Showing papers in "Bell System Technical Journal in 1959"
TL;DR: Upper and lower bounds are found for the error probability in decoding with optimal codes and decoding systems for a continuous channel with an additive gaussian noise and subject to an average power limitation at the transmitter.
Abstract: A study is made of coding and decoding systems for a continuous channel with an additive gaussian noise and subject to an average power limitation at the transmitter. Upper and lower bounds are found for the error probability in decoding with optimal codes and decoding systems. These bounds are close together for signaling rates near channel capacity and also for signaling rates near zero, but diverge between. Curves exhibiting these bounds are given.
836 citations
TL;DR: The paper shows the relationship between switching circuits and binary-decision programs and gives a set of simple rules by which one can transform binary- Decision programs to switching circuits, and shows that binary-Decision programming representation is superior to the usual Boolean representation.
Abstract: A binary-decision program is a program consisting of a string of two-address conditional transfer instructions. The paper shows the relationship between switching circuits and binary-decision programs and gives a set of simple rules by which one can transform binary-decision programs to switching circuits. It then shows that, in regard to the computation of switching functions, binary-decision programming representation is superior to the usual Boolean representation.
773 citations
TL;DR: In this article, the problem of finding the best integer x for the first test and to find a rule for choosing the best subsequent test-groups (which may depend on results already observed), in order to minimize the expected total number of group-tests required to classify each of the N units as good or defective is considered.
Abstract: In group-testing, a set of x units is taken from a total starting set of N units, and the x units (1 ≤ × ≤ N) are tested simultaneously as a group with one of two possible outcomes: either all x units are good or at least one defective unit is present (we don't know how many or which ones). Under this type of testing, the problem is to find the best integer x for the first test and to find a rule for choosing the best subsequent test-groups (which may depend on results already observed), in order to minimize the expected total number of group-tests required to classify each of the N units as good or defective. It is assumed that the N units can be treated like independent binomial chance variables with a common, known probability p of any one being defective; the case of unknown p and several generalizations of the problem are also considered.
359 citations
TL;DR: This paper gives a theoretical treatment of several properties which describe certain variable-length binary encodings of the sort which could be used for the storage or transmission of information, such as the prefix and finite delay properties.
Abstract: This paper gives a theoretical treatment of several properties which describe certain variable-length binary encodings of the sort which could be used for the storage or transmission of information. Some of these, such as the prefix and finite delay properties, deal with the time delay with which circuits can be built to decipher the encodings. The self-synchronizing property deals with the ability of the deciphering circuits to get in phase automatically with the enciphering circuits. Exhaustive encodings have the property that all possible sequences of binary digits can occur as messages. Alphabetical-order encodings are those for which the alphabetical order of the letters is preserved as the numerical order of the binary codes, and would be of possible value for sorting of data or consultation of files or dictionaries. Various theorems are proved about the relationships between these properties, and also about their relationship to the average number of binary digits used to encode each letter of the original message.
358 citations
TL;DR: In this paper, the stability of silicon surfaces when they are provided with a chemically bound solid-solid interface was investigated and the application of the process to devices and resulting device characteristics were discussed.
Abstract: A study has been carried out of the stability of silicon surfaces when they are provided with a chemically bound solid-solid interface. Stable surfaces have been obtained with the system silicon-silicon dioxide when the oxide is thermally grown. This latter system has been studied in some detail. In this paper the following phases of our investigation are presented: (i) some aspects of the thermal oxidation process and properties of the oxide; (ii) the electronic properties of the resulting silicon-silicon dioxide interface; (iii) the application of the process to devices and resulting device characteristics.
215 citations
TL;DR: TASI as mentioned in this paper is a high-speed transmission and switching system that assigns a talker to a channel as soon as he starts to talk and disconnects him when he pauses, if someone else needs the channel.
