Search

Consideration is directed toward electromagnetic compatibility as viewed from the control aspect. Emphasis is placed upon an inter-system approach to the improvement of the over-all electromagnetic environment.

The major growth area within the printed circuit market is expected to be as a result of the development of flexible and multi-layer circuits. The continuing pressures on designers of electrical and electronic equipment for smaller and lighter packaging has greatly increased the demand for thin, multi-layer and flexible, wiring packages. In the automotive field, dashboard wiring is rapidly being replaced with flexible circuitry. Computer, appliance and meter manufacturers are among the many circuit users finding benefits in flexible multi-layer circuitry. Mechanical pattern generation methods lend themselves to the production of flexible circuits, where the insulating film to which the pattern is bonded is tough, flexible, and possesses good electrical properties.

Recent years have seen the advent and application of many new and valuable insulating materials. We have seen new applications for older materials and increased emphasis on certain insulating properties such as surface track resistance and operation at higher temperatures. These new materials and new applications should provide the users of high voltage equipment with electrical apparatus which is more reliable and which requires less maintenance. However, this can only be achieved through careful study and control of all of the properties of the insulating material important to the application. For example, care must be exercised to insure that good surface tracking resistance of an insulation is not achieved at the cost of other dielectric properties. A material application, or substitution, without sufficient study of all of the properties can create new problems and in some cases provide a dielectric structure more prone to failure even though "improved" materials are applied. One important property, insulation dissipation factor, (often called power factor), whose neglect may lead to insulation failure in a number of ways, has been under study and control in the application of insulation to switchgear for a number of years. This report will deal with that activity.

The reliability of rotating electrical plant depends greatly on the quality of the insulation system.

This paper will present a technique for simulation of large circuit configurations using circuit level modeling which converts integro-differential equations to simple algebraics. A configuration of 1300 interconnected FET logic circuits has been analyzed using this approach.

This paper will describe a proposed CCD model introducing a nonlinear drift concept which disclosed that the applied voltage waveform to the last transfer electrode had great influence on transfer efficiency. Operation modes and structure will be cited.

This paper will discuss a nonvolatile 2048-bit MNOS block-oriented RAM array with a serial clear/write or read/cycle data rate of over 2 MHz with less than 400-mW power dissipation.

The paper will discuss a low-power, high-performance static RAM using a compact inverted transistor flip-flop memory cell. Small chip size and conventional processing have been found to yield an economical device suited for mainframe storage applications.

This paper examines the use of linearly optimal policies in systems which are in fact nonlinear. Any linear representation of a non-linear system involves some loss of information about the system, and this loss of information in general leads linearly optimal policies to differ from corresponding nonlinearly optimal policies which are truly optimal. This paper indicates the nature of the resulting discrepancies in a stochastic framework and reports on results of several numerical experiments dealing with macroeconomic stabilization policy.

The use of gold in integrated circuit packages has become prohibitive because of cost. Of the current alternatives available, silver plating offers a favorable economic advantage to the MOS-LSI manufacturer. This paper discusses the economics of various gold substitute systems. In addition, the paper presents data which demonstrates the reliability of molded packages using silver plated lead frames. Reliability data on Life, Temperature Cycling, 85°C/85% R.H. and Solderability Testing is presented. Silver plating is an economical and reliable system for use in molded packages for MOS-LSI devices.

In recent years there has been an increasing interest shown by engineers and scientists in problems of ecology and sociology. Many of these problems when viewed as a set of interacting variables take on the characteristics of a large scale system of the man-machine-environment type, and are in many ways similar to those found in the aerospace and process control industries. Because of his special skills in system modeling and optimization, the control systems engineer can make a significant contribution in this area. This paper is directed toward getting control system engineers to exhibit a greater interest in this expanding area. After establishing a reference framework, the paper categorizes problems and gives selective bibliography to assist newcomers in this area. The bibliography is in no way considered exhaustive nor complete, but should serve as an introduction to the problems existing in this area.

