Showing papers in "Bell System Technical Journal in 1956"
TL;DR: The maximum exponential rate of growth of the gambler's capital is equal to the rate of transmission of information over the channel, generalized to include the case of arbitrary odds.
Abstract: If the input symbols to a communication channel represent the outcomes of a chance event on which bets are available at odds consistent with their probabilities (i.e., “fair” odds), a gambler can use the knowledge given him by the received symbols to cause his money to grow exponentially. The maximum exponential rate of growth of the gambler's capital is equal to the rate of transmission of information over the channel. This result is generalized to include the case of arbitrary odds. Thus we find a situation in which the transmission rate is significant even though no coding is contemplated. Previously this quantity was given significance only by a theorem of Shannon's which asserted that, with suitable encoding, binary digits could be transmitted over the channel at this rate with an arbitrarily small probability of error.
1,398 citations
TL;DR: A systematic procedure is presented for writing a Boolean function as a minimum sum of products and specific attention is given to terms which can be included in the function solely for the designer's convenience.
Abstract: A systematic procedure is presented for writing a Boolean function as a minimum sum of products This procedure is a simplification and extension of the method presented by W V Quine Specific attention is given to terms which can be included in the function solely for the designer's convenience
1,103 citations
TL;DR: In this article, the properties of electrolyte-semiconductor barriers are described, with emphasis on germanium, and the use of these barriers in localizing electrolytic etching is discussed.
Abstract: Properties of electrolyte-semiconductor barriers are described, with emphasis on germanium. The use of these barriers in localizing electrolytic etching is discussed. Other localization techniques are mentioned. Electrolytes for etching germanium and silicon are given.
1,039 citations
TL;DR: In this article, the authors present a blend of theory and experiment, and describe developments in this field during the past few years, including acid-base neutralization, complex ion formation, and ion pairing.
Abstract: Interactions among defects in germanium and silicon have been investigated. The solid solutions involved bear a strong resemblance to aqueous solutions insofar as they represent media for chemical reactions. Such phenomena as acid-base neutralization, complex ion formation, and ion pairing, all take place. These phenomena, besides being of interest in themselves, are useful in studying the properties of the semiconductors in which they occur. The following article is a blend of theory and experiment, and describes developments in this field during the past few years.
405 citations
TL;DR: In this paper, a theory for permitting direct dialing access to high-delay toll trunk groups is presented, and a method is devised using equivalent random traffic, which has good loss predictive ability under the lost calls cleared assumption, for a diverse field of alternate route trunking arrangements.
Abstract: Present toll trunk traffic engineering practices in the United States are reviewed, and various congestion formulas compared with data obtained on long distance traffic. Customer habits upon meeting busy channels are noted and a theory developed describing the probable result of permitting subscribers to have direct dialing access to high delay toll trunk groups. Continent-wide automatic alternate routing plans are described briefly, in which near no-delay service will permit direct customer dialing. The presence of non-random overflow traffic from high usage groups complicates the estimation of correct quantities of alternate paths. Present methods of solving graded multiple problems are reviewed and found unadaptable to the variety of trunking arrangements occurring in the toll plan. Evidence is given that the principal fluctuation characteristics of overflow-type of non-random traffic are described by their mean and variance. An approximate probability distribution of simultaneous calls for this kind of non-random traffic is developed, and found to agree satisfactorily with theoretical overflow distributions and those seen in traffic simulations. A method is devised using “equivalent random” traffic, which has good loss predictive ability under the “lost calls cleared” assumption, for a diverse field of alternate route trunking arrangements. Loss comparisons are made with traffic simulation results and with observations in exchanges. Working curves are presented by which multi-alternate route trunking systems can be laid out to meet economic and grade of service criteria. Examples of their application are given.
399 citations
TL;DR: A class of binary signaling alphabets called “group alphABets” is described, which are generalizations of Hamming's error correcting codes and possess the following special features: all letters are treated alike in transmission; the encoding is simple to instrument; maximum likelihood detection is relatively simple to Instrument.
Abstract: A class of binary signaling alphabets called “group alphabets” is described. The alphabets are generalizations of Hamming's error correcting codes and possess the following special features: (1) all letters are treated alike in transmission; (2) the encoding is simple to instrument; (3) maximum likelihood detection is relatively simple to instrument; and (4) in certain practical cases there exist no better alphabets. A compilation is given of group alphabets of length equal to or less than 10 binary digits.
187 citations
TL;DR: In this article, a transistor with a maximum alpha of 0.97 and an alpha-cutoff of 120 mc/sec is presented. And the manner in which some of the electrical parameters are determined by the distribution of the doping impurities is discussed.
Abstract: Silicon n-p-n transistors have been made in which the base and emitter regions were produced by diffusing impurities from the vapor phase. Transistors with base layers 3.8 × 10−4-cm thick have been made. The diffusion techniques and the processes for making electrical contact to the structures are described. The electrical characteristics of a transistor with a maximum alpha of 0.97 and an alpha-cutoff of 120 mc/sec are presented. The manner in which some of the electrical parameters are determined by the distribution of the doping impurities is discussed. Design data for the diffused emitter, diffused base structure is calculated and compared with the measured characteristics.
99 citations
TL;DR: In this article, impurity diffusion and alloying have been used to construct p-n-p junction transistors utilizing a diffused surface layer as a base region for high-frequency devices.
Abstract: Techniques of impurity diffusion and alloying have been developed which make possible the construction of p-n-p junction transistors utilizing a diffused surface layer as a base region. An important feature is the high degree of dimensional control obtainable. Diffusion has the advantages of being able to produce uniform large area junctions which may be utilized in high power devices, and very thin surface layers which may be utilized in high-frequency devices. Transistors have been made in germanium which typically have alphas of 0.98 and alpha-cutoff frequencies of 500 mc/s. The fabrication, electrical characterization, and design considerations of these transistors are discussed.
63 citations
TL;DR: In this paper, a theory for the forward current-voltage characteristic of the PIN diffused junction silicon diode was given, which predicts that the device should obey a simple PN diode characteristic until the current density approaches 200 amp/cm2.
Abstract: A theory is given for the forward current-voltage characteristic of the PIN diffused junction silicon diode. The theory predicts that the device should obey a simple PN diode characteristic until the current density approaches 200 amp/cm2. At higher currents an additional potential drop occurs across the middle region proportional to the square root of the current. A moderate amount of recombination in the middle region has little effect on the characteristic. It is shown that the middle region cannot lead to anomalous characteristics at low currents.
62 citations
TL;DR: In this paper, a method for determining whether a Boolean function possesses any group invariance is presented, that is, whether there are any permutations or primings of the independent variables which leave the function unchanged.
Abstract: A method is presented for determining whether a Boolean function possesses any group invariance; that is, whether there are any permutations or primings of the independent variables which leave the function unchanged This method is then extended to the detection of functions which are totally symmetric
60 citations
TL;DR: In this article, a wafer-type silicon point-contact rectifier and holder designed primarily for use as the first detector in millimeter wave receivers are described, and a pilot production group of one hundred wafer rectifier units yielded the following average performance data at a wavelength of 5.4 millimeters.
Abstract: A wafer-type silicon point-contact rectifier and holder designed primarily for use as the first detector in millimeter wave receivers are described. Measurements made on a pilot production group of one hundred wafer rectifier units yielded the following average performance data at a wavelength of 5.4 millimeters: conversion loss, 7.2 db; noise ratio, 2.2; intermediate frequency output impedance 340 ohms. Methods of estimating the values of the circuit parameters of a point-contact rectifier are given in an Appendix.
TL;DR: The zone leveling process has been developed into a simple and effective tool, capable of growing large single crystals having high lattice perfection and containing an essentially uniform distribution of one or more desired impurities as mentioned in this paper.
Abstract: The zone-leveling process has been developed into a simple and effective tool, capable of growing large single crystals having high lattice perfection and containing an essentially uniform distribution of one or more desired impurities. Experimental work with germanium is discussed, and the possibility of broad application of the principles involved is indicated.
TL;DR: In this paper, the nonlinear operating characteristics of a traveling wave tube have been studied using a tube scaled to low frequency and large size, and the most important conclusions are: I. There is an optimum set of parameters (QC = 0.2 and γ r 0= 0.5) giving the greatest efficiency.
Abstract: The non-linear operating characteristics of a traveling wave tube have been studied using a tube scaled to low frequency and large size. Measurements of electron beam velocity and current as a function of RF phase and amplitude show the mechanism of power saturation. The most important conclusions are: I. There is an optimum set of parameters (QC = 0.2 and γ r 0 = 0.5) giving the greatest efficiency. II. There is a best value of the gain parameter “C” which leads to a best efficiency of about 38 per cent. III. A picture of the actual spent beam modulation is now available which shows the factors contributing to traveling wave tube power saturation.
TL;DR: In this article, the theory of Cutler and Hines is extended to permit an analysis of beam-spreading in electron guns of high convergence, and a lens correction for the finite size of the anode aperture is also included.
Abstract: The theory of Cutler and Hines is extended in this paper to permit an analysis of beam-spreading in electron guns of high convergence. A lens correction for the finite size of the anode aperture is also included. The Culter and Hines theory was not applicable to cases where the effects of thermal velocities are large compared with those of space charge and it did not include a lens correction. Gun design charts are presented which include all of these effects. These charts may be conveniently used in choosing design parameters to produce a prescribed beam.
TL;DR: In this article, the non-linear behavior of the traveling-wave amplifier is calculated by numerically integrating the motion of the electrons in the presence of the circuit and the space charge fields.
Abstract: The non-linear behavior of the traveling-wave amplifier is calculated in this paper by numerically integrating the motion of the electrons in the presence of the circuit and the space charge fields. The calculation extends the earlier work by Nordsieck and the small-C theory by Tien, Walker and Wolontis, to include the space charge repulsion between the electrons and the effect of a finite coupling between the circuit and the electron beam. It however differs from Poulter's and Rowe's works in the methods of calculating the space charge and the effect of the backward wave. The numerical work was done using 701-type I.B.M. equipment. Results of calculation covering QC from 0.1 to 0.4, b from 0.46 to 2.56 and k from 1.25 to 2.50, indicate that the saturation efficiency varies between 23 per cent and 37 per cent for C equal to 0.1 and between 33 per cent and 40 per cent for C equal to 0.15. The voltage and the phase of the circuit wave, the velocity spread of the electrons and the fundamental component of the charge-density modulation are either tabulated or presented in curves. A method of calculating the backward wave is provided and its effect fully discussed.
TL;DR: In this paper, a set of charts has been developed from which one can read power gain and input impedance as functions of the load termination in tetrode transistor amplifiers, and a method to select, on simple bases, the kinds of terminations that will be suitable.
Abstract: The design of tetrode transistor amplifiers encounters problems of the type that occurs with other transistor uses. Desired frequency characteristics, limitations of parasitic elements, and other practical considerations impose constraints on the range of terminations that can be employed. With many transistors, one can terminate a transistor so that it will oscillate without external feedback; this oscillation or other exceedingly sensitive terminations must be avoided. The two-port parameters of the transistor in any orientation in which it is to be used constitute the fixed or given information which is the starting point of the amplifier design. Using this starting point, methods are developed by which one can select, on simple bases, the kinds of terminations that will be suitable. To facilitate the design of amplifiers, a set of charts has been developed from which one can read power gain and input impedance as functions of the load termination. Illustrative tetrode amplifiers are described. These include a common base 20-mc video amplifier, a common-emitter 10-mc video amplifier, an IF amplifier centered at 30 mc, and an IF amplifier centered at 70 mc. Predicted and measured gains are compared.
TL;DR: In this article, a traveling-wave amplifier with 30 db gain at 5 watts output in the 5,925-to 6,425-mc common carrier band is described and a description of the tube and detailed performance data are given.
Abstract: This paper discusses a traveling-wave amplifier which gives 30 db of gain at 5 watts output in the 5,925- to 6,425-mc common carrier band. A description of the tube and detailed performance data are given.
Journal Article•
[...]
TL;DR: In this paper, the propagation constants of some of the lossy modes in helix waveguide of zero pitch were calculated numerically, as functions of the jacket parameters and the guide size, in regions where the approximate formulas are no longer valid.
Abstract: Helix waveguide, composed of closely wound turns of insulated copper wire covered with a lossy jacket, shows great promise for use as a communication medium. The properties of this type of waveguide have been investigated using the sheath helix model. Modes whose wall currents follow the highly conducting helix have attenuation constants which are essentially the same as for copper pipe. The other modes have very large attenuation constants which depend upon the helix pitch angle and the electrical properties of the jacket. Approximate formulas are given for the propagation constants of the lossy modes. The circular electric mode important for longdistance communication has low loss for zero-pitch helices. The propagation constants of some of the lossy modes in helix waveguide of zero pitch have been calculated numerically, as functions of the jacket parameters and the guide size, in regions where the approximate formulas are no longer valid. Under certain conditions the attenuation constant of a particular mode may pass through a maximum as the jacket conductivity is varied.
TL;DR: In this paper, a method for the thickness measurement of p-type or n-type surface layers on semiconductors is presented, where the surface is lapped at a small angle in order to expose the p-n junction.
