来源: WILEY

摘要:

AbstractThis paper proposes a method of constructing an RC active network to realize a voltage transfer function matrix without using an integrator. the method is based on the fact that an m‐input system of multivariable controllable canonical form can be decomposed into m subsystems of controllable canonical form using state feedback and input transformation. the required m‐input p‐output voltage transfer function matrix is constructed using m passive RC networks and (m + p) summers. the network has the following structure. the capacitor edge voltage of the passive RC network is taken out as the state variable, and the state differential equation is realized by the feedback through the input‐side summers. the output equation is realized by the feedforward through the output‐side summers. the feature of the proposed method is that it is applicable to any proper rational transfer function matrix of any size, not only of the square form. Another feature is that the transfer function matrix can be realized with a fewer number of active

ISSN:8756-6621    

DOI:10.1002/ecja.4400670608

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1984

数据来源: WILEY

摘要:

AbstractThis paper presents a new method of analysis with numerical results for planewave scattering by a layer with inhomogeneous permittivity μ(x) and permeability (x). the analysis is based on the following idea: (1) we construct a periodic structure, an iteration of the given layer and apply Maxwell's equation to derive a differential equation for this structure; (2) the electromagnetic fields in this periodic structure are expanded into Fourier series, and the obtained characteristic equation is transformed into an eigenvalue equation; (3) the electromagnetic field in the original inhomogeneous layer is expanded by the solutions of the eigenvalue equation, and the reflected and the transmitted waves are computed by using the boundary conditions. We call this method the Fourier series expansion method (FEM). When the differential equation obtained in the first step of FEM contains dϵp(x)/dx (or dμp(x)/dx), Hill's matrix method cannot be applied for the profile for which d2μp(x)/dx2(or d2ϵp(x)/dx2) is discontinuous. By contrast, FEM in this paper can be applied even if ϵp(x) (or μp(x) is continuous. dϵp(x)/dx (or dϵp (x)/dx) is discontinuous. Even if ϵ(x) (or μ(x)) of the original inhomogeneous layer is discontinuous we can always construct a periodic structure with continuous ϵ(x) (or μ(x)), so that FEM can be applied. By numerical analysis, the usefulness of the method is evaluated from the viewpoints of the accuracy of the solution vs. the number of terms in the expansion of the comparison with already reported results. As an example of application, reflectivity is computed for the concrete layer with inhomogeneous permittivity and permeability containing magnet

ISSN:8756-6621    

DOI:10.1002/ecja.4400670609

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1984

数据来源: WILEY

摘要:

AbstractThis paper proposes R(t), an approximation for the auto‐covariance function of virtual waiting time in an M/G/1. Calculation by R(t) is much easier and faster than the exact formula R(t). By the use of this approximation, not only the computation for arbitrary model becomes possible, but also the computation time can be reduced below 10‐3time even for the model which can be calculated by the exact formula, such as M/M/1. Comparisons between R(t) and R(t) for several models show that R(t) approximates R(t) closely enough for practical use. We have applied R(t) to the accuracy estimation of waiting time measurement by test calls (test call method) and obtained several important results for waiting time measurement. They are: i) the accuracy is improved as the interval of test calls is set shorter, but is hardly improved even if the interval is set shorter than ten times the mean waiting time (saturation characteristic); ii) better measurement accuracy is obtained by the test call method if the interval of test calls is properly set than by the real call method; and iii) the coefficient of variation of service time distribution greatly affects the measurement accur

ISSN:8756-6621    

DOI:10.1002/ecja.4400670610

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1984

数据来源: WILEY

摘要:

AbstractA problem inherent in discussing the antenna radiation mechanism is that the electromagnetic energy flux density can not be deterministically defined. the practice has been to interpret the Poynting vector as the flux density vector and to use it to explain the various phenomena. In this paper, as the equivalent of the Poynting vector, the Nukiyama vector derived from the energy theorem is interpreted as the flux density vector. This interpretation is applied to the radiation mechanism of a perfectly conducting antenna, and it is shown that the radiation can be explained as “an exchange of energy between the electromagnetic wave radiated into space and the conductor on which the power is impressed, with the energy exchange taking place as the energy transverses the conductor.” It is also shown that the new concept of locally‐stored energy derived from Nukiyama's vector for use in this exposition is a broad concept which also encompasses the representation of observable stored energy as defined by R

ISSN:8756-6621    

DOI:10.1002/ecja.4400670611

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1984

数据来源: WILEY