Digital Commons Network^™

An Artificial Neural Approach To The Decomposition Problem, Chandrashekar L. Masti

Electrical & Computer Engineering Theses & Dissertations

The goal of this thesis is to develop an artificial neural approach toward addressing the intractability involved with the decomposition problem. The search for the lattice of substitution property (s. p.) partitions essential to decompositions is cast into the framework of constraint satisfaction. An artificial neural network is developed to provide solutions by performing optimization of a mathematically derived objective function over the problem space. The issue of transitivity is verified to belong to a class of problems beyond the scope of solvability for conventional quadratic-order constraint satisfaction neural networks. A theorem is stated and proved establishing that third-order correlations …

Visual Speech Training Aid For The Deaf, Subhashri Venkat

Electrical & Computer Engineering Theses & Dissertations

A computer-based vowel articulation training aid has been developed. A "continuous" acoustic-phonetic transformation is performed to map speech parameters to a lower dimensionality display space. There are two possible approaches to this transformation problem. The transformation could be either linear or a combination nonlinear/linear. The nonlinear transformation is performed using a multi-layered feedforward neural network with linear output layers. Speech parameters are extracted either from an analog filter bank arrangement (band energies) or by a digital signal processing procedure (Discrete Cosine Transform Coefficients). The speech parameters obtained from both methods correspond to the spectral envelope of the speech signals. The …

Encoding Phonetic Knowledge For Use In Hidden Markov Models Of Speech Recognition, Danming Qian

Electrical & Computer Engineering Theses & Dissertations

Hidden Markov models (HMM's) have achieved considerable success for isolated-word speaker-independent automatic speech recognition. However, the performance of an HMM algorithm is limited by its inability to discriminate between similar sounding words. The problem arises because all differences between speech patterns are treated as equally important. Thus the algorithm is particularly susceptible to confusions caused by phonetically-irrelevant differences. This thesis presents two types of preprocessing schemes as candidates for improving HMM performance. The aim is to maximize the differences between phonologically-distinct speech sounds while minimizing the effect of variations in phonologically-equivalent speech sounds. The preprocessors presented are a discrete cosine …

Improving Solids Conveying In Plasticating Extruders : An Investigation Into Manufacturing Processes And Manufacturing Systems For Profiled Thermoplastic Pellets, Oommen Cherian

Theses

The objective of the thesis is the manufacture of thermoplastic bilobal and trilobal profiled pellets for the determination of the interparticulate co-efficient of friction exhibited by solid beds of such pellets. The enhancement of the interparticulate coefficient of friction is expected to lead to ameliorated solids conveying rates in plasticating extruders.

Predicting the product shape from a profile die is very difficult due to the elastic memory of plastics and the complex boundary conditions. Consequently the cost involved in experimenting with different profile dies is high. Therefore a standard die with several die inserts of varying cross-sections were used to …

Small Disturbance Stability As Affected By Excitation Controls, Alfredo Flores

Theses

The effect of excitation control systems on the power system's small disturbance stability limits has been investigated in this thesis. Three excitation control schemes namely the proportional, fast?acting and the modified fast?acting have been studied.

The stability results were predicted directly in terms of the AVR gain controlling parameters following one?dimensional and two?dimensional frequency domain separation technique. The necessary algorithms and computer simulation programs have been developed.

In case of a modified fast?acting excitation control with three unknown variables, a procedure has been proposed to obtain the stability results by extending the two?dimensional frequency domain separation technique.

Coke Oven Emissions: A Case Study Of Technology-Based Regulation, John D. Graham, David R. Holtgrave

RISK: Health, Safety & Environment (1990-2002)

After examining and analyzing the experience with coke ovens, the authors conclude that attempts to force technology beyond its demonstrated competence can be both expensive and ineffective in controlling hazards. They also suggest implications for pending proposals to further control air pollution.

Computer Aided Localization Of Neurological Diseases, Yusuf Parlar

Dissertations

Computers in medicine has brought new dimensions and better understanding of uncertainties in the field of medical sciences in the last decade. This thesis is focused upon the localization of neurological lesions in the human nervous system. It relates clinical neurological test outcomes to pathways of function or malfunction. Certain methods are proposed-empirical. stochastic. Deterministic-to estimate the spatial distributions of lesion probabilities.

First. a Bayesian model is presented to estimate the posterior probability of lesion from a priori information, based on the test outcomes. Due to unavailable data alternative methods and models are presented: Regression Analysis. Monte Carlo simulation, and …

On The Performance Of Minimum Redundancy Array For Multisource Direction Finding, Byung-Seub Lee

Dissertations

As an application of power spectrum estimation, the multi-source direction finding has been evolved from conventional FFT method to Superresolution methods such as Multiple Signal Classification(MUSIC) algorithm. Uniform Regular Array(URA) was mainly used in all these approaches.

The Minimum Redundancy array(MRA); a non-uniform thinned array which results in an input signals covariance matrix with minimum redundancy has been shown to have certain interesting properties for spectrum estimation. Only recently it was suggested to use the MRA for spatial estimation. The purpose of this research was to study the performance of this array in multi-source direction finding estimation and compare it …

Computer Modeling And Analysis Of Biological Rhythms, Shi Xiong Yang

Dissertations

Biological rhythms are an important phenomenon and feