Engineering Commons^™

An Approximate Min-Max Solution To The Identification Problem, Richard Loyd Warren

Masters Theses

"A technique based on the min-max error criterion is proposed for identifying a dynamic system in terms of a discrete-time model from the system response to a deterministic input. A linear, single-input/single-output, lumped-parameter, time-invariant (at least during the measurement interval) system is assumed. Although the resulting problem is non-linear, it is shown that linear programming techniques are applicable if a realistic approximation is made. The technique is implemented on the digital computer and evaluated by considering a number of typical discrete-time systems. The effects of additive noise are also considered"--Abstract, page ii.

A Digital Simulation For Optimum Nonlinear Detection Of Binary Signals In Impulsive Noise, Ralph Bernard Fluchel

Masters Theses

"The problem of reducing the effects of impulsive noise on binary communication systems is considered. A digital simulation of such a system is developed in order to find a nearly optimum (minimum probability of error criterion) nonlinear device to precede detection. The level of the Gaussian noise power is found to be the most important parameter for determining the nearly optimum nonlinear device"--Abstract, page ii.

The Effects Of Magnetic Fields Upon The Steady State And Short Circuit Performance Of Low Voltage Multi-Link Fuses, John Thomas Schultz

Masters Theses

"The steady state and short circuit performance of multi-link fuses is investigated, as it is affected by the magnetic field of the nearby returning current and the current flowing in the fuse itself. The fuses investigated are low voltage (600 volts or less), multi-link fuses with ratings between 400 and 4000 amperes. The effects of the above mentioned magnetic fields upon the steady state current distribution within the fuse and upon the mechanical forces acting on the individual fuse links are calculated, while the effects of close proximity of the returning current upon short circuit fuse performance are shown by …

Digital Simulation Of A Signal Conditioner/Bit Synchronizer, Quentin Reed Webb

Masters Theses

"A bit rate filter type signal conditioner/bit synchronizer used in a split-phase PCM system is modeled digitally in three stages to determine overall system performance. General computer algorithms are developed to approximate independent operation of each major segment within the device. Bit rate effects and signal time-base error are investigated yielding the characteristic G-curves for the phase detector. These characteristics facilitate subsequent analysis of a linear equivalent digital-data-transition tracking phase-locked loop under several typical time-base error input conditions. Using bit error probabilities as performance criteria, the conditioner/synchronizer is found to perform adequately at signal-to-noise ratios exceeding +10dB. For inputs below …

Probe Detection Of Microwave Perturbation In A Magnetoplasma, Show-Ran Peter Wang

Masters Theses

"When a microwave signal with a frequency near the electron upper hybrid resonance is applied to a magnetoplasma, it will excite the plasma. The perturbation of plasma can be detected by using Langmuir probes. The particular plasma used in this experiment is an argon discharge type and the density is relatively low (n~10⁸ cm⁻³). The electron temperature is about 7000º K. Because a magnetic field is applied to the plasma, it is necessary to use the probe theory in a magnetic field to calculate the electron number density from the electron temperature. In order to determine the electron upper hybrid …

Next-State Equation Generation For Asynchronous Sequential Circuits - Normal Mode, Gregory Martin Bednar

Masters Theses

"This paper describes the known methods of generating next-state equations for asynchronous sequential circuits operating in normal fundamental mode. First, the methods that have been previously developed by other authors are explained and correlated in a simple and uniform language in order that the subtle differences of these approaches can be seen. This review is then followed by a development of a new method for generating minimal next-state equations which has some advantages over the previous methods. From the comparison of the previous known methods, it is noted that any one of these methods may be desirable for certain designs …

The Dinary Numerical System And Its Applications To Modern Parallel Processing Computers, Hai Chi Koo

Masters Theses

"The Dinary Numerical System (DNS) is an extension of Binary Numerical System. By utilizing the newly developed Large-Scale-Integration technology, DNS can be implemented as a new concept of computer organization which has the merits of parallel processing. DNS consists of three field elements: they are -1, 0, and +1. Since DNS representation is not necessarily unique, new algorithms to generate the number of representations and all equivalent representations are presented in this paper. A Block-Oriented-Computer (BOC) is developed as an array of computing systems. For each of the systems, the arithmetic unit is essentially a modified Digital Differential Analyzer. The …