Abstract: The number of telephone circuits carried by a submarine cable system can he doubled by using the normal gaps in speech to interpolate additional conversations. TASI is a high-speed transmission and switching system that assigns a talker to a channel as soon as he starts to talk and disconnects him when he pauses, if someone else needs the channel. Switching from channel to channel may occur many times during a typical call, but the effect on transmission quality is negligible because the switching time is fast compared with the syllabic rate.
144 citations
[...]
TL;DR: A wealth of devices which depend on the Hall effect for their operation have been proposed in the last decade as mentioned in this paper, and a survey of these devices can be found in this paper.
Abstract: A wealth of devices which depend on the Hall effect for their operation have been proposed in the last decade. This paper gives the results of a survey of these devices. Original work in this field is included in those sections which describe the circulator, one-piece gyrator, switch, frequency spectrum analyzer, phase discriminator and digital-to-analog encoder. Semiconductor materials are discussed in terms of what type of material is most desirable and how currently available materials limit the usefulness of Hall effect devices.
134 citations
TL;DR: In this paper, the authors found that X-irradiation effects a lowering of the amplitude of the 50°K absorption, while a new absorption is introduced at 100°K.
Abstract: Certain imperfections in alpha-quartz are found to introduce anelastic absorption at low temperatures in crystal resonators. For a shear wave of 5 mc, absorption peaks are observed at approximately 20° and 50°K in unirradiated samples, X-irradiation effects a lowering of the amplitude of the 50°K absorption, while a new absorption is introduced at 100°K. The results of acoustic absorption measurements on a number of selected samples of synthetic quartz resonators suggest that, whereas the 50°K absorption is attributable to excess oxygen, the 100°K defect involves aluminum which is present substitutionally in the crystal structure. X-irradiation-induced lowering of the resonant frequency of crystal resonators is clearly the result of a lowering of the elastic modulus associated with the 100°K anelastic absorption.
98 citations
TL;DR: In this paper, a traveling-wave maser consisting of a ruby-loaded comb structure was tested and a gain of 23 db at 6 kmc with a bandwidth of 25 mc was obtained.
Abstract: Broadband very-low-noise microwave amplification can be obtained from solid state maser action in a propagating microwave structure. Such a traveling-wave maser produces unilateral amplification with a high degree of gain stability. The theory of the traveling-wave maser is developed and used to compare the gain, bandwidth and gain stability of the traveling-wave maser with that of the cavity maser. The general requirements for traveling-wave maser slow-wave structures are discussed. Theoretical analysis and experimental results are presented for the comb-in-wareguide slow-wave structure. A traveling-wave maser consisting of a ruby-loaded comb structure was tested. A gain of 23 db at 6 kmc with a bandwidth of 25 mc was obtained. Further performance characteristics of this amplifier and one using gadolinium ethyl sulfate are given. Experimental verification of the low noise temperature of solid state masers was obtained.
84 citations
TL;DR: In this paper, the paramagnetic resonance properties of Cr+++ ions in Al 2 O 3 (ruby) were investigated theoretically and experimentally in order to obtain information necessary for the application of this material as active material in a three-level solid-state maser (3LSSM).
Abstract: The paramagnetic resonance properties of Cr+++ ions in Al 2 O 3 (ruby) were investigated theoretically and experimentally in order to obtain information necessary for the application of this material as active material in a three-level solid-state maser (3LSSM). Numerically computed energy levels, together with their associated eigenvectors, are presented as a function of applied magnetic field for various orientations of the magnetic field with respect to the crystalline symmetry axis. A more detailed discussion is devoted to energy levels, eigenvectors and transition probabilities at angles O°, 54.74° and 90°, where certain simple relations and symmetries hold. Paramagnetic spectra for signal frequencies between 5 and 24 kmc are shown; agreement between computed and measured resonance fields is satisfactory.
79 citations
TL;DR: A class of codes capable of correcting multiple errors is described and the logical design of the encoder and decoder, as well as the, guard space requirement of good digits between bursts of errors, is described.
Abstract: A class of codes capable of correcting multiple errors is described. Some of these codes can be implemented with considerably less hardware than was needed for previous multiple error-correcting codes. A general method is shown for constructing a code of redundancy 1/b that will correct error bursts of Kb or fewer digits (K and b integers). The logical design of the encoder and decoder, as well as the, guard space requirement of good digits between bursts of errors, is described.
TL;DR: A new experimental artificial larynx, which makes use of transistors and miniaturized components to provide a voice for those who have lost the use of their vocal cords by surgical removal or paralysis, is described.
Abstract: A new experimental artificial larynx, which makes use of transistors and miniaturized components to provide a voice for those who have lost the use of their vocal cords by surgical removal or paralysis, is described. The larynx operates by introducing a substitute for the sound of the vocal cords into the pharyngeal cavity by means of a vibrating driver held against the throat. The acoustic principles of normal and artificial speech production that were followed in arriving at the new design are presented, along with descriptions of the transistor circuit and its operating characteristics.
TL;DR: In this paper, an extended series of experiments in beyond-the-horizon propagation on a 171-mile overland path using 460 and 4110 cm was described, where the effect of antenna size on signal level and fading characteristics, wavelength dependence, seasonal and diurnal effects, a new form of diversity reception, and bandwidth capability of the medium were investigated.
Abstract: This paper describes an extended series of experiments in beyond-the-horizon propagation on a 171-mile overland path using 460 and 4110 mc. The following aspects of the propagation were investigated: the effect of antenna size on signal level and fading characteristics, wavelength dependence, seasonal and diurnal effects, a new form of diversity reception, the bandwidth capability of the medium. Many of the experiments were directed toward a better understanding of the mechanism of propagation.
TL;DR: In this article, the authors investigated radio propagation at 35 and 150 cm into large city buildings and found that, on the average, a loss in the order of 20 to 25 db may be encountered on the first floor.
Abstract: Investigations of radio propagation at 35 and 150 mc into large city buildings have disclosed that, on the average, a loss in the order of 20 to 25 db may be encountered on the first floor. This loss, which represents the reduction from the median field in the city streets at the same distance from the transmitter, is known as building loss. Losses were found to be slightly smaller and more uniform at 150 mc than at 35 mc. Losses also were found to be appreciably less on higher floors in a building. Methods of using this information for engineering radio systems to serve people in buildings are described. Some sample problems demonstrate that, with equal receiver performance, the effective coverage range in buildings for a 150-mc system will be greater than that for a 35-mc system.
TL;DR: In this paper, the splitting factors for the Cr+++ ion in emerald were determined at X-band (8.2 to 12.4 kmc), K-band and M-band.
Abstract: Paramagnetic resonance for the Cr+++ ion in emerald has been observed at X-band (8.2 to 12.4 kmc), K-band (18 to 26.5 kmc) and M-band (50 to 75 kmc). From spectra observed at these frequencies, the spectroscopic splitting factors g $$ , g$$ and D have been determined. The large value of D observed suggests the possible use of emerald as an active material in relatively high microwave-frequency solid-state masers.
TL;DR: A method is described for transmitting digitalized video signals to reduce channel capacity front that needed for standard PCM by taking advantage of the inability of the human eye to notice the exact amplitude and shape of short brightness transients.
Abstract: A method is described for transmitting digitalized video signals to reduce channel capacity front that needed for standard PCM. This method takes advantage of the inability of the human eye to notice the exact amplitude and shape of short brightness transients. The transmitted information consists of the amplitudes and times of occurrence of the “edge” points of video signals. These selected samples are coarsely quantized if they belong to high-frequency regions, and the receiver then interpolates straight lines between the samples. The system was simulated on the IBM 704 computer. The processed pictures and obtained channel-capacity savings are presented.
