交大學刊;;Science Bulletin National Chiao-Tung University

Permanent URI for this collectionhdl:11536/137391

Browse

Now showing 1 - 20 of 121

A Dynamic Storage Allocation Algorithm for Multiprocessor Computer Systems
(交大學刊編輯委員會, 1972-01)
鄭武皇
;
楊超植
;
W.H.Cheng
;
C.C.Yang
In this paper a dynamic storage allocation algorithm is developed. The algorithm is very suitable for multiprocessor computer systems under multiprogramming and/or time-sharing environment. It may be considered as a dynamic Bayesian type. Since all the statistical information used in the algorithm is dynamically extracted, prerunning of programs is not required at all. The technique of biasing the working parameters and the concept of dynamic activity are both applied for developing the algorithm. It is shown that the system performance is improved on the average.Although some additional cost is small. In addition the algorithm is simple in implementation.
A FORTRAN Program for Calculating the Mean Square Values of the Electron Momenta in N-Electron Atoms He-Kr
(交大學刊編輯委員會, 1974-03)
李維銘
;
Lee Wei Ming
The one-electron part of the total mean square values of electron momenta in N-elevtron atoms can be expressed as <Ψ|Σ1 to N P|Ψ>. The total wave function Ψis a linear combination of spin-orbitals ø. Ø is expanded in terms of Slater type orbital (STO) . In this paper, a Fortran program was designed to solve this lengthy calculation.
A Generalized Function-Theoretic Technique for the Solution of Some Diffraction Problems
(交大學刊編輯委員會, 1972-01)
郭南宏
;
N.H.Kuo
A straightforward approach to the solution of a general class of Wiener-Hopf equations is developed. The advantage of the present procedure is to apply the topological concept so that a rather difficult step of performing factorization in the usual procedure can be avoided. A new theorem regarding additive decomposition is deduced and verified in some detail; a theorem of approximating a function in a topological space is presented; in addition, an alternative view of Laplace transformation is given. The solutions of two typical diffraction problems are used to demonstrate the method; moreover application of the concept of a generalized operator to simplity problem formulation is well illustrated.
A Generalized Technique in the Solutions of Some Diffraction Problems Concerning Dielectric Wedges
(交大學刊編輯委員會, 1969-04)
郭南宏
;
Nan-Hung Kuo
A modified technique in dealing with the generalized Wiener-Hopf equations is presented. The basic procedure is to apply the topological concept, by which a transform equation is reduced without performing product factorization to a set of matrix equations. Moreover, the theory of distribution is applied to simplify the problem formulation. A new theorem concerning additive decomposition is deduced and used to verify the authenticity of the method are discussed. In this paper an example, diffraction of a plane wave by a right-angled dielectric wedge, is included to illustrate the application of the method.
A Hybrid Method for Solving Parabolic Partial Differential Equations
(交大學刊編輯委員會, 1973-02)
黃瑞龍
;
Lung-Rui Huang
The temperature distribution across an one-dimensional thermally conductive, uniform slab was found at discrete space points using continuous-time, discrete-space(CTDS) technique on the university of Cincinnate AD-4 hybrid computer facilities. In addition, a pure analog computer solution was also investigated, so that a comparison can be made between these two methods.The agreement between the analog and hybrid methods is satisfactorily good.
A New FORTRAN Processing Program-FORTRAN Processor(19000)-for the IBM-1620 Computer
(交大學刊編輯委員會, 1971-04)
蔡中川
;
Jong-Chuang Tsay
The existing FORTRAN processor of IBM 1620 digital computer at National Chiao Tung University uses a two-stage translation process to execute a FORTRAN program. Considerable time and paper tape are wasted in this process. This paper describes an one-stage translation process that avoids these disadvantages. This new translation program called FORTRAN processor (19000) is now used in the computer center of National Chiao Tung University.
A New Light in Nerve Theory
(交大學刊編輯委員會, 1969-04)
魏凌雲
;
Ling Y. Wei
The most recent discoveries of negative fixed surface charge, infra-red emission and birefringence change in nerves strongly suggest the important role of surface dipoles on axon membrane in nerve mechanisms.
A New Model of Negative Photocurrent__Invited Paper
(交大學刊編輯委員會, 1972-01)
薩支唐
;
陳偉文
;
C.T.Sah
;
L.Forbes
;
W.W.Chen
A one-deep-energy- level impurity, space-charge neutral and steady-state model is proposed and nerified in n-type silicon N+IN+ diode samples which are overcompensated but not inverted by gold acceptors in the slightly n-type I-region. The model is as follows. If the photoionization rate of trapped holes at the impurity centeris greater than electrons, then an extrinsic monochromatic light will increase the trapped electron concentration. To maintain quasi-neutrality and a constant steady-state trapped electron concentration (equal to the concentration of the shallow level donor), the injected electron concentration into the compensated n-type I-region must decrease . Thus, the electron or the total conduction current must also decrease since the hole concentration is very low in the presence of a large sweep-out electric field and the hole conduction current may be neglected. This model has successfully predicted the observed large negative photocurrent (NPC)which is as much as 2/3 of the dark current , the large intrinsic optical gain (about 50) and the sublinearity in dark current.
A New Sensitive Method for Measuring Photosensitive Gas Pollutants by Means of Concentration Modulation
(交大學刊編輯委員會, 1973-02)
胡家倫
;
Chia-lun Hu