An Explicit Simulation Of Overland Flow, Glendon Taylor Stevens

Masters Theses

"Computer simulation of overland flow requires a mathematical model which converges to a true solution. This paper presents a method, based on measurable physical parameters, which insures convergence of a fixed grid explicit numerical solution. Analysis of a wide range of overland flow plane parameters has resulted in an equation of the form Δt = 0. 2 ΔX/S_o, for the time step increment. This equation provides values for Δt that will insure stability and convergence of the fixed grid explicit numerical method of solving the spatially varied unsteady flow equations. The effect of subarea shape on the rising …

Computerized Methods For Estimating Heating-Cooling-Ventilating System Usage In All-Electric Buildings, Carl William Glaser

Masters Theses

"This thesis develops a method and a resulting computer program for estimating the energy input required by the heating-cooling-ventilating system of a building. The computer program has the ability to model the ventilating system and all the inefficiencies introduced when the heating system is allowed to fight the cooling system to provide temperature control within the conditioned space. The program also can consider mechanical heat reclaim accurately, making the consideration on a time of need versus a time of occurrence basis. The program is written to utilize input data in rather simplified form. This allows the program to be used …

A Study Of Microstrip Transmission Line Parameters Utilizing Image Theory, Joseph Louis Van Meter

Masters Theses

"This paper is a theoretical investigation of the potential, electric field, capacitance, and characteristic impedance of the open strip transmission line or microstrip configuration based upon the classical Thomson Image Technique. It provides the basis for determination of the charge distribution on the strip and reports impedance values which compare favorably both with experimental values and theoretical work in the current literature"--Abstract, page ii.

Computer-Controlled Processing Of Radar Precipitation Signals And Lightning Data, Gary Wayne Neal

Masters Theses

"A system for the simultaneous processing of radar precipitation signals and lightning stroke location data has been designed. The system is to be used for the purpose of determining spatial and temporal correlations of lightning and thunderstorm precipitation. This system has been developed to gain further information about the cause-effect relationship between lightning stroke activity and thunderstorm precipitation. Data acquisition and processing is accomplished through the use of an AN/FPS-18 radar set, an SCC-650 digital computer system, a radar interface, and a set of lightning detectors. The radar set provides information on precipitation activity in a circle of radius fifty …

Fault Detection In Asynchronous Sequential Circuits, Jeng-Chuan Kau

Masters Theses

"As the asynchronous sequential circuit has become more and more important to digital systems in recent years high reliability and simple maintenance of the circuit is stressed. This paper presents a fault-detection algorithm which will be applicable to most of the practical asynchronous sequential circuits. The asynchronous sequential circuit is treated from the combinatoric point of view. First the minimal set of states, both stable states and unstable states, sufficient to detect all possible faults of the circuit is found from the fault table. Then a test sequence is generated to go through these states. It is assumed that testing …

A Mathematical Model For The Neutron Dependence Of Neutral Base Region Recombination, Jonas Bereisa Jr.

Masters Theses

"A discrete, deterministic, mathematical model is developed for the neutral bulk base region of graded base devices which accounts for both the effects of recombination and the built-in electric field in the bulk base region at low, intermediate, and moderate current/injection levels. This model is applied to a study of neutron radiation effects in double diffused silicon transistors. Exact expressions are developed for the injected emitter, base and collector currents, the base recombination term, and the ratio of base current increase to collector current decrease as functions of device geometry, impurity profile, and neutron fluence. Neutron-induced current gain degradation is …

An Analysis Of A Quadrature Double-Sideband/Frequency Modulated Communication System, Denny Ray Townson

Masters Theses

"A QDSB/FM communication system is analyzed with emphasis placed on the QDSB demodulation process and the AGC action in the FM transmitter. The effect of noise in both the pilot and message signals is investigated. The detection gain and mean square error is calculated for the QDSB baseband demodulation process. The mean square error is also evaluated for the QDSB/FM system. The AGC circuit is simulated on a digital computer. Errors introduced into the AGC system are analyzed with emphasis placed on nonlinear gain functions for the voltage controlled amplifier"--Abstract, page ii.

