Showing papers in "Ibm Journal of Research and Development in 1961"
TL;DR: It is shown that a relaxation process which exhibits a Gaussian distribution in the logarithm of the relaxation times (a "lognormal" distribution) can be specified completely by three parameters: the mean relaxation time (τm), width of the distribution (β), and the magnitude of the Relaxation (δJ).
Abstract: Such phenomena as dielectric, magnetic, and anelastic relaxation are often described in terms of a distribution of relaxation times. It is shown that a relaxation process which exhibits a Gaussian distribution in the logarithm of the relaxation times (a "lognormal" distribution) can be specified completely by three parameters. These are: the mean relaxation time (τm) the, width of the distribution (β), and the magnitude of the relaxation (δJ). The relationships of these parameters to experimentally measurable functions are usually complicated. These relationships were obtained in numerical form by machine computation. Finally, a simple formula is derived which expresses the parameter β in terms of the widths of the distribution of the activation energies and that of the attempt frequencies.
174 citations
125 citations
TL;DR: Theories of several elastic effects resulting from the contribution of electrons to the strain energy function of germanium are presented and it is found that the elastic properties are appreciably changed by doping with donor and acceptor impurities.
Abstract: Theories of several elastic effects resulting from the contribution of electrons to the strain energy function of germanium are presented. They show that the elastic properties are appreciably changed by doping with donor and acceptor impurities. The properties considered are: (1) volume, (2) elastic constants of degenerate n-type material, (3) third-order elastic constants of degenerate n-type material, (4) elastic constants of degenerate p-type material, (5) elastic constants of material containing electrons bound to donors, (6) elastic constants of material containing electrons bound to pairs of donor atoms. The most striking effect is found for the case of degenerate n-type germanium, in which the theory predicts that c44 can be lowered by 8% with attainable doping levels.
107 citations
TL;DR: The playback process of a magnetic ring head with finite permeability of head and tape are studied, using the theorem of reciprocity, and a simplified gap-loss function is given for the case of finite parameters for tape and head.
Abstract: The playback process of a magnetic ring head with finite permeability of head and tape are studied, using the theorem of reciprocity. In order to obtain an accurate result for the playback process, the field around a magnetic gap is studied by a Fourier method. The shift of the gap null for infinite permeability of the head as calculated by Westmijze is confirmed, and the new shift is found when the tape permeability is greater than one. A simplified gap-loss function is given for the case of finite parameters for tape and head.
97 citations
TL;DR: A method of digital power spectrum estimation involving the direct combination of the sample time function with sines and cosines and the practical design details necessary for the planning of a spectral estimation program are treated.
Abstract: This paper discusses a method of digital power spectrum estimation involving the direct combination of the sample time function with sines and cosines. This treatment is in contrast to the normal indirect digital method which proceeds through the intermediary of the autocovariance function. All the practical design details necessary for the planning of a spectral estimation program are treated.
71 citations
TL;DR: The method for solving the problem of N queues of unrestricted length served in cyclic order by a single server is based on the notion of a "self-consistent" probability distribution and is actually applicable to a general class of multiqueue situations of which the one considered here is typical.
Abstract: The following problem is considered: N queues of unrestricted length are served in cyclic order by a single server. Input to each queue is Poisson, the service time distribution may be arbitrary, and a finite time is required by the server to go from one queue to the next. Supposing that at any queue the server serves all units which he finds when he arrives, what is the probability Pn that in a stationary state he finds exactly n units? The method for solving this problem is based on the notion of a "self-consistent" probability distribution and is actually applicable to a general class of multiqueue situations of which the one considered here is typical.
65 citations
TL;DR: The experimental determination of the basic parameters appearing in the theory is shown to require the application of a number of corrections to the usual type of dynamical data.
Abstract: The present paper deals with the use of the theory and computations given in Part I for a relaxation process governed by a lognormal distribution of relaxation times. The experimental determination of the basic parameters appearing in the theory is shown to require the application of a number of corrections to the usual type of dynamical data. General expressions are derived for these corrections, and the data necessary for their use are presented in graphical form. Specific examples are given of the application of the theory to the analysis of anelastic relaxation phenomena.
56 citations
TL;DR: A mathematical analysis is developed for nonlinear circuits which have at least two stable steady states, and therefore are of interest as computing or memory elements, and results in quantitative restrictions on the parameters.
Abstract: A mathematical analysis is developed for nonlinear circuits which have at least two stable steady states, and therefore are of interest as computing or memory elements. Circuits containing one or two tunnel diodes will be analyzed in detail as applications of the theory. The method is based on the study of certain “potential function” whose extrema are the steady states of the circuit and whose minima correspond to the stable switching states. This study leads to a qualitative description of all solutions in the large and results in quantitative restrictions on the parameters (R, L, C and nonlinear characteristics) which seem of practical importance.
