Digitized Theses and Dissertations (1940 - 2009)
Permanent URI for this communityhttps://hdl.handle.net/10657/3932
About UH Libraries' Digitized Theses and Dissertations Project
University of Houston (UH) Libraries is engaged in a multi-year project to digitize and deliver online its collection of print theses and dissertations dating back to 1940, making the full breadth of scholarship produced by UH students more readily accessible around the world. There is no cost to the author for this service.
Alumni and other readers will be able to view these works as they are processed and made available through UH's open access repository. Works that are presumed to be under copyright will be restricted only to users who have an active CougarNet ID.
Please note, text may be faint or difficult to read, and pages may be missing or misnumbered in the print copies of theses or dissertations. UH Libraries staff have made every effort to provide the highest possible quality representation of the original works. To protect privacy and other rights, some personally identifiable information and/or copyrighted material is redacted from the works in this collection.
Theses and dissertations will continue to be made available through interlibrary loan (ILL) to other libraries, as when they were only available in print.
Requests for withdrawing works (except electronic theses and dissertations) must be directed to the online Takedown Request Form. Any other questions about this project may be directed to cougarroar@uh.edu.
Browse
Browsing Digitized Theses and Dissertations (1940 - 2009) by Department "Electrical and Computer Engineering, Department of"
- Results Per Page
- Sort Options
Item A basic driven microprocessor based experimentation system(1978) Kong, Tai Loy; Tavora, Carlos J.; Linton, Thomas D.; Huang, Jung-ChangThis thesis describes a set of programs designed to extend the capability of the BASIC language to allow periodic acquisition of signals and control of devices. These programs allow a system user to define sets of inputs and sets of outputs from a group of signal lines. Output data are converted into the data formats of external devices and input data are converted into a data format compatible with programs written in BASIC. The extensions to BASIC described in this thesis provide a casual programmer with the means to perform data acquisition and control with minimum knowledge of system software. An application of the software described in this thesis to a system that will test student's proficiency in the use of laboratory equipment is presented and evaluated.Item A basic interactive system for image digitization, processing, and display(1985) Lee, Ming-Chih; Pai, David M.; Gardner, Gerald H. F.; Pyle, Leonard Duane; Simpson, Anne L.Many applications in science and industry require convenient tools for converting analog images into digital form and for performing some desired processing by computer. This thesis describes the design of a basic interactive system for image digitization, processing, and display. The system provides several frequently used image processing operations: digitization, normalization, segmentation, contrast enhancement, contouring, and image enlargement by splined interpolation. Each operation is performed by issuing a proper command provided by the system. These operations are described in detail with illustrations. The system has been implemented on a VAX-11/780, and utilizes a digitizer, two interactive graphic devices, and one electrostatic plotter. Source code listings are included.Item A calibration system for a Sordhi tube(1980) Manseur, Rachid; Everett, Robert L.; Paskusz, Gerhard F.; Yonovitz, AlbertA Sondhi tube is a reflectionless acoustic tube that is used for the measurement of the glottal waveform. One of its major applications is the detection of laryngeal pathologies. The operation of the tube is based on the assumption that no reflections take place and that the tube behaves as an infinite prolongation of the vocal tract. Therefore, the undisturbed glottal wave can be recorded ty means of a microphone positioned at any point inside the tube. This Thesis is about the design of a calibration system for a Sondhi tube. In order to verify and eventually adjust the operation of the tube, the latter must be excited by a sound source with known characteristics and the sound wave excitation must be measured in order to obtain a reference signal to which the corresponding tube microphone measurement can be compared. In the system that we propose here, the sound source was implemented as a loudspeaker driven by a power amplifier and the reference measurement was obtained through measurement of the loudspeaker cone displacement. The acoustic wave produced by this sound source is set propagating inside the tube by acoustical coupling so that the tube microphone would record the sound pressure level signal corresponding to the loudspeaker cone displacement recorded through the calibration device circuitry.Item A comparative analysis of cascade and feedback compensation(1966) Norling, Donald K.; Parker, Sydney R.; Whitaker, Thomas N.; Gunderson, Richard H.; Kittinger, William T., Jr.This thesis presents a comprehensive analysis and comparison of the relative effects of cascade and feedback compensation upon the steady-state and dynamic performance of feedback control systems. In particular, system sensitivity, steady-state system error and actuating signal, log-modulus response, and pole-zero considerations are investigated for cascade compensation and various forms of feedback compensation. Equations relating equivalent feedback and cascade compensators for a given uncompensated plant and an overall system transfer function are developed. Conditions are specified for the realizability of feedback compensators as R-C networks. Specific advantages and limitations of the various modes of compensation are noted and general insight is provided into the relative