TL;DR: The analysis indicates that, with binary FM and appropriate postdetection low-pass filters, it is possible in principle to realize an improvement in signal-to-noise ratio over bipolar double-sideband AM with synchronous detection (phase reversal), for equal channel bandwidths, average signal power and pulse transmission rates, although this may not be feasible with practicable filters.
Abstract: In binary pulse transmission by carrier amplitude or frequency modulation it is ordinarily desirable, both for efficient bandwidth utilization and for improved performance under adverse noise conditions, to use bandpass channels of the minimum practicable bandwidth, as determined by considerations of intersymbol interference and filter design. It is shown that intersymbol interference can be avoided in binary pulse transmission by FM without the need for a wider channel band than in double-sideband AM, for equal pulse transmission rates. Explicit general expressions are derived for the appropriate shaping of the bandpass channel and for the shapes of received pulses, for cases in which rectangular binary pulses are transmitted by FM, without premodulation or postdetection pulse shaping by low-pass filters. Illustrative comparisons are made of binary pulse transmission by AM and FM for two special cases of general interest in communication theory and pulse-system design. The more general case of partial pulse shaping by premodulation and postdetection low-pass filters is also considered. The performance of FM and AM systems in the presence of noise depends on the division of channel shaping between transmitting and receiving filters. The optimum division with FM and AM is determined for random noise, and comparisons are made of signal-to-noise ratios for optimized FM and AM systems. It is shown that there is a single universal relation between error probability and signal-to-noise ratio, applying to an infinite universe of optimized baseband systems and optimized AM systems with ideal synchronous detection, and that this relation is the same as for baseband transmission over an idealized flat channel of minimum bandwidth. The analysis indicates that, with binary FM and appropriate postdetection low-pass filters, it is possible in principle to realize an improvement in signal-to-noise ratio over bipolar double-sideband AM with synchronous detection (phase reversal), for equal channel bandwidths, average signal power and pulse transmission rates, although this may not be feasible with practicable filters.
TL;DR: In this paper, the Nernst effect and the variation of thermoelectric power with magnetic field for single-crystal samples of n-type germanium of various orientations and impurity concentrations, at fields up to 18,000 gauss and temperatures from 275° to 60°K and below.
Abstract: A study has been made of the Nernst effect and the variation of thermoelectric power with magnetic field for single-crystal samples of n-type germanium of various orientations and impurity concentrations, at fields up to 18,000 gauss and temperatures from 275° to 60°K and below. Except at the highest temperatures, both effects arise predominantly from that part of the thermoelectric power which is due to phonon drag. All observations can be quantitatively accounted for by theory. They yield information about the dependence of the relaxation times for phonon-phonon scattering on the frequency of the phonons, and establish with some certainty the conclusion that four-phonon collisions are much less important than three-phonon collisions in the pertinent range of temperatures and phonon frequencies. A uxiliary investigations have shown that the quantization of electron orbits in a magnetic field has only a small effect on transport properties when the cyclotron level spacing is less than thermal energy. The mean free path of electrons is shown to be energy-independent, as acoustic-scattering theory predicts. The absolute mobility can be predicted to within 10 per cent or better from data on the fractional changes of resistance with stress and magnetic field. A more detailed summary of the conclusions and implications of the present work is given in Section IX.
TL;DR: A basic theorem is derived for RC networks containing active elements that it is shown that no more than one active element, embedded in a passive RC network, is needed to realize any driving-point function.
Abstract: A basic theorem is derived for RC networks containing active elements. It is shown that no more than one active element, embedded in a passive RC network, is needed to realize any driving-point function. Sufficiency of only one active element is shown by developing a synthesis method. A synthesis technique for n-port passive RC networks is developed in order to establish the sufficiency proof of the basic theorem. A more practical method of realizing driving-point functions, using active RC networks, termed the “cascade” method, is also presented. This method is applied to the design of a tenth-order Tchebycheff parameter filter.