Abstract: A method is given for the thickness measurement of p-type or n-type surface layers on semiconductors. This method requires the use of samples with optically flat and reflecting surfaces. The surface is lapped at a small angle in order to expose the p-n junction. After detecting and marking the p-n junction, the thickness is measured by an interference microscope. Another application of the equipment is the measurement of steps in a surface. The thickness range measurable is from 5 × 10−6 cm to 10−3 cm.
TL;DR: In this article, combined measurements of surface recombination velocity, surface photovoltage, and surface conductance were made on etched germanium surfaces, and the results were interpreted in terms of the surface space-charge region and of the fast surface states.
Abstract: Combined measurements have been made of surface recombination velocity, surface photo-voltage, and the modulation of surface conductance and surface recombination velocity by an external field, on etched germanium surfaces. Two samples, cut from an n-type and a p-type crystal of known body properties, were used, the samples being exposed to the Brattain-Bardeen cycle of gaseous ambients. The results are interpreted in terms of the properties of the surface space-charge region and of the fast surface states. It is found that the surface barrier height, measured with respect to the Fermi level, varies from −0.13 to +0.13 volts, and that the surface recombination velocity varies over about a factor of ten in this range. From the measurements, values are found for the dependence of charge trapped in fast surface states on barrier height and on the steady-state carrier concentration within the semiconductor.
TL;DR: In this article, a mathematical analysis of a heterodyne conversion transducer in which the nonlinear element is made up of a nonlinear resistor and a non linear capacitor in parallel is presented.
Abstract: This paper gives a mathematical analysis of a heterodyne conversion transducer in which the nonlinear element is made up of a nonlinear resistor and a nonlinear capacitor in parallel. Curves are given showing the change in admittance and gain as the characteristics of the nonlinear elements are varied. The case where a conjugate match exists at the terminals is treated. It is shown that when the output frequency is greater than the input frequency, modulators having substantial gain and bandwidth are possible, but when the output frequency is less than the input frequency, the converter loss is greater than unity and is little affected by the nonlinear capacitor. The conditions under which a conjugate match is possible are specified and it is concluded that a nonlinear capacitor alone is the preferred element for modulators and that a nonlinear resistor alone gives the best performance in converters.
TL;DR: In this article, a frequency-modulation radar technique especially suited to measurement of atmospheric attenuation at millimeter wavelengths is described, which employs a single klystron, a single antenna and a set of spaced corner reflectors whose relative reflecting properties are known.
Abstract: A frequency-modulation radar technique especially suited to measurement of atmospheric attenuation at millimeter wavelengths is described. This two-way transmission method employs a single klystron, a single antenna and a set of spaced corner reflectors whose relative reflecting properties are known. Since the method does not depend on measurements of absolute antenna gains and power levels, absorption data can be obtained more readily and with greater accuracy than by the usual one-way transmission methods. Application of the method is demonstrated by measurements in the 5-mm to 6-mm wave band. The results have made it possible to assign an accurate value for the line-breadth constant of oxygen at atmospheric pressure; the constant appropriate to the measurements lies between 600 and 800 MCS per atmosphere.
TL;DR: In this paper, a non-reciprocal ferrite device with reverse to forward loss ratios of about 150 in the region from 5,925 to 6,425 cm/sec was constructed.
Abstract: A nonreciprocal ferrite device (field displacement isolator) has been constructed with reverse to forward loss ratios of about 150 in the region from 5,925 to 6,425 mc/sec. The forward loss is of the order of 0.2 db while the reverse loss is 30 db. These results are obtained by using a single ferrite element, spaced from the sidewall of the guide. The low forward loss suggests the existence of an electric field null at the location of a resistance strip on one face of the ferrite. We discuss the various conditions, derived theoretically, under which the electric field null may be obtained and utilized. Furthermore, a method of scaling is demonstrated which permits ready design to other frequencies.
TL;DR: In this paper, a theory for the evaluation and control of the effect of spurious modes in a multi-mode transmission system under certain conditions has been derived and checked experimentally in the particularly interesting case of a TE 01 transmission system, where mode conversion to TE 2, TE 3, and TE 4 is produced by tapered junctions between two sizes of waveguide.
Abstract: In a multi-mode transmission system the presence of spurious modes which resonate in a closed environment can produce an appreciable loss to the principal mode. The theory for the evaluation and control of this effect under certain conditions has been derived and checked experimentally in the particularly interesting case of a TE 01 transmission system, where mode conversion to TE 02 , TE 03 … is produced by tapered junctions between two sizes of waveguide.
Journal Article•
[...]