A rigorous and exact theory of four-port stripline circulator based on Green's function method is presented in this paper. Three circulation conditions were derived and numerical results are presented which can be used for practical four-port circulator design. Since the solution derived is exact, metal pin tuning is not necessary to fabricate the four-port circulator described in this paper.

This paper will describe a semiconductor display with inherent memory whose central element is a monolithic integrated circuit that performs the logic function in additon to the optical ouput on the display surface. Matrix address is permitted.

In the first part of this paper, an algorithm is derived for testing a given two state machine for the property of being universal. Various characterizations of universality are also obtained, and are stated in the form of necessary and sufficient conditions. The second part of this paper is concerned with the economical realization of sequential machines as networks of identical modules. Bounds are derived for the number of copies of a certain universal modules needed for the realization of an arbitrary machine selected from a given class.

This paper is a continuation and supplement of a previous paper on this same subject,1 which deals with the accelerated life testing of small low voltage electronic power transformers. In this paper are reported the results of life tests which were incomplete at the time of the previous paper. Additional life tests under different conditions of the same insulation system and modified with a different layer insulation are also reported here for the first time. To present an integrated picture, the results previously reported are summarized here together with the additional and new data.

Computers and other data processing equipments usually contain a basic pulse source (or clock) which keeps the various portions of the equipment in synchronization. When the size of the equipment increases significantly, however, or the clock frequency increases, the clock pulse can no longer be assumed to be at all portions of the hardware simultaneously and propagation delays for the clock pulse itself become important. An important problem is thus how to construct a logic net of nontrivial depth using real components (there are no zero delays and all delays can vary with time even if only slightly) into which a second (or-third, etc.), set of waveforms can be introduced before the first set of waveforms has propagated through to the end of the net. It has been shown by R. McNaughton that it is possible to construct such a net of unlimited depth without danger of one set of waveforms being destroyed by another set catching up with it, if the net contains signal feedback in which each logical element tells the preceding element or elements when it is ready to receive another waveform. The present paper proves that such a net cannot be achieved without such feedback. It is easy to show that the net of logic elements where each element merely delays the signal by a fixed amount (with some uncertainty in duration of the delay) can get into trouble, as one signal can catch up to another. However, if the element can contain memory elements (flipflops) to remember the relative timings of previous signals or if the element can buffer some number of successive input signals and be affected in its operation by their relative timing before passing on the first signal, or if the element can pass on control data to following (not preceding) elements, then the problem of showing a failure is more complicated. The present paper first defines rigorously the possible bounds on a physically-realizable element without feedback from following elements, and then shows that an arbitrarily, large chain. of such elements cannot be constructed to pass signals without a possible error.

This paper is concerned with the problem of detecting when two asynchronous control systems are equivalent. The systems investigated in the paper are first represented by means of a formal model called an asynchronous control structure (ACS). This model specifies the constraints imposed on the generation of control signals by a system by means of a simple graphical model called a marked graph. Behavioral equivalence is then characterized in terms of the set of all possible sequences of control signals that can be generated by the system. These sequences are represented by means of another (infinite) marked graph, called a behavior graph. Finally, it is shown that two control systems are equivalent if and only if their behavior graph representations have identical (finite) generating sets.

This paper consists of two parts. Part I, by Giguere and Jamison, discusses transistor circuits capable of regenerating 6.25-m?sec pulses at a 160-mc bit rate. Part II, by Noll, discusses techniques for multiplexing 16 digital signals with a 10-mc clock rate into a single signal with a 160-mc clock rate. Two methods of performing the regeneration function are presented. One method consists of dc level restoration for recognition of the signal and constant current coincident circuitry for the reconstruction of the pulses. The second method consists of operating on changes in the signal for pulse recognition and the use of a bistable circuit for pulse reconstruction. Timing in the second case is obtained by a constant current coincidence gate. Parallel-to-serial multiplexing techniques have been developed to combine sixteen parallel 10-mc clock-rate signals into a 160-mc clock-rate pulse train. The sixteen synchronous signals are applied to sixteen AND gates along with a 10-mc narrow gate pulse. The space separations of the resulting regenerated and timed AND gate output pulses is converted to time separation with only a small amount of signal loss. This is done by injecting the pulses at sixteen equally separated points on a broad-band delay line. Methods have been developed to reduce spurious responses resulting from multiple reflections on the delay line. The current mode transistor AND gates are suitable for AND/OR functions for individual 4-m?sec logic. The multiplexer may also be used as a piece of test equipment to generate repetitive 16-bit binary.

This paper considers the model-referenced adaptive control problem. An adaption technique that is extremely simple to implement is derived analytically. The simplicity of this technique gives it a distinct advantage over other techniques that have been described in the literature and makes it attractive for practical applications. A direct approach to the problem is taken, employing the state-space point of view. By solving the differential equations of the reference model and the adaptive control system, an expression is obtained showing the explicit functional dependence of the performance error on the adaptive parameters. Manipulation of the expression for performance error yields the adaption equations which are subsequently shown to be very simple to implement. To illustrate the theory developed in this paper, a simple example is discussed and a stability analysis employing Lyapunov's second method is undertaken.

The increased complexity of aerospace vehicles, with emphasis on automatic controls, has considerably altered the role of the human operator in such systems. Routine, burdensome tasks previously performed by man are now automated. As a result, the human task requires a greater emphasis on the monitoring and decision making aspects than on the control problem. Any attempt to automate these decision processes or to augment them for semi-automatic procedures must necessarily be based on a quantitative understanding of human capabilities in complex decision and control tasks. Considerable data are available on the information gathering and processing aspects of human behavior. From these results mathematical models of human decision processes and adaptive behavior have been proposed for specific control situations. In this paper we survey accepted techniques and models for analyzing and predicting human performance in complex multi-control and multidisplay situations commonly found in aerospace systems. The models that we consider have been developed or proposed for the related human functions of information processing, decision making and control. This paper discusses the relative advantages, disadvantages and limitations of each of the modeling schemes. Prospects for mechanizing all or part of the decision functions performed by human operators are considered - specific examples being in the automation of human failure detection and adaption to sudden changes in the system operating conditions.

Electrical and thermal interactions in silicon integrated circuits provide a means for realizing large effective time constants and, correspondingly, low cutoff frequencies in filters. This paper describes a method of analyzing the behavior of a general configuration of lumped or distributed heaters and sensors in an integrated circuit. The method yields results that are amenable to computer evaluation, and that also indicate how heaters and sensors should be shaped to obtain a particular type of response. A practical synthesis procedure is given for filter applications, and experimental results are presented that show close agreement with calculated values for examples of both analysis and synthesis of filters.

This paper presents some aspects of a special multimodality in a two-person game-renju (a Japanese alignment game on a board); the authors believe such a renju multimodality will surely be a new base of ideas and understanding of adaptive processes in artificial intelligence requiring multimodal optimum search. After introducing renju rules, and basic renju technical terms and strategies, this paper develops the necessary concepts such as distance between any two points, four directions, string, jumstring, costring, dead (jum-)string,and criterion function of a placed stone (piece or chip) on a renju board. The essential aspects of renju multimodality studied here are as follows: possible values of the criterion function, the realized number of the highest peak (lowest valley) formed by the criterion function within a game, height of maximum peak vs. the issue of the game, possible number of the highest peak (lowest valley), illustration of the criterion function, developing processes of peak and valley, adaptive mobility of the optimum point, and practical appearance of the multimodality of renju problems.

This paper will describe a new differential amplifier, the dual of conventional amplifiers, whose output is proportional to the differential-mode input current, while common-mode currents are rejected. High common-mode rejection can be maintained over wide bandwidths.

The papers summarized in this discussion reflect some of the characteristics of the human as an information processor and as a user of this processed information in a subsequent decision-making type of situation. In the first of the papers presented here reporting research carried out at the decision-making laboratories at the Pennsylvania State University, it is shown how the human adjusts his rate of responding to particular ratios of reinforcement to particular stimuli not only as a function of the reinforcement ratio but also as a function of the context of other reinforcement ratios he is learning at the same time. The second paper reviews the adaptation of the human information processor in learning and decision-making to programmed changes in variability of reinforcement ratios. These ratios range from zero change in material programmed for learning to small and large changes in the variability of the reinforcement ratios. During these changes overall mean proportions of reinforcement are constant. Of particular importance is the stability of the decision-maker in seemingly disregarding the rather gross changes in the programmed variability of the stimuli he is required to learn and on which he must base subsequent decisions. The third study summarized extends results described earlier. Unlike the earlier study, however, materials in the form of meaningful words are used here as stimuli. Of special importance is the rather "floating" characteristic of formerly neutral, or nonresponse-eliciting, stimuli as a function of the training schedule of un-related words.

This paper is a report on a problem which is thought to have rather general interest in the field of radar and communications systems. This is the problem of interference caused by the reflection of pulsed radar signals from the terrain iii the transmitter-receiver environs. This problem has proved to be rather intractable to handle analytically and therefore the empirical approach presented in this paper might be of special interest to those who have tried some analytic approach. Basically the problem to which this paper addresses itself is as follows: Suppose it is desired to predict the number of pulses and their power levels, arriving at an unintended receiver from a radar transmitter. The task of predicting the direct pulses is difficult, but can be approached analytically; however, what can be done about predicting those pulses reflected from the terrain (or other objects) in the vicinity, especially when no detailed terrain information is available? During the study from which this paper arose, measured data was obtained at several selected sites in the San Diego area by tuning a test receiver to one of the identified, operating radar transmitters at a time. The number of pulses and their 'Power levels were measured by the test equipment and recorded. This procedure was reported for several different transmitters at three test sites. At the same time, a computer program (called MSS-l) has been developed to predict the number of direct pulses and their power ievels.

This paper constitutes Part II of a series of papers on adaptive pattern recognition and its applications. It pertains to optimal, unsupervised learning, adaptive pattern recognition of "lumped" gaussian signals in white gaussian noise. Specifically, both deterministic decision directed learning as well as random decision directed learning algorithms for continuous data are obtained. It is shown that the supervised learning results [1], in particular the partition theorem are applicable in the directed learning approach to the unsupervised case [2].

Amid the highly technical papers presented at a conference such as this, it has been suggested that it might be well if there were a few comments from a representative of those who are the ultimate beneficiaries of these developments, the end-users. While no one end-user can fully represent all of the others, this paper will review one particular process by which transformers are selected and new developments evaluated. This should suggest the general problems of the users, although others might make different evaluations.

The theme of the first technical session was methods of analysis as applied to a variety of electronic power processing systems. The seven papers, all of which could be classified as research papers, presented new concepts and ideas relating to the analysis, design and management, of power by electronic means and provided a theoretical basis upon which many of the circuits and systems presented in subsequent sessions could be examined and interpreted.

Summary form only given. This paper describes a voltage comparator which uses a junction transistor and a silicon diode. The comparator circuit is essentially a blocking oscillator which generates a sharp output pulse when the input waveform passes through the reference level. By taking advantage of the special properties of the semiconductor devices, the comparator can be designed to have an accuracy of ± 5 millivolts throughout a temperature range of 20°C to 4OO.C. A theoretical analysis of the comparator is given with particular emphasis placed on accuracy and switching speed. It is shown that by preceding the voltage comparator with a transistor dc summing amplifier, the reference level for the comparator may be adjusted over a range of ± 25 volts

This paper summarizes the presently-known solutions to the linearity problem. The paper includes not only linear assignment techniques which require the output to be linear, but also linear assignment techniques for the next-state behaviour. For this latter category, it has been shown that state splitting may be required, and the current results in this area are included.

This paper presents systematic synthesis procedures for Lupanov decoding networks. Decoding networks find applications in computer memories for the selection of a particular item of data addressed by a binary code, in counter circuits, in character-by-character code conversion circuits, in the synthesis of switching functions and also in the theoretical derivations of complexity bounds for arbitrary switching networks. In particular, a complete decoding network of n input variables provides a distinct output for each of the 2n possible combinations of the variables. These networks have been studied extensively, both in terms of relay contact realizations1,2 and in terms of gate-type element realizations3. Moore4 has shown that for disjunctive contact networks, hence for networks of branch-type elements, the familiar transfer trees using CM = 2(2n-1) contacts are the only possible minimal n variable decoding networks. In applications where disjunctivity of the outputs is not required, Lupanov5 has shown that the required number of contacts can be reduced considerably (by a factor of 2 as n grows arbitrarily large). Short6 has examined specific design procedures using Lupanov's approach. The main idea is based on the design of a r-variable partial decoder, where r = 2k, k = 2,3,... whose 2r/r outputs consist of mutually disjoint groups of r vertices, collectively exhausting all the vertices of the r-cube such that any vertex is ? distinguished from the rest of the vertices in the group by the value of a single variable. The problem that has been posed7 and solved in this paper is the following: can we take r ≠ 2k for the partial decoder and find Lupanov trees whose costs are smaller than those obtained using r = 2k? We shall prove that this is possible and present systematic procedures for the design of Lupanov trees. The Lupanov trees obtained by our method have the best minimal costs known so far for all values of n. As an example, we have designed a 5-variable decoder which uses only 58 contacts compared to 60 contacts used by Lupanov.

The U.S. Army Corps of Engineers is annually faced with the problem of determining which of several proposed civil works projects to fund. Each project has a specified cost. Furthermore, each project can contribute benefits in a number of different categories; for example, a water reservoir project may contribute benefits in the categories of flood control, power generation, river navigation, general recreation, conservation of fish and wildlife, water quality control, municipal and industrial water supply, irrigation, and other categories. Previous work by the Corps of Engineers has resulted in the development of methods for quantifying the benefit produced in a specific category by a particular project. However, because each project produces a multiplicity of benefits of a variety of types, a straightforward project selection procedure based on benefit cost ratios has obvious shortcomings. Many benefits have natural physical units of measure, but are difficult to measure in dollars. Even if all benefits could be assessed in dollars, they often accrue to different population segments and are hence qualitatively different. An uncritical summation even in this case may result in a loss of information relevant to the decision maker(s). Finally, a yes or no decision on a particular project should be made in the context of other projects with similar benefits which are competing for a part of the same limited supply of capital resources. Thus the desirability of a given project is dependent upon decisions made with respect to other related projects. Explicit consideration should therefore be given to sets of projects acting in concert. The purpose of this paper is to describe a computer program developed by the authors for use by Corps of Engineers planners as an aid in the project selection process. This program is based on a formulation of the multibenefit project selection problem as an integer programming problem. A specialized integer programming code developed by the first author is used to obtain specific solutions. The program can consider as many as 125 projects, each of which can produce benefits in any of 50 benefit categories. Computation times for typical problems are on the order of one minute for a UNIVAC 1108. The paper first summarizes the basic problem formulation. The integer programming algorithm for solving the program is then described. Finally, a summary of computational experience is prescribed.

In order to manufacture high quality electrical equipment for our customers and to maintain a strong competitive position in the electrical industry, it has always been necessary to maintain very rigid standards in the selection of new insulating materials for electrical equipment. This paper discusses the best evaluation techniques which have been developed or adopted during the past ten years and are being used in the selection of insulating varnishes for Class H electrical apparatus. This paper also contains a description of the testing methods used in the evaluation of varnishes and includes data which illustrates these methods. These methods can be modified to include all classes of insulating varnishes.