TL;DR: In this article, the authors describe some recent work in which single-crystal gallium arsenide, with resistivity adjusted to fit the application, is used for point-contact rectifiers which operate efficiently as frequency converters at frequencies as high as 60 kmc, and for switching diodes which show no minority carrier storage effects for switching time of the order of 10−10 seconds.
Abstract: Gallium arsenide, one of the Group III-V intermetallic compounds, appears to be an excellent semiconductor for use in point-contact devices. This paper describes some recent work in which single-crystal gallium arsenide, with resistivity adjusted to fit the application, is used for point-contact rectifiers which operate efficiently as frequency converters at frequencies as high as 60 kmc, and for switching diodes which show no minority carrier storage effects for switching time of the order of 10−10 seconds. These devices will operate over a considerable range in temperature.
TL;DR: A research model, called ESSEX (Experimental Solid State Exchange), which combines remote line, concentration, time-separation switching and PCM transmission is introduced to demonstrate the concept.
Abstract: A new communication system concept which is an important step toward an all-digital telecommunication plant is discussed. A research model, called ESSEX (Experimental Solid State Exchange), which combines remote line, concentration, time-separation switching and PCM transmission is introduced to demonstrate the concept. The model, which uses solid state devices, works at the speed of a full-size system.
TL;DR: In order to derive rules for engineering radio relay systems at 11 kmc, a one-year experiment was conducted in a region of frequent heavy rainfall to measure the attenuation of paths 27 and 12 miles long.
Abstract: Radio waves at 11 kmc are attenuated by rain In order to derive rules for engineering radio relay systems at 11 kmc, a one-year experiment was conducted in a region of frequent heavy rainfall The attenuation of paths 27 and 12 miles long was measured, together with rainfall at two-mile intervals along the paths The instrumentation and the test results are described, and some implications related to systems engineering are pointed out
TL;DR: In this article, the equilibrium delay distribution for a single-server queueing system with Poisson input, random service and constant holding time was found for various occupancy levels, and these were compared with their queued-service constant-holding-time and random-service exponential-hold-time counterparts.
Abstract: The equilibrium delay distribution is found for a single-server queueing system with Poisson input, random service and constant holding time. Curves are presented for various occupancy levels, and these are compared with their queued-service constant-holding-time and random-service exponential-holding-time counterparts.
TL;DR: The Laddic as mentioned in this paper is a ladder-like structure cut out of a rectangular hysteresis-sloop ferrite, where the sides of the ladder and all of the rungs are equal in minimum cross section so that all possible paths are flux-limited.
Abstract: The Laddic is a ladder-like structure cut out of a rectangular hysteresisloop ferrite. The sides of the ladder and all of the rungs are equal in minimum cross section so that all possible paths are flux-limited. The structure presents a large number of possible flux paths. By controlling the actual switching path through the structure any Boolean function of n variables ran be produced. A number of methods of operation are discussed, and design formulae and experimental results presented. One of the attractive features of this device is that the operating currents are not critical. Therefore, it can be operated at speeds limited essentially only by the current drives available. The output may be taken during the input variable phase or during a subsequent reset phase. Switching speeds of a few tenths of a microsecond and repetition rates of a few hundred kilocycles hare been achieved.
TL;DR: Two theorems are proved that characterize the matrices used to construct systematic error-correcting codes that are identical with Plotkin's bound on the size of any error-Correcting code.
Abstract: Two theorems are proved that characterize the matrices used to construct systematic error-correcting codes. A lower bound on the number of required check bits is derived, and it is shown that, in certain cases, this bound for systematic codes is identical with Plotkin's bound on the size of any error-correcting code. A linear program whose solutions correspond directly to a minimum-redundancy error-correcting code is derived. This linear program can be solved by an algorithm that is essentially the simplex method modified to produce integer solutions. Explicit solutions in closed form that specify the codes directly are derived for the cases when the specified code parameters satisfy certain restrictions. Several theorems are proved about minimum redundancy codes with related parameters.
