摘要:

AbstractThe stability of a scalar polynomial can be tested by the Routh‐Hurwitz method, but a stability criterion for polynomial matrices is not known yet. A square polynomial matrix D(s) is said to be (continuous‐time) stable when all the zeros of det D(s) have negative real parts. In this paper, orthogonal polynomial matrices (Rj(s); j = 0, 1,…; associated with an inner product are considered in studying the stability of D(s). A recursive algorithm to produce {Rj(s)} is presented which elucidates the relation between the stability and the polynomial matrix structure of D(s). When D(s) is a scalar polynomial the algorithm is shown to be equivalent to the Routh‐Hurwitz test. The results are applied to the construction of the Schwartz‐form realizations and the Routh approximations of multi‐variable lin

ISSN:8756-6621    

DOI:10.1002/ecja.4410690801

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

年代:1986

数据来源: WILEY

摘要:

AbstractActive R filters consisting of resistors and finite‐GB‐product operational amplifiers have been reported in the literature. Most of them are of second order. Filter functions of higher order can be realized by second‐order blocks connected in cascade. A second‐order highpass filter or a filter with poles can not be cascaded if it is realized with two operational amplifiers since the filter output is not the amplifier output. This paper proposes a new active R filter which realizes an nth order filter function with n operational amplifiers. Lowpass, bandpass, and highpass filters with or without poles can be realized by selecting input and grounded terminals. The proposed method is featured with simple circuitry and ease of design. A fifth‐order lowpass filter will be shown to have theoretical and measured characteristics which are in good

ISSN:8756-6621    

DOI:10.1002/ecja.4410690802

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

年代:1986

数据来源: WILEY

摘要:

AbstractThis paper analyzes the tap weight variance for the FIA‐ZF algorithm and the ADA‐MSE algorithm. The covariance matrix of the FIA algorithms is solved and the features of these algorithms are clarified. Covarianec matrices are also solved statistically for both the FIA Zero Forcing (ZF) algorithm and the FIA Mean Squared Error (MSE) algorithm. Comparisons are made with the simulation result. Finally, tap weight variance is studied for various noise levels and for the case of distorted input sign

ISSN:8756-6621    

DOI:10.1002/ecja.4410690803

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

年代:1986

数据来源: WILEY

摘要:

AbstractIt is known that the sampling theorem applies to the waveform after band‐limited transformation. It frequently happens as in the case of the far visual field of an image where only a finite number of sampling points are available, in the waveform after Hankel transform, for which the spectrum is limited to a finite region. Yen introduced a restoration formula based on the minimum‐norm principle for the finite Fourier transform waveform to restore the original waveform from such a finite number of sampled points. The method was discussed in detail by Kishi and Sakaniwa. This paper applies the idea of Hankel transform and presents several theorems indicating that the interpolation ability of the formula is better than that of the past formula by Dini's expansion. The effectiveness of the method is demonstrated by a numerical example. The proposed restoration formula requires a little more complex numerical calculation but the realized interpolation accuracy is far better both in waveform and in n

ISSN:8756-6621    

DOI:10.1002/ecja.4410690804

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

年代:1986

数据来源: WILEY

摘要:

AbstractThurstone's discriminal process model is geometrized and discussed from the information‐system theoretical point of view. That is, a probability density function (p.d.f.) based on the discriminal processes is defined in the state space (this is the concept of grasping simultaneously the physical stimulus and its response. A parameter space spanned by parameters of the p.d.f. is defined, and fundamental metrics are introduced into the space based on Fisher's information; thus, the space is considered to be a Riemannian space. In the space, the psychophysical functions are derived as certain geometrical invariants. As the results, when the “discriminal dispersion,” i.e., the distribution of the response (the response function) to the input stimulus is to be Gaussian, the power law and the logarithmic law are obtained as psychophysical laws with respect to the mean values of the distributions. From this, relations between difference thresholds, the discriminal process and the psychophysical function are clarified. At the same time, the formalization of the problems and conclusions in [2] are justified from the information‐system theoretical point of view. Further the functional forms of the response functions are also investigated, and it is shown that almost the same conclusions can be obtained even if the Gaussian assumption is repalced by the exponential, the two‐sided exponential, or the logarithmic Gaussian dist

ISSN:8756-6621    

DOI:10.1002/ecja.4410690805

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

年代:1986

数据来源: WILEY

摘要:

AbstractRecently, it has become important, with the development of electroacoustic devices, to provide a systematic analysis of the electrical circuit, the electromechanical transducer part and sound field, taking their interactions into consideration. When the boundary condition of the sound field or the shape of the scattering object is complicated, numerical computation is more effective, especially when the analysis of time variation is required. This paper presents a numerical method of computation for the electrical circuit, transducer part and sound field, in which the entire system is represented by an electrical equivalent circuit. A formulation is made of the wave propagation and circuit element characteristics by Bergeron's method, with the time‐axis analyzed by the node equation. As a result, it became possible to analyze the entire system by electrical circuit analysis, clearly indicating the interactions among various parts. The method is based on the time‐response analysis, and the response to the pulse wave, for example, can be calculated on the time axis. As examples, some basic problems with sinusoidal input in two‐ and three‐dimensional space are analyzed, and the results are compared with those of theoretical analysis. A computation is also made for the case where the voltage and current at the transducer input are affected by the reflected wave of the sound field through the vibrating membrane, indicating the feature of the proposed method where the interactions in the system can clearly be demon

ISSN:8756-6621    

DOI:10.1002/ecja.4410690806

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

年代:1986

数据来源: WILEY

摘要:

AbstractIn the monolithic IC switched capacitor filters (SCF) it is desirable to use as small a member of operational amplifiers (OPA) as possible to save the chip area. This paper presents a circuit configuration of single‐OPA SCF. In the synthesis, a given transfer function is realized by active RC circuits containing ideal phase inversion transformers and then the RC circuits portion is replaced by equivalent SC networks. The characteristics of the realized SCF are sensitive to element variations and its application should be limited to transfer functions of the second to third order with a relatively low Q. High‐order transfer functions may be realized with a relatively small number of OPA's by the cascade realization with this configuration and biquad secti

ISSN:8756-6621    

DOI:10.1002/ecja.4410690807

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

年代:1986

数据来源: WILEY

摘要:

AbstractWhen a channel falls out of synchronism, the approximate frame sync position can normally be estimated and this fact can be applied in the design of frame synchronization systems. This paper presents a new multicounter method for the realization of a fast frame synchronization system taking into account the probability distribution of the sync position. There are two typical ways of realizing a multicounter system: the intensive method, and the division method. The refraining characteristics of a multicounter system are demonstrated for particular probability distributions of the sync position. It is shown that the multi‐counter method provides a flexible synchronization strategy by taking account of the state of the channe

ISSN:8756-6621    

DOI:10.1002/ecja.4410690808

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

年代:1986

数据来源: WILEY

摘要:

AbstractThe reflection coefficient of the guided mode and near‐field radiation pattern in front of the truncated end of a dielectric waveguide with rectangular cross section are analyzed numerically. The procedures are as follows: (1) The electromagnetic field along the dielectric waveguide is exprssed in terms of the guided mode and radiation mode. (2) Two field equations which must be satisfied by the boundary conditions at the end surface of the waveguide are defined. (3) The two‐dimensional Fourier transformation is applied to the field equations to obtain the spectral representations of each mode. (4) By determining the ratio of the electric and magnetic field components under the conditions that the mode spectrum must be satisfied with mutual orthogonality, the reflection coefficients of the guided and radiation modes at the truncated end can be obtained. (5) Finally, applying inverse Fourier transformation to the spectral electric distribution at the guide end, we calculate the near‐field electric distribution pattern in front of the truncated end of the rectangular dielectric waveguide. The analytical results are found to agree with experimental results. Several investigators have analyzed the problem of a dielectric waveguide discontinuity by using the approximated slab waveguide model. However, it is difficult to use the slab model to calculate the effects of transverse displacement of rectangular dielectric waveguides.The results obtained in this paper are very advantageous for analysis of the problem of a waveguide discontinuity displaced in the transverse plane. The analytical method of this paper is applicable to the design of rectangular dielectric waveguide components such as couplers, dividers

ISSN:8756-6621    

DOI:10.1002/ecja.4410690809

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

年代:1986

数据来源: WILEY

摘要:

AbstractWe have been studying the propagation characteristics of the electromagnetic wave at high elevation angle between ETS‐II and Kashima station, aiming at the precise and continuous analysis by computer. We have succeeded in the computer analysis of the propagation characteristics. However, this was time consuming and a real‐time analysis could not be realized. From such a viewpoint, approximate expressions are derived for the size distribution of the precipitation to be calculated from the measured attenuation and the radar echo, to analyze the precise attenuation characteristics of the waves. In the analysis, the measured attenuations of 34.5 and 11.5 GHz, radar echo at 5.33 GHz and the ground temperature are used. The layered precipitation model is employed in the calculation of the rainfall and snowfall in the slant propagation path. The location of the bright band is determined from the ground temperature, decreasing rate of atmospheric temperature, and reflection coefficient of the radar echo. An approximate expression is introduced which determines the size distribution of the precipitation from the radar echo reflection coefficient and the attenuation measured at 34.5 GHz. By comparing the measured values of attenuation at 11.5 GHz with the theoretical ones, it was verified that the estimation is accurate to within 0.3 dB. Thus, a good result was obtained for real‐time precise ana

ISSN:8756-6621    

DOI:10.1002/ecja.4410690810

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

年代:1986

数据来源: WILEY