46 citations
TL;DR: A system in which data flows in and out of the memory on the basis of content rather than location (address) is described, using ferrite cores as novel associative memory storage elements.
Abstract: This paper describes a computer storage system in which data flows in and out of the memory on the basis of content rather than location (address). In addition, a small experimental model of this system is described, using ferrite cores as novel associative memory storage elements.
45 citations
TL;DR: Whitfield's generalization of the deformation potential theorem is used to calculate the electron-phonon interaction; his method is, however, extended to include the spin-lattice coupling.
Abstract: Matrix elements are calculated for acoustic-mode scattering of holes in the valence band structure typified by germanium. Whitfield's generalization of the deformation potential theorem is used to calculate the electron-phonon interaction; his method is, however, extended to include the spin-lattice coupling. A general expression for the electron-phonon interaction matrix element is obtained, and calculations are presented for some special directions in k-space.
33 citations
TL;DR: From results, the plastic behavior of germanium can be explained in terms of dislocation velocity and dislocation multiplication, and is analogous to the interpretation given by Johnson and Gilman for deformation of lithium fluoride.
Abstract: Single-crystal specimens of germanium have been tested in tension over a range of temperature and strain rate. Dislocation density has been determined as a function of plastic strain, and some direct measurements of dislocation velocity have been made. From these results the plastic behavior of germanium can be explained in terms of dislocation velocity and dislocation multiplication, and is analogous to the interpretation given by Johnson and Gilman for deformation of lithium fluoride. Effects of crystal purity and orientation have been examined; strain-aging effects were absent. Conjectures are made as to the causes of work hardening.
TL;DR: The problem of the voltage oscillations in a transmission line when a diode represented as a nonlinear capacitance is placed in shunt in that line is considered and the response of this line to periodic driving voltages is considered.
Abstract: In this paper we consider the problem of the voltage oscillations in a transmission line when a diode represented as a nonlinear capacitance is placed in shunt in that line. In particular we consider the response of this line to periodic driving voltages and study the periodic responses. This physical situation is shown to lead to the mathematical model of the wave equation in the voltage in a domain with an internal boundary (the interface), at which the voltage is required to satisfy a nonlinear jump condition. By an application of Gauss' theorem, the problem is reduced to a nonlinear difference-differential equation. In the case that the generator driving the line is matched to it, this family of equations reduces to a family of nonlinear differential equations. The paper is concerned with a study of the periodic solutions of these two classes of equations
TL;DR: The design and performance of a newly developed magnetic core memory is described, which features a unique combination of a current-steering diode matrix and a load-sharing magnetic switch for an economical and high-performance drive system.
Abstract: The design and performance of a newly developed magnetic core memory is described A two-dimensional array organization and partial switching of toroidal cores were employed in the design of this low-power, high-speed memory The memory features a unique combination of a current-steering diode matrix and a load-sharing magnetic switch for an economical and high-performance drive system The operating memory has a storage capacity of 73,728 bits and executes instructions reliably up to a repetition rate of 147 mc The discussion will include a description of the organization, the series-parallel delay line clock, the control of critical timing pulses, and the actual measured performance
TL;DR: A table look-up machine based upon a photographic memory is being optimized for the processing of natural languages that makes use of automatic retrieval of lexical information by means of novel addressing features which allow look- up of phrases regardless of length.
Abstract: A table look-up machine based upon a photographic memory is being optimized for the processing of natural languages. It makes use of automatic retrieval of lexical information by means of novel addressing features which allow look-up of phrases regardless of length. Linguistic determinations made on the basis of the lexical information retrieved govern subsequent operations. In addition a sentence buffer holds a sentence long enough to make the backward and forward passes which will be required to make grammatical and contextual analyses.
TL;DR: A method of addressing memories is described which is very powerful in the processing of natural languages, where the arithmetic or logical operations are either nonexistent or do not lend themselves to algorithmic description.
Abstract: A method of addressing memories is described which is very powerful in the processing of natural languages, where the arithmetic or logical operations are either nonexistent or do not lend themselves to algorithmic description. The main feature is the guarantee of initiation of an exhaustive search for a linguistic word at a point just beyond the desired address. Sequential search backwards not only locates an address if it is there but also provides identification of a longest match first. The method is further extended to provide "conditional" addressing by prefixing subsequent addresses from information obtained in earlier searches.
TL;DR: A phase reversal data transmission system is described, capable of operating over private telephone lines and over switched networks at 1200 bauds, and the clocking problem in data transmission is considered and several approaches are indicated.
Abstract: A phase reversal data transmission system is described, capable of operating at 2000 bauds over private telephone lines and at 1200 bauds over switched networks. The advantages and limitations of the system are discussed from a theoretical and practical standpoint. The clocking problem in data transmission is considered and several approaches are indicated. A summary is given of extensive field tests over switched and private lines in Europe and on private (SAGE) lines in the U. S.
