摘要:

AbstractThis paper discusses the global asymptotic stability and existence of periodic solutions of periodically varying nonlinear compartmental systems with time delay. We exploit the inherent property that any bounded solution approaches a periodic solution. Based on this result, we derive a necessary and sufficient condition for a homogeneous (zero‐input) system to be globally asymptotically stable and derive a necessary and sufficient condition for the system forced by a periodic input to admit a periodic solution. These conditions do not include time delay and are stated in terms of the interconnections among compartments. This amounts to showing that such qualitative properties of solutions are preserved irrespective of the introduction of arbitrary time dela

ISSN:8756-6621    

DOI:10.1002/ecja.4410650102

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

年代:1982

数据来源: WILEY

摘要:

AbstractA general form of sampling theorem is considered in which the signal is uniquely interpolated by sample values taken at a number of uniformly spaced sample points for a band‐limited time function of the form f(t) such that ∫−∞∞|f(t)|2dt<+∞ and whose Fourier spectrum contains components within |ω|<ω0. One of the necessary and sufficient conditions establishing such an interpolation function is derived. According to the theorem, this interpolation function can be given in terms of time functions that correspond to the coefficients obtained by Fourier series expansion with respect to a function of angular frequency and time, called the generating function. Next, for a signal of the same form considering a generalized form of the sampling theorem that deals with nonuniform sampling points, the necessary and sufficient condition for the interpolation function that ordinarily establishes this is shown. In this case it is also shown that the concept of the generating function can be extended. Further, by combining the well‐known Silverman‐Toeplitz theorem for summation of series and the sampling theorem as derived above from the outline of the generating function, several sampling theore

ISSN:8756-6621    

DOI:10.1002/ecja.4410650103

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

年代:1982

数据来源: WILEY

摘要:

AbstractThe minimal realization problem of two‐dimensional transfer functions remains unsolved despite numerous efforts to solve it. Most conventional approaches have taken the form of seeking real number solutions for nonlinear equations that have system coefficients as their unknown variables and are therefore not amenable to realization. This paper presents sufficient conditions by which the possibility of minimal realization can be judged without necessarily solving nonlinear equations from the coefficients of optimally given two‐dimensional transfer functions. The sufficient conditions presented here are broad ones that include the well‐known types of separable‐denominator and separable‐numerator. We also introduce the canonical forms of state space representation for the separable denominator type and separable‐numerator type, which are very useful in practical applications, and additionally show that there are cases where realization dimensions can be reduced even for the general transfer function, which is difficult in finding the minimal

ISSN:8756-6621    

DOI:10.1002/ecja.4410650104

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

年代:1982

数据来源: WILEY

摘要:

AbstractA well‐known analytical method for linear systems is that which employs the Laplace transformation. Such method is applicable to analytical nonlinear systems in various ways, as has already been reported [1, 2]. A problem in this method is calculation of the multidimensional inverse transformation, which is normally performed after reduction to one dimension. However, if higher‐order terms have to be taken into account, the amount of calculation for the reduction becomes excessive, so that manual calculation is impossible. Thus, in this report we propose to achieve the reduction operation by computer. Algorithms derived are practical for application to systems that consist of the sum and product of linear subsystems, cascade connection, and feedback. These are quite efficient algorithms, fully utilizing the system characteristics. We show that the advantage of the algorithm is to have the output in the form of frequency compone

ISSN:8756-6621    

DOI:10.1002/ecja.4410650105

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

年代:1982

数据来源: WILEY

摘要:

AbstractSurface acoustic wave parametric oscillation due to the backward nonlinear process in a surface acoustic wave convolver is an interesting phenomenon and various applications have been considered. This paper discusses the theory and experimental results with a SAW reflector design with a low oscillation threshold, using a Si/LiNbO3substrate, which yields the highest convolution efficiency. The results show that the oscillation threshold power can be greatly reduced by using reflectors (threshold power is reduced to 1/30 for a reflection coefficient of 0.75, as compared with the reflectionless case). In the experiment oscillation could be achieved with such low input power that the oscillation threshold was only 14.4 dBm, and thus SAW parametric oscillation is possible. Also, the experimental oscillation characteristics agree well with theory.