The Evaluation Of Radar Guided Missile Performance Using Various Target Glint Models, Franklin Delano Hockett

Masters Theses

"A radar guided missile seeker approaching a target arrives at a crossover point beyond which the entire target is illuminated by the missile seeker's radar antenna beam. When this occurs, for a complex target, the radar senses a time changing target center location which may greatly complicate the terminal tracking portion of the seeker's flight. This thesis documents various glint models in use today and compares the performance of these models by using a seeker model to determine the effect of glint on terminal tracking performance. A large volume of data has been compiled describing various radar characteristics of complex …

Iddap -- Interactive Computer Assistance For Creative Digital Design, Richard Franklin Crall

Doctoral Dissertations

"A new computer-aided design program to assist in the initial phases of logical design is described. The program, intended for use via an on-line remote terminal, will allow the designer to study and experiment with design alternatives during the initial creative design phases. An ALGOL-like language is used for specifying the system being designed. In addition to simulating the design, the program allows the user to perform on-line design changes, reorganize data and generate timing diagram information"--Abstract, page ii.

Synthesis Heuristics For Large Asynchronous Sequential Circuits, Robert Judson Smith

Doctoral Dissertations

"Many well-known synthesis procedures for asynchronous sequential circuits produce minimal or near-minimal results, but are practical only for very small problems. These algorithms become unwieldy when applied to "large" circuits with, for example, three or more input variables and twenty or more internal states. New heuristic procedures are described which permit the synthesis of very large machines. Although the resulting designs are generally not minimal, the heuristics are able to produce near-minimal solutions orders of magnitude more rapidly than the minimal algorithms. A method for specifying sequential circuit behavior is presented. Input-output sequences define submachines or modules. When properly interconnected, …

Model-Referenced Adaptive Control Of Plants With Noise And Inaccessible State Vector, Donald James Schooley

Doctoral Dissertations

"The problems of process noise, measurement noise, and inaccessible process state vector in the model-referenced system of Shackcloth-Butchart-Parks (SBP) are investigated. Since the system no longer possesses a static equilibrium point, the concept of asymptotic stability does not apply. However, by using Liapunov-like ideas, the error between the plant (process plus controller) and the model is shown to enter a region about the origin if certain bounds are known. A possible technique for reducing the size of this region is to feedback a function of the error into the plant's input. Several methods of circumventing the inaccessible state vector problem …

A Model For Predicting The Energizing Transients Of Station Capacitor Banks, Dennis Oliver Wiitanen

Doctoral Dissertations

"A model for predicting the transients associated with energizing ungrounded wye-connected station capacitor banks was developed. The model has the capability of simulating non-simultaneous operation of the breaker poles, pre-insertion resistors, and the presence of previously energized capacitor banks at the same station. To investigate various system representations, a computer solution of the transient response of an arbitrary network and the computer solution of the generalized Thevenin impedance of a complex network are presented. A simple power system is defined and used to investigate the modeling of various system components, using the computer solution methods. Simplification of the model due …

A Simulation And Diagnosis System Incorporating Various Time Delay Models And Functional Elements, David Michael Rouse

Doctoral Dissertations

"The application of digital simulation to all phases of digital network design is considered here as oppossed [sic] to development of simulation for one or two restricted parts of the digital process. For this reason a simulator is presented which can be consistent by varying the level of expression from the simulation of architectural structures to such detailed simulation requirements as race analysis of asynchronous sequential circuits. In order to make system simulation more than just an idea, it must be capable of handling large circuits in reasonable times. It is demonstrated that functional simulation has the potential to increase …

Time-Domain Analysis And Measurement Techniques For Distributed Rc Structures. Ii. Impulse Measurement Techniques, Robert C. Peirson, Edward C. Bertnolli

Electrical and Computer Engineering Faculty Research & Creative Works

Transient analysis for uniform RC structures is considered in this paper. A method is presented for determining the parameters of such structures. The measurements are obtained using impulse excitations in open-circuit and short-circuit configurations. The theoretical results obtained predict fairly the experimental results. © 1969 The American Institute of Physics.

An Approadch For Suboptimal Feedback Control., Ronald D. Andreas

Electrical and Computer Engineering ETDs

This dissertation presents an approach for achieving suboptimal feedback control of a class of time-invariant systems, with primary emphasis on linear, time-optimal regulator systems. The approach involves the specification of a family of suboptimal feedback control policies that has the unique feature that each policy in the family will produce a stable system and a suboptimal performance that can be measured against optimal performance. In particular, the performance deviation from optimal afforded by any of these policies is parametrically bounded, and by varying the parameter a hierarchy of policies can be determined. For bang-bang systems, the resulting suboptimal controller design …

A Note On The Free Distance Of A Convolutional Code, Alexander Miczo, Luther D. Rudolph

Electrical Engineering and Computer Science - All Scholarship

A counterexample to a conjecture on the number of constraint lengths required to achieve the free distance of a rate l/n systematic convolutional code is presented.