suitability of cascade and feedback compensation for a given system and set of performance specifications.Item A comparative analysis of international color television systems in use today(1968) Tomaras, Nicholas K.; Michaels, Edward L.; Whitaker, Thomas N.; Schneider, William P.This thesis represents a comprehensive analysis and comparison of the different color T.V. systems now in use. I was motivated to select this topic from the fact that very limited literature is available and not systematic. I hope the study of the individual systems and their comparison with each other will enable the T.V. engineer to utilize some of the ideas which others are using, thus contributing to the advancement of T.V. technology. Television is not an area well-known yet, the outer space exploration has introduced new technological problems to the field as: bandwidth compensation techniques for real time deep space picture transmission, etc. Since the modern T.V. technology is based on the presently used commercial T.V. systems, it is important to be familiar with them and utilize any of the advantageous characteristics they have.Item A comparative study of feedforward and feedback control systems for processes with dead time(1969) Wise, Jerald D.; Schneider, William P.; Whitaker, Thomas N.; Michaels, Edward L.Various reasons for improving the response of a feedback type control system are given. The best response of a conventional feedback type of control system is described and plotted. For a given dead time process, three conventional modes of controller action are adjusted and the output response determined for each system. These output response curves are shown in such a manner, in the time domain, that differences in controllability are easily seen. A process example of a typical industrial blending system is discussed and analyzed, using conventional feedback techniques and feedforward elements. The thesis closes with an evaluation of the results obtained for each of the process systems that were analyzed.Item A computational algorithm for direct generation of difference equations coefficients utilizing matrix series approximations with checks by Tustin transformation(1975) House, Millage, Jr.; Schneider, William P.; Denman, Eugene D.; Sinkhorn, Richard D.This research is directed at the basic need of all who endeavor to represent a system by a set of difference equations for computer applications. Generation of coefficients that are valid for an n'th degree system in both the s-state domain and the z-discrete domain utilizing matrix series approximations are the goals of this research. There has been extensive research in the use of digital computers as controllers. There is a need to provide a computer program based on a sampling time that will give a desired discrete matrix representation and the companion set of difference equations for a discrete system.Item A computer system for determining the dynamic equivalent of a synchronous generator(1981) Sombar, Stephen Michael; Lee, Kwang Y.; Markenscoff, Pauline; Marquina, Nelson E.There exist today state-space or other similar models to adequately represent physical apparatus or systems. Although these models are general, industry today has the problem of determining the exact specifics of the model to be used (i.e., the parameters). This has been a major stumbling block to model usage, and has limited its application. The parameters of a given synchronous generator are to be determined in the laboratory. They are found by a computer determination process, called the weighted least squares estimation (WLSE) method. A comparison between a generator model with WLSE determined parameters and the original generator system is made. The WLSE method is discussed in general. Further applications of WLSE are shown via applying it to two other systems (simulated). A microprocessor based computer system is used for collecting generator data in the laboratory. The hardware and software used are described.Item A computer system for the mugfile problem(1976) Nguyen, Ha Hoang; Batten, George W., Jr.; Townes, James R.; McInnis, Bayliss C.The Oakland. Police Department CRIME system is a computerized, version of the familiar police mugfile. It is essentially an automatic information storage and. retrieval system of criminal records. This project involves the incorporation of a so-called "look-alike" algorithm into the searching capability of the CRIME system. This algorithm can be used to search through a mugfile and to bring out records that closely resemble a particular facial image. Necessary modifications have been made to the system to achieve the above stated objective. Subjects in the hit list - a list of suspects compiled by the system using descriptive characteristics input by an user - can be ordered according to the degree of their facial similarity with respect to some input image.Item A computer-aided method for the factorization of matrix polynomials(1980) Chahin, Nicolas; Shieh, Leang-San; McInnis, Bayliss C.; Donaghey, Charles E.An extended multidimensional Newton-Raphson method is proposed for the factorization of matrix polynomials. A root-locus approach and a matrix continued fraction approach are presented to make initial guesses for rapid convergence of the Newton-Raphson method. The computer-aided method can be used to determine the spectral factors of a matrix polynomial for the analysis and synthesis of kinematic and dynamic systems and to obtain the spectral factorization of a matrix polynomial for optimal control and filtering problems. The same approach can be applied to determine the nth root of a real or complex matrix.Item A computerized method for fast time-domain analysis of EEG data, and its comparison with spectral analysis(1979) Gosalia, Atul Premchand; Ktonas, Preiklis Y.; Bargainer, James D., Jr.; Salamy, Joseph G.A digital system for fast time-domain analysis of sleep electroencephalograms (EEGs) is described, A TI-980 minicomputer-based