ISSN:8756-6621    

DOI:10.1002/ecja.4410650106

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

年代:1982

数据来源: WILEY

摘要:

AbstractThis paper introduces a computer analytical technique for obtaining the element sensitivities of switched‐capacitor networks (SC networks) with good efficiency by using adjoint networks similar to those for conventional linear networks. The SC networks treated here consist of capacitors, ideal two‐mode switches and operational amplifiers represented by nullator‐norator models. The network equations for the SC networks with amount of charge movement and voltage as variables are so formulated that the network equations of the adjoint networks can be obtained by a simple transformation such as the transpose of their coefficient matrices and with minimum order in many cases. It is also shown that by using Tellegen's theorem in the SC networks, the element sensitivities for capacitors can be obtained as the product of the amount of charge movement in the original SC networks and the voltages in their adjoint networks. Therefore, the element sensitivities can be determined by network analysis of the SC networks and their adjoint networks and the analysis can be performed efficiently with the abovementioned formulation of network equations. Finally, an example of sensitivity analysis by implementing a computer program is shown and its effectiveness is conf

ISSN:8756-6621    

DOI:10.1002/ecja.4410650107

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

年代:1982

数据来源: WILEY

摘要:

AbstractThis paper describes the optical cord and stranded cable used in cable transmission systems, stations or relays. The allowable stress, bending radius and lateral load of the fiber core are explained first, and then component designs are presented that ensure these characteristics. Tensile strength requirements on the optical cable necessitate using a cord of more than 5,500 denier and a central steel wire with 1‐mm radius. On the basis of the deformation of the PVC sheath due to lateral load, the standard outer diameter of the cable is determined to be 3 mm. The buckling phenomenon of nylon pipe is discussed as it applies to a tube‐type cord, and it is shown that the Young's modulus of the pipe must be below 70 Kg/mm2. The transmission properties, strain, tensile and bending properties of cords and cables, fabricated on the basis of the above‐described design satisfy the desired objectives, confirming the effectiveness of the d

ISSN:8756-6621    

DOI:10.1002/ecja.4410650108

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

年代:1982

数据来源: WILEY

摘要:

AbstractPel Adaptive DPCM Coding is an encoding method in which an appropriate DPCM coding loop is selected for every picture element. This is done by referring to only the past decoded values in the previously available multiple coding loops. Since each DPCM loop has been determined experimentally, it has been difficult to achieve efficient encoding, and the location of its compression limit remains unclear. In this paper, we have modified the previously proposed Block Adaptive DPCM coding method based on image partition and introduced a Pel adaptive DPCM coding method using multiple DPCM coding loops. In this method, nonhierarchical cluster analysis plays an important role. The coding performances of a Pel adaptive DPCM coding method differ depending on the subject images. The Pel adaptive DPCM designed by the above method yields a maximum performance advantage of 2.9 dB for the Girl and Couple images, 1.5 dB for the Aerial image, and none for the Moon. This is an improvement over the optimally designed block adaptive DPCM coding method proposed previously by the authors.

ISSN:8756-6621    

DOI:10.1002/ecja.4410650109

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

年代:1982

数据来源: WILEY

摘要:

AbstractThe constant impedance characteristics of self‐complementary antennas are theoretically established; however, the wideband nature of their radiation patterns is not in general assured. A number of works have been published in which widebanding of radiation patterns is attempted by creating axially symmetric self‐complementary monopole‐notch array antennas on the edge of a conductor. However, for the most part these studies are based on experimental procedures. In this paper the radiation characteristics of the self‐complementary monopole‐notch antenna are analyzed theoretically and the array structure with wideband constant gain characteristics is discussed. A monopole‐notch antenna created on the edge of an ideal semiinfinite conducting plane is first analyzed and found to possess wideband radiation characteristics. A more practical monopolenotch antenna created on a finite conducting plate is then analyzed by the induced e.m.f. method. The results show that the gain is higher for the structure with a longer edge in the main beam direction. It is also found that an antenna with constant impedance and constant gain characteristic can be obtained by setting up a monopole‐notch array on a triangular conducting plate. The validity of the theoretical analysis is experimentally confirmed and the limitation of the induced e.m.f. method

ISSN:8756-6621    

DOI:10.1002/ecja.4410650110

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

年代:1982

数据来源: WILEY

摘要:

AbstractIn the design of monolithic filters using multiple energy‐trapped resonators on a piezoelectric plate, the conventional method uses operating parameter theory in lumped‐element equivalent circuits. This method cannot, however, be used for designing wideband monolithic filters employing high‐coupling piezoelectric plates such as piezo‐electromagnetic material and LiTaO3because the error introduced by the lumpedelement equivalent circuit is large. Therefore, this paper proposes a new method using the distributed‐element equivalent circuit, which is suitable for image parameter design theory. First, it is pointed out that in the design of monolithic filters based on the image parameter method which is adopted for designing multimode filters, the resonant and antiresonant arrangement method is difficult to apply to high‐coupling piezoelectric plates. Next, we point out that in order to deal with this problem a new resonant and antiresonant frequency arrangement method is suitable for high‐coupling plate filters and thus good filter properties ca

ISSN:8756-6621    

DOI:10.1002/ecja.4410650111

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

年代:1982

数据来源: WILEY