A High Voltage Regenerative Pulse Modulator, James D. Doss

Electrical and Computer Engineering ETDs

This paper describes the design of a High Voltage Pulse Modulator that employs positive feedback to generate output pulses of variable width. The modulator is intended for use as a switching device to drive the modulating anode of a five-cavity klystron. The klystron develops more than one million watts of peak power at a frequency of 805 MHz, with duty factors up to 12%. The modulating anode voltage drive requirement is approximately 80 kV with the current drain being of the order of 0.01 ampere. The total modulator current output will be approximately one ampere with most of the current …

A Noise Figure Lower Bound And A Frequency Converter, Grant Myers

Department of Electrical and Computer Engineering: Faculty Publications

This article establishes a lower bound for the noise figure of two networks in cascade. This is illustrated by means of a passive low-level frequency converter which utilizes a back diode as a nonlinear conductance. Other investigators have analyzed the properties of low-level frequency converters. However, the greatest lower bound of a combination of the converter and intermediate frequency amplifier system was not considered previously and is the subject of this paper. Knowledge of this lower bound has practical significance because if a measured value of noise figure is less than this lower bound, it follows that the measurement is …

Simulation And Analysis Of A Second-Order System With An Added Finite Pole And/Or Zero, William Robert Bodwell

Electrical & Computer Engineering Theses & Dissertations

An original system whose transfer functions consist of two infinite zeros and two finite poles is modified by the addition of a finite pole and/or finite zero. Statespace techniques are used to obtain equations for simulating the transfer functions on both analog and digital computers. Sample cases were selected and simulated to demonstrate the various types of responses. Analysis of the computer results and mathematical analysis of the transfer functions provide nomographs for estimating the response of the systems to a step-input and methods for obtaining equivalent pole/zero configurations.

Correlation Of 1/F Noise With Threshold Drift In Mosfet's, John Peter Rooney

Theses

A brief review of the fundamentals of MOSFET's is given. A correlation between threshold drift and surface states is made, with a following correlation between surface states and 1/f noise. It therefore follows that MOSFET's with a high 1/f noise level will drift more than those with a low level of 1/f noise. Experiments were carried out to show this effect, but no clear cut conclusions can be drawn from the experimental work.

Stability Analysis Of Nuclear Reactor Feedback System, Ashok K. R. Bizzul

Electrical and Computer Engineering ETDs

A nuclear reactor feedback control system is analyzed for stability conditions in frequency domain by two methods. The nuclear reactor is modeled by a single delayed neutron group model, and the methods of analysis are Popov's analysis and describing function analysis. The feedback system with linearized nuclear reactor model is analyzed by Popov's method. A more exact non-linear reactor model is used in the feedback control system for describing function method of anal­ysis. It is found by both methods that the system is absolutely stable. But describing function analysis gives greater insight to the study of the relative stability of …

Concepts In Lsi Servo-Control-Electronics, Alfons Alfred Tuszynski

Dissertations

This thesis deals with the engineering aspects of control electronics. It examines modern concepts of servo-control theory in the light of recent developments in the technology of monolithic circuits. Applicational considerations are slanted towards Aerospace standards of reliability and power-consumption economy.

Conclusions drawn from the discussion of fabrication constraints and performance requirements lead to a preference for digital implementations. Yield problems on one hand and aging effects on the other greatly reduce the feasibility rating of analog arrays.

Current practice in servo-control electronics revolves around purely analog implementations, sampled-data systems and Primitive on-off arrangements. The motivation behind the status quo …

Reliability Aspects Of Micropower Majority-Charge Logic, James T. Henderson

Electrical and Computer Engineering ETDs

This thesis is a study of the reliability aspects of micropower majority-charge techniques as applied to fundamental-mode clocked-se­quential circuit design. A· sample sequential circuit is used as the basis for the extension of previously developed circuitry. This previously developed circuitry is applied to the sample sequential circuit in the form of Micropower Majority-Charge Flip-Flops. These flip-flops are very immune to transient noise and offer reduced power consumption over standard integrated circuit flip-flops by pulsed power application. The use of AND gates and OR gates with amplifiers and NAND gates or NOR gates for memory elements is reported. It is shown …