TL;DR: In this article, the three-level excitation method for solid state masers, including some background material on paramagnetic resonance, is reviewed and the experimental application of the maser material, two cases can be distinguished, In the first case, maser action is based mainly on a favorable relaxation time ratio in signal and idler transitions.
Abstract: The three-level excitation method for solid state masers, including some background material on paramagnetic resonance, is reviewed. With respect to the experimental application of the maser material, two cases can be distinguished, In the first, maser action is based mainly on a favorable relaxation time ratio in signal and idler transitions. It is shown that the relaxation time ratio can be changed artificially by a doping technique. Experimental evidence is presented for two such doping techniques, one self-doping, the other impurity doping. In the second case, maser action is based primarily on a favorable frequency ratio of signal and idler transitions. Maser experiments using this approach are described. In addition, excitation of unidirectional gain and attenuation by circular polarization is discussed. Properties of practical isolator materials are surveyed; they include high-concentration paramagnetic and polycryslolline ferrimagnetic materials.
TL;DR: In this article, the number of trunks found busy by the kth arriving customer is studied as a Markov process imbedded in a (usually) non-Markov process, and a study of bounds for, and approximations to, the probability of loss is made.
Abstract: A group of N trunks serves calls arriving in a renewal process, and lost calls are cleared. The number, N(k), of trunks found busy by the kth arriving customer is studied as a Markov process imbedded in a (usually) non-Markov process N(t), the number of trunks busy at t. Results of C. Palm and F. Pollaczek on the distribution of N(k) are generalized, and a study is made of bounds for, and approximations to, the probability of loss. The probability of loss is studied as a functional of the interarrival distribution function, and certain extremal properties are proven. Formulas for the mean of N(k) and for the covariance function are given, together with equilibrium curves for the probobility of loss, for the mean and variance of N(k), and for the first four values of the covariance function. Some applications to switch counting are discussed.
TL;DR: In this article, a broadband horn reflector antenna has been used for simultaneous radiation and reception of radio signals on different frequencies in the three common-carrier bands in which the Bell System has developed radio relay systems.
Abstract: Development of the broadband horn reflector antenna has permitted the simultaneous radiation and reception of radio signals on different frequencies in the three common-carrier bands in which the Bell System has developed radio relay systems. A necessary adjunct to the antenna is a network to combine or separate the common carrier bands and also to combine or separate the two polarizations of any one band. The particular form of the network that is described was designed to meet strict system requirements on impedance match, insertion loss and cross-coupling between ports.
TL;DR: In this article, the problem of handling errors in testing and measuring is considered and formulas involving the bivariate normal are given for the solution of a problem of dealing with errors in test and measurement.
Abstract: Charts and formulas are presented from which bivariate normal probabilities may be computed. Formulas involving the bivariate normal are given for the solution of a problem of handling errors in testing and measuring. These formulas include, in addition to previously published cases, two new cases. In one, the product is not necessarily centered relative to two-sided specification limits; in the other, one-sided specification limits are considered.
TL;DR: In this article, two common design approaches to a solution of the problem are discussed and Nyquist's criterion of stability and Bode's feedback theory are then used to obtain an engineering evaluation of the relative merits of these two design approaches from a stability standpoint.
Abstract: The major problem in the design of high-frequency transistor amplifiers is the interaction between the output and the input of the amplifier caused by the internal feedback of the transistor. This problem is illustrated and the two common design approaches to a solution of the problem are discussed. Nyquist's criterion of stability and Bode's feedback theory are then used to obtain an engineering evaluation of the relative merits of these two design approaches from a stability standpoint. The positive nature of the internal transistor feedback is established in this stability evaluation. Finally, Bode's feedback theory is used to consider the relative merits of some of the broad banding techniques used in transistor video amplifier design. The over-all analysis shows that many of the most practical and stable linear transistor amplifiers are very simple and can be built with a minimum of design effort.