In this paper we report some results of a study on the range of possible structure of programming languages. The main emphasis is on the range of graphical ("topological" or flowchart) and syntactic structure. For the sake of simplicity and precision we rather severely limit the "semantic structure" of the languages--we restrict ourselves to command (instruction) languages for Turing machines. As we show, this apparently strong limitation imposes very little restriction on the graphical and syntactic structure. The bulk of the paper consists of the presentation of six Turing machine languages. These languages serve to illustrate the range of possible structure and, more important, they allow us to establish the range of a number of structural parameters. All the languages are universal in the sense that in each one we can program every computable function. However, they differ greatly in syntax, graphical structure, ease of compilation (assembly), and in the type of machine, if any, which can operate directly in the language. In brief, we present languages with finite-state, context-free and more complex syntax; languages with "conventional" graphical structure, with block structure and only one transfer per block, with only nested transfers (nested loops), with transfers only to the immediately neighboring instructions, and with only one transfer per program.

This paper will describe a dynamic high-speed three-line access memory cell which occupies the area of a single bipolar transistor. New design is compatible with bipolar processing, with the addition of one extra masking step.

A bounded-transient finite automaton is an automaton for which a single change of an input will not affect the output "far away." This paper investigates the class of events which bounded-transient automata can compute and shows that this class of events is a generalization of the class of definite events. A method of determining whether or not a finite automaton is bounded-transient is described, and the connection between the results of this paper and Kilmer's results is indicated.

This paper will discuss an externally controllable operational amplifier which can be used not only for conventional applications, but for others, by the use of an analog gating technique.

Previous papers have shown that for any given n-input synchronous sequential machine there exists a circuit realization in which the circuit consists of a finite number of identical copies of one module and in which the modules are interconnected in a uniform manner. This paper shows that additionally the signal fan-in to every module and the signal fan-out from every module and from the input can be bounded by a constant and that the modules can be interconnected in a planar structure. This paper also investigates several properties of these circuits and establishes several necessary conditions that these circuits must have. Two desirable design goals are shown to be simultaneously unachievable

The algorithm, model and results of computer programs with a built-in large signal MOS model that provides DC and transient analysis of P, N or complementary MOS circuits will be covered in this paper.

The purpose of this paper is to demonstrate that representations of numbers other than positional notation may lead to practical hardware realizations for the digital calculation of classes of algorithms. It is the authors' opinion that practicality of the use of continued products has been demonstrated. This paper describes current research in the use of continued fractions. Although practicality has not been demonstrated, theoretical results are promising, and the results thus far are presented as a case study of the difficulties which arise when use of a new representation is attempted.

The paper outlines a particular computer aided design philosophy for the control of single input-single output discrete systems subject to disturbances with rational spectral density. The design procedure is based on identification and synthesis; computer aided methods of off-line and on-line process identification provide sufficient information for the synthesis of a constrained input, minimum variance control algorithm of particularly simple form. In this sense, the present approach to control systems design provides a possible systematic alternative to more conventional methods, such as three term control. Since the complete design procedure is composed in part of certain existing techniques, the paper can also be considered as a short review of the subject.

The objective of the work described in this paper was to investigate techniques for reducing the electromagnetic coupling between antennas on aerospace vehicles. In earlier work a theoretical and experimental study was carried out in order to determine the coupling between two antennas, either similar or dissimilar the early part of this paper some results of this analysis are presented. In the appendix some methods are presented to show simple calculating procedures for the determination of coupling. Experimental studies were also conducted where different techniques of reducing coupling were developed and tested. Particular emphasis was placed upon the possible effect of different coupling reduction methods on the far field radiation pattern and gain of the representative antenna. Another important aspect was the bandwidth over which a particular decoupling method is effective. The analysis and measurements reported here have been made on the basis of continuous wave propagation.

The construction of a qualitative theory of climate change based on systems theory is discussed. Specifically, the importance of initial conditions, nonlinearities, feedback mechanisms, forcing functions and stability concepts, as they pertain to the climatic behavior of the Earth are shown. Suggestions regarding areas of prime importance are advanced and feedback mechanisms are identified. The paper serves three goals: (1) provides a first approximation qualitative model outline, useful for subsequent quantitative calculations regarding climate change, (2) introduces systems theorists to the subject of climate, and (3) summarizes current literature in the field of climate behavior, its modification and simulation. A first-approximation, quantitative model based on the principles discussed in this paper will be the subject of a second paper.

This paper will describe a computer study of two-dimensional domain behavior for the design of new high-speed carry generators. Using these devices, the computation time of 30-40 bit full adder and subtractor can be reduced to less than 1 ns.

This paper will commence with a description of the present use and dependence upon microwave systems within the United States, particularly from the point of view of the Federal Government. Secondly, a description will be given of the kinds of advances projected in microwave technology during the 70's and the kind of new systems these advances will bring into use. The paper will then discuss the two major aspects of environmental impact that both present and proposed systems will make. These include the impact of new systems on existing systems and secondly, the impact of a microwave system on the biological environment. Finally, this paper will discuss the implications of environmental impacts insofar as adequacy of boundary conditions in each area.

This paper reviews the state of the art of methods and techniques suitable for use in the automatic optimization of microwave circuits by digital computer. A number of significant recent developments in the rapidly developing area of computer-oriented circuit theory and design which may be directly applied to high frequency or distributed circuits are emphasized. A fairly extensive classified bibliography on optimization methods and on some applications, primarily in the microwave area, is included. The principal aim of this paper is to bring the microwave circuit designer up to date with developments in computer-oriented optimization methods and techniques and their application to microwave circuit design. To achieve this, the state of the art is reviewed in a tutorial manner, highlighting the talk with material taken from a variety of recent research papers, published during the last year or so. Promising areas for future work are indicated. A fairly extensive classified bibliography on optimization methods and on some applications, primarily in the microwave area, concludes the paper. Papers and books not necessarily in the microwave area are referenced if it is felt that the subject matter is, nevertheless, relevant. The first question that a microwave circuit designer might have is what material should he read to convince himself that he can solve a wide variety of design problems using optimization techniques. Five articles [1]-[5] are referenced under the heading of Reviews of Network Optimization which should probably be found most useful.

In this paper we report some results of a study on the range of possible structure of programming languages. The main emphasis is on the range of graphical ("topological" or flowchart) and syntactic structure. For the sake of simplicity and precision we rather severely limit the "semantic structure" of the languages--we restrict ourselves to command (instruction) languages for Turing machines. As we show, this apparently strong limitation imposes very little restriction on the graphical and syntactic structure. The bulk of the paper consists of the presentation of six Turing machine languages. These languages serve to illustrate the range of possible structure and, more important, they allow us to establish the range of a number of structural parameters. All the languages are universal in the sense that in each one we can program every computable function. However, they differ greatly in syntax, graphical structure, ease of compilation (assembly), and in the type of machine, if any, which can operate directly in the language. In brief, we present languages with finite-state, context-free and more complex syntax; languages with "conventional" graphical structure, with block structure and only one transfer per block, with only nested transfers (nested loops), with transfers only to the immediately neighboring instructions, and with only one transfer per program.

The present paper covers a test procedure for the evaluation of thermal endurance of enamelled wires at constant voltage. In the NEMA Magnet Wire Technical Committee Presentation of 19711 it was proposed in clause 7 of “Additional work recommended” of the “Joint Sections Committee Report” to develop meaningful test methods for voltage endurance of magnet wire at elevated temperature. In the EIC 1971, the author presented a paper on statistical analysis applied to data obtained by IEEE 57.2 In the conclusion of this paper, recommendations for test development for magnet wire evaluation incorporating voltage and temperature were given. This paper proposes such a test method.

This paper will discuss methods of generating high-voltage waveforms (up to several thousand volts) at modest power levels (several hundred watts). Circuit techniques and system configurations, which involve parallel series chains of light triggered bilateral switches, employing solid-state components, will be described.

In this paper we are concerned with the decomposition of synchronous sequential machines into interconnections of component machines. The term interconnection is taken literally; the input terminals of a component machine must be directly connected to either i) a logical constant, ii) an output terminal of a component machine, or iii) an external input line. A fundamental question regarding such decompositions is "Does there exist a finite set of component machines such that any given machine can be realized by an interconnection of copies of the component machines?" The answers to this and related questions comprise the scope of the paper.