TL;DR: In this paper, an analysis of coupled helices is presented, using the transmission line approach and also the field approach, with the objective of providing the tube designer and the microwave circuit engineer with a basis for approximate calculations.
Abstract: An analysis of coupled helices is presented, using the transmission line approach and also the field approach, with the objective of providing the tube designer and the microwave circuit engineer with a basis for approximate calculations. Devices based on the presence of only one mode of propagation are briefly described; and methods for establishing such a mode are given. Devices depending on the simultaneous presence of both modes, that is, depending on the beat wave phenomenon, are described; some experimental results are cited in support of the view that a novel and useful class of coupling elements has been discovered.
TL;DR: In this article, a theoretical treatment of the field effect, surface photovoltage and surface recombination phenomena has been carried out, starting with the Hall-Shockley-Read model and generalizing to the case of a continuous trap distribution.
Abstract: A theoretical treatment of the field effect, surface photo-voltage and surface recombination phenomena has been carried out, starting with the Hall-Shockley-Read model and generalizing to the case of a continuous trap distribution. The theory is applied to the experimental results given in the previous paper. One concludes that the distribution of fast surface states is such that the density is lowest near the centre of the gap, increasing sharply as the accessible limits of surface potential are approached. From the surface photo-voltage measurements one obtains an estimate of 150 for the ratio (σ p /σ n ) of the cross-sections for transitions into a state from the valence and conduction bands, showing that the fast states are largely acceptor-type. On the assumption that surface recombination takes place through the fast states, the cross-sections are found to be: σ p ∼ 6 ×10−1 cm2 and σ n ∼ 4 ×10−17 cm2.
TL;DR: In this paper, the electric field and the hole and electron concentrations are found for reverse biased junctions in which one side is either intrinsic (I) or so weakly doped that the space charge of the carriers cannot be neglected.
Abstract: The electric field and the hole and electron concentrations are found for reverse biased junctions in which one side is either intrinsic (I) or so weakly doped that the space charge of the carriers cannot be neglected The analysis takes account of space charge, drift, diffusion and non linear recombination A number of figures illustrate the penetration of the electric field into a PIN structure with increasing bias for various lengths of the I region For the junction between a highly doped and a weakly doped region, the reverse current increases as the square root of the voltage at high voltages; and the space charge in the weakly doped region approaches a constant value that depends on the fixed charge and the intrinsic carrier concentration The mathematics is greatly simplified by expressing the equations in terms of the electric field and the sum of the hole and electron densities
TL;DR: In this paper, a simple device has been produced for regenerating binary pulses directly at microwave frequencies, and the results indicate that even in the presence of serious noise and bandwidth limitations pulses can be regenerated many times and still show no noticeable deterioration.
Abstract: A simple device has been produced for regenerating binary pulses directly at microwave frequencies. To determine the capabilities of such devices one of them was included in a circulating test loop in which pulse groups were passed through the device a large number of times. Results indicate that even in the presence of serious noise and bandwidth limitations pulses can be regenerated many times and still show no noticeable deterioration. Pictures of circulated pulses are included which illustrate performance of the regenerator.
TL;DR: In this paper, a simple transistorized device has been constructed for amplifying and regenerating binary code signals as they are transmitted over substantial lengths of transmission line, and means are provided whereby the distortion in the output of one repeater due to low frequency cutoff is compensated in the next repeater.
Abstract: A simple transistorized device has been constructed for amplifying and regenerating binary code signals as they are transmitted over substantial lengths of transmission line. By the use of simple circuitry, means are provided whereby the distortion in the output of one repeater due to low frequency cutoff is compensated in the next repeater. Furthermore, the repeater is effectively and simply timed from its own regenerated output. A brief discussion of the theory of the circuit is presented along with measured results and oscillograms showing its performance. The effects of extraneous interference on the production of errors in such a repeater are reported. These results are in substantial agreement with theory.
TL;DR: Diffused p-n junction silicon rectifiers incorporating the feature of conductivity modulation are being developed as mentioned in this paper, which are made by the diffusion of impurities into thin wafers of high-resistivity silicon.
Abstract: Diffused p-n junction silicon rectifiers incorporating the feature of conductivity modulation are being developed. These rectifiers are made by the diffusion of impurities into thin wafers of high-resistivity silicon. Three development models with attractive electrical characteristics are described which have current ratings from 0 to 100 amperes with inverse peak voltages greater than 200 volts. These devices are attractive from an engineering standpoint since their behavior is predictable, one process permits the fabrication of an entire class of rectifiers, and large enough elements can be processed so that power dissipation is limited only by the packaging and mounting of the unit.