system utilized for period analysis is extended for simultaneous measurement of amplitude. This is accomplished through the development of the necessary hardware and software interfaces between an A/D converter and the TI-980 minicomputer. Use is made of the fast direct memory access channel (DMAC). A simultaneous period-amplitude analysis on sleep stage 4 EEG epochs is performed to demonstrate the applicability of the system. The results of the period amplitude analysis are compared to the results obtained by EFT-based spectral analysis procedures applied on the same EEG epochs.Item A continuous incremental impedance resolving system(1966) Chavez, Rufus; Badger, A. S.; Geddes, Leslie A.; Kittinger, William T., Jr.; Whitaker, Thomas N.Electrical impedance is a property of living tissues and organs. It has been noted by physiologists that this impedance and its variations reflect the function of cells and organs. Some of the conditions that may cause a variation in the electrical impedance are: thyroid function, estrogenic activity, respiration, cardiac activity and blood flow. Variations in the total impedance, the resistive component and the reactive component, have all been detected at different frequencies. However, simultaneous detection of both the resistive and reactive components has not been reported in the literature. It was the purpose of the research described in this thesis to design and build the necessary electronic equipment to simultaneously detect both components without the need to make intermediate adjustments of balance controls commonly found in bridge instruments. Chapter I is a discussion of the significance of electrical impedance in living tissues and organs. The second chapter is devoted to instrumentation techniques and important considerations. Chapter III presents a technical discussion of all the electronic equipment developed to achieve the previously stated goals. The fourth chapter presents the results of tests of the system which show that the system has the capability of detecting both resistive and reactive variations of less than 1 ohm with a stability of better than 1 part in 5000 for a time interval of greater than thirty minutes.Item A control system design for stabilization in three-dimensional space of a cart mounted inverted pendulum(1985) Hoang, Le Gia; Schneider, William P.; Shieh, Leang-San; Denman, Eugene D.; Rajaram, Navaratna S.; McInnis, Bayliss C.A system of an inverted pendulum mounted on a cart such that the inverted pendulum can fall in three-dimensional space is modeled. In this thesis, the system differential equations of motion are derived and a linearized transfer function is developed from which the natural characteristic of the system is studied. The system dynamics are analyzed and a control system is designed to stabilize the inverted pendulum in a vertical position at a given cart location. Also, an analog control technique is synthesized to control the movement and the position of the cart. The control system performance is evaluated, first by using the definition of the "Initial-Value Theorem" and the "Final-Value Theorem" to study the natural characteristic and time response of the system, and then using a system simulation computer program which provides a graphic analog output of the motion of the inverted pendulum and the motion of the cart.Item A controller for the Ampex TM-7 magnetic tape transport(1973) Hart, Robert Neil; Bargainer, James D., Jr.; Whitaker, Thomas N.; King, Willis K.A controller has been designed and built to interface an Ampex TM-7 magnetic tape transport to the Hewlett-Packard 2100 computer. The design is essentially the same as the Hewlett Packard 13181A controller (1), which is designed for a Hewlett Packard tape machine. The design of the controller was modified as required to adapt to the Ampex tape machine. This allows the use of all Hewlett Packard software for the 2100 computer (2). In addition, a design will be described which will allow use of the controller and tape machine with the SDS-92 computer.Item A data compaction technique for continuous gray-tone digital images(1980) Earickal, Indusekar B.; Batten, George W., Jr.; Glover, John R., Jr.; Rhodes, Benjamin T., Jr.This thesis is the result of initial research work done towards a data compaction technique with the aim of reducing redundant two- dimensional image data. An attempt has been made to facilitate post compression application of image processing techniques on the compressed image in a manner as to achieve improvement in processing speed. The image is subdivided into smaller subpictures which are then represented in the compressed image by a few descriptors. Provisions for storing high-contrast regions, where the above approach fails, and for preventing excessive subdivision of the image are an integral part of the technique. Results obtained with 20 images, using a redimentary implementation of the algorithm are given. The scope and nature of further work to be done and the advantages and disadvantages of the approach are also discussed.Item A dedicated intelligent peripheral processing system with intelligent A/D for the PDP-11 computer(1980) Beste, Randal T.; Glover, John R., Jr.; Ktonas, Preiklis Y.; Kleis, Stanley J.A distributed processing system for the PDP-11 family of computers is described. The system is based upon the use of Dedicated Intelligent Peripheral Processing modules to enhance the processing power of the host computer. The modules provide unique pre-processing and parallel processing capabilities so that less of a burden is placed on the host computer. Therefore, more of the system resources are made available for data computation. The various modules communicate with the host computer by using a single DMA controller. A description of the system's architecture is given along with a design for an Intelligent A/D Peripheral Module. Three applications on the use of the A/D module are also presented. The system