TL;DR: Throughout this investigation radioisotopes were used to identify the comnounds studied and the use of alkanol amines as solder fluxes has been investigated.
Abstract: When rosin flux activated with amine halides is used in soldering printed circuit boards, metal halides are formed; these halides cannot be removed with organic solvents and thus may cause corrosion. The use of alkanol amines as solder fluxes has been investigated. After soldering, the flux can be removed by rinsing with water. Throughout this investigation radioisotopes were used to identify the comnounds studied.
TL;DR: A mathematical analysis of the dynamic behavior of subharmonic oscillators (parametrons) is made assuming a nonlinear reactance but a linear dissipation or resistance.
Abstract: A mathematical analysis of the dynamic behavior of subharmonic oscillators (parametrons) is made assuming a nonlinear reactance but a linear dissipation or resistance. Simple equations of motion for the subharmonic and pump amplitudes are derived in the quasistatic, or high-Q, approximation. Numerical solutions are obtained for two cases. The first shows the subharmonic amplitude changing from a small value to its steady state value when a constant pump signal is applied. The second shows decay when the pump signal is removed.
TL;DR: It is shown how, for a given number of bits per character and a given minimum distance, the probability of undetected error in an asymmetric channel may be reduced by many orders of magnitude merely by the proper selection of coded characters.
Abstract: The choice of a code for a given application is influenced by many factors, such as economics, compatibility, and reliability. This paper is concerned solely with the reliability of codes, and shows how, for a given number of bits per character and a given minimum distance, the probability of undetected error in an asymmetric channel may be reduced by many orders of magnitude merely by the proper selection of coded characters. For a given minimum distance, an optimum selection of characters requires, as nearly as possible, the same number of "one" and "zero" bit failures to change one character to another. The concept of polarized distance is introduced, and it is shown how the probability of undetected error is related to the minimum distance of a code only in a symmetric channel, while the probability of undetected error is related to the minimum polarized distance in both symmetric and asymmetric channels.
The purpose of this paper is to present new theoretical concepts useful in the evaluation of codes, and not to recommend one code over another. The codes in the paper are used only as examples to illustrate the theoretical concepts involved.
TL;DR: The feasibility in analyzing general queuing problems in this manner is stated, together with the results of a Monte Carlo simulation used for comparison purposes.
Abstract: A model which describes a basic junction, or queuing structure, arising in a general computing system is subjected to a mathematical analysis. The results consist of several formulas describing the performance of various parts of the system. The feasibility in analyzing general queuing problems in this manner is stated, together with the results of a Monte Carlo simulation used for comparison purposes.
TL;DR: It is reasoned that the large photovoltages observed by Cheroff and Keller on ZnS crystals are theoretically plausible and can result from a wide class of different mechanisms.
Abstract: It is reasoned that the large photovoltages observed by Cheroff and Keller on ZnS crystals are theoretically plausible and can result from a wide class of different mechanisms. Conversely, the conditions under which a photoconductor with periodic inhomogeneities does not show a cumulative photovoltage are shown to be very restrictive and improbable. General theorems concerning the magnitude of the photovoltage are proved, and remarks are made on its directionality.
TL;DR: This paper surveys some numerical methods applicable in obtaining the distribution of performance parameters associated with the transient behavior of switching circuits, and estimates are given of the IBM 704 machine time necessary for implementation on a ten-transistor switching circuit.
Abstract: This paper surveys some numerical methods applicable in obtaining the distribution of performance parameters associated with the transient behavior of switching circuits. The methods considered are: (1) Monte Carlo, in which sample circuits are simulated on the IBM 704 and their performance computed and tabulated. (2) Parameter sensitivity methods, including propagation of error, in which the first and second moments of the output distribution are estimated from parameter sensitivities; and a method using the total differential of the performance parameter to estimate the deviation of circuit behavior from its component deviations. (3) A surface fifting method, in which there is developed a formula for delay time in terms of circuit component values. Each method is used to determine the delay time of a simple system. Their merits and drawbacks are compared and discussed, and estimates are given of the IBM 704 machine time necessary for implementation on a ten-transistor switching circuit.
TL;DR: An analysis of the system and its application to a paper form feeder carriage in a printer, and the effect of system parameters on system acceleration, steady state volumetric efficiency, and peak pressure are shown.
Abstract: An analysis of the system is discussed and its application to a paper form feeder carriage in a printer is described. A nonlinear differential equation is derived based on stated assumptions and lumped oil parameters. The equation is linearized and solutions for motor velocity and peak pressure, both versus time, are determined. The analytical results show the effect of system parameters on system acceleration, steady state volumetric efficiency, and peak pressure. A limited amount of experimental data has been collected and good correlation between theoretical prediction and experimental results for acceleration time and peak pressure has been achieved.