In order to discriminate thr IR spectroscopic absorption lines of a certain pollutant from the IR absorption lines of the background molecules(H2O,CO2,etc.), it is proposed here that a flash light is applied to the sampled air such that the concentration of the pollutant under measuring is subject to a step change. This step modulation of concentration can be discriminatively detected from the background noise by a differentiator connected to the output of a spectroscopic IR detector. The final output can be calibrated to read directly the concentration of the pollutant. The improvement of signal to noise ratio due to the background molecular noise has been calculated to be six-order increase in magnitude over the conventional IR spectroscopic schemes. The physical reason for this large improvement is due to the fact that the imposed concentration modulation proposed here is spatially coherent, while the fluctuation of background molecular concentrations is not, therefore the"signal" is additive spatislly while the background noise is not. A practical scheme is described and signal to noise ratios due to background molecular noise as well as due to circuit noise are calculated in detail. Research in the related fields has been surveyed and evaluated.
A Note on the Solution of a Difference Equation with Three Indices
(交大學刊編輯委員會, 1974-03)
郭南宏
;
王樂羣
;
Nan-Hung Kuo
;
Leh-Chyun Wang
The generating-function technique for the solution of a linear three-index difference equation with constant coefficients is proposed. In addition, the algorithm for the proposed technique is presented.
A Relay-Tree Method for Simplifying Boolean Functions
(交大學刊編輯委員會, 1975-07)
羅浩榮
;
Hao-Yung Lo
Generally there are two methods for simplifying Boolean functions; one is the Karnaugh Map Method, and the other, the Quine-Mccluskey Tabular Method.
A Self-biased Replacement Algorithm using the Characteristics of Markovean Program Model
(交大學刊編輯委員會, 1974-03)
鄭武皇
;
許濬
;
Wu-Haung Cheng
;
Jiunn Hsu
In this paper a self-biased replacement algorithm is developed. The algorithm is mostly suited for allocating system programs and library programs under multiprogramming and/or time-sharing environment. With biasing, the system performance is improved. It is also proved that the algorithm is optimal on macroscopic viewpoint. Since each reference string behaves a static referencing pattern, so it is feasible that the algorithm makes the optimal choice in selecting the candidate for replacement by using the characteristics of Markovean program model. And since in the present time and in the future time , system programs and library programs constitute dorminant part of software system, so it is worthy to develop a replacement algorithm using the characteristics of Markovean program model, although pre-runnings of programs are required.
A Study on Measures on Nonmeasurable Subsets
(交大學刊編輯委員會, 1974-03)
林妙英
;
Maiow-Ing Lin
A Study on Orbits of Artificial Stellites
(交大學刊編輯委員會, 1973-02)
陳煦
;
Chen,Hsu
Technological developments in the fields of rocket propulsion and guidance have made possible the launching of artificial satellites into one of the many possible space orbits. Once a body has been put into an orbit beyond the influence of the atmosphere, its total energy remains constant and equal to that at rocket burnout. The total energy establishes the class of orbit, which may vary from an ellipse to a hyperbola, the circle and the parabola being limiting cases. However, an infinite number of orbits of the same class are possible for a given total energy at a specified point in space, and the heading angle of the velocity vector at rocket burnout must spacify the particular orbit from the many possible orbits passing through this point.
A Study on Periodic Sequences
(交大學刊編輯委員會, 1972-01)
謝清俊
;
C.C.Hsieh
This paper contains two parts. In one part, the minimum length required to identifying two periodic sequences will be formulated. In the other, two methods for decomposing a periodic sequence into two or more periodic sequences which have periods shorter than that of the original sequence will be presented. The theorems and methods derived in this paper may be applied to many fields, such as coding and decoding implementations,data processing, numerical analysis, data compression,etc.
A Study on the High Energy Ballistics
(交大學刊編輯委員會, 1969-04)
陳煦
;
Hsu Chen
The purpose of this study is twofolds : to present systematically the development of such dynamics of motion as related to the external ballistics of long-range missiles, and to construct a gunnery table for the firing angle, angular range, altitude and flight time for firing velocities approaching the satellite region. The importance of this study is to treat the falling body problem as the motion which relates the transition from bound state to free state , and therefrom to calculate the firing angle and range as well as the time of flight in terms of the firing velocity. The scope of this study is limitted to the discussion under the following assumption, that the burn-out time for energizing the projectile is small compared to flight time, thatt the dissipative effect is ignored , that the spheroidalsphape of the earth and its nonuniform density is neglected, and that the coriolis force is not considered.
A Study on the Range and Accuracy of Long-Rang Ballistic Missiles
(交大學刊編輯委員會, 1971-04)
陳煦
;
Hsu Chen
The flight of a ballistic missile is idealized by the assumptions that the speed is acquired impulsively at sea level and that drag is absent. With practical long-range rockets the distance traversed under power is usually 5-10% of the total range, and the length of re-entry path in which drag is significant is even smaller. Consequently the impulsive assumption gives a good first approximation both for the range and for the impact errors of long-range rockets.
An Application of Vector Space to Sampling Theorem
(交大學刊編輯委員會, 1971-04)
林坤洲
;
K.C.Lin
There are some important concepts in communication theory which are purely of a vector space nature. One of them is the sampling theorem, which says that in a certain class of signals a particular signal is completely determined by its values(samples) at an equally spaced set of times extending forever.
An Extension of A Method For Solving The Equation x=ø(x)
(交大學刊編輯委員會, 1968-10)
張澤仁
;
朱潤祖
;
J.R.Chang
;
J.T.Chu
An Extension of the Dieudonne's Mean Value Theorem
(交大學刊編輯委員會, 1971-04)
鄭希灝
;
Shi-Haw Cheng
In this paper we extend the Dieudonne's mean value theorem to the f(x) which is a mapping of 〔a,b〕into a topological linear space and the Ø(x) which is a continuous realvalued function on 〔a,b〕. One of the sufficient conditions is the existence of the right derivative of f. Particulary, if f(x) is a mapping of 〔a,b〕into〔c,d〕, the sufficient condition can be relaxed to the existence of the Dini derivative.