has been implemented successfully on a PDP-11/34 computer system using the Intelligent A/D Peripheral Module described.Item A digital procedure for finding all directed subgraphs of a signal flow graph(1969) Greenwade, Jack D.; Chen, C. F.; Schneider, William P.; Childs, S. BartThe algebra of flow graphs is well defined and several techniques are available for obtaining input-output relations. A brief review is presented with the addition of the application to sampled data systems. However, all of the techniques for obtaining input-output relations depend upon the users ability to identify all subgraphs within his system flow graph. Several papers have been presented concerning error free identification of these loops and paths, however, the techniques involved are usually laborious and lengthy. Through the use of the connection matrix approach of Mr. C. V. Ramamoorthy a digitial computer program is developed that finds all loops and paths that exist within a flow graph. This research was restricted to only continuous variable systems, however, techniques are suggested for handling the hybrid case, containing discrete as well as continuous variables. Several examples are preseneted to demonstrate the capability of the program. Suggestions are also made for handling multi-input-output problems.Item A dominant data matching method for identification, model reduction and design of multivariable control systems(1980) Ayed-Kralfa, Djillali; Shieh, Leang-San; Donaghey, Charles E.; Lee, Kwang Y.A dominant data matching method is presented to fit the matrix coefficients of a matrix transfer function from the selected frequency response data of a multivariable system. The problem of modelling transfer functions of single-input single-output systems from frequency response data has been investigated by Shieh and Cohen. Their interpolation method has been applied to the identification of minimum and non-minimum phase systems. The method utilizes a relatively limited amount of experimental data to obtain the model transfer function of a single variable system. In this thesis, Shieh and Cohen's method is extended to develop a dominant data matching method for the identification of multivariable systems. A set of selected frequency response data points is used to solve a matrix equation CX = D, where X contains the matrix coefficients of the unknown matrix transfer function and, C and D are not square in general. To solve for X, a generalized inverse of C in the least squares sense is sought. The same method is then applied to determine the reduced-degree model of a high-degree multivariable control system and to design a dynamic controller for a coupled multivariable system.Item A fast computing technique of time-and frequency domain data conversion(1976) Nusrullah, Dost Mohammad; Chen, Chih-Fan; Schneider, William P.; Graff, William J.The objective of this research is to develop a fast computing technique for the Inverse Fourier transform approximation of a given frequency response data. Frigyes and Langer's formula involving eight straight line segments for the Fourier transform is investigated. The new approach is particularly suitable for the middle frequency range. The numerical differentiation and integration process involved in the approach is carried out by using Walsh functions and an operational matrix. Several applications are investigated.Item A feasibility study in applying read-only memories to the analysis of nonlinear analog function responses(1972) Dohmann, Edgar Lee; Whitaker, Thomas N.; Lee, Samuel C.; Schneider, William P.; White, Ardis H.The measurement of physical qualities such as temperatures, pressures, and flow rates is both desirable and essential in many industrial processes. Devices such as thermocouples and other transducers have been developed which yield a predictable voltage or current response to the external stimulus. The use of such devices which produce voltage and current signals provide economical and convenient capabilities to measure the desired physical qualities because many signals may be terminated and monitored at a central location such as a control room. Thermocouples for temperature measurement are the most widely used devices in such industrial process applications because of their simplicity, accuracy, and low cost. As most such devices which respond to an external stimulus, the relationship between the voltage generated between thermocouple junctions and the temperature of the junctions is nonlinear. This nonlinear relationship poses the greatest problem in designing and constructing accurate temperature indicating devices. Several analog-to-digital conversion techniques have been developed which compensate for the nonlinear function being measured and produce temperature readouts in proper engineering units. All techniques developed for commercial applications strive to attain satisfactory degrees of speed, simplicity, accuracy, economics, and reliability. Most techniques offer some advantages in two or more of these qualities but no presently available techniques offers a clear advantage in all five qualities. Recent developments in the field of semiconductor memory devices have made it possible to design a digital temperature indicator using read-only memories (ROMs) with accuracies approaching those attainable with computer based techniques. The speed, simplicity, and reliability of such a ROM technique would be better than any other presently available technique. Predicted trends in future semi-conductor developments also make such a technique very attractive economically. A computer program has been developed to generate ROM truth tables for a nonlinear response system. The truth tables generated by this program are based on the system of quadratic equations which describe the nonlinear function. The accuracies are limited only by the inherent accuracy of the equations used, the size of the ROM, and the dynamic range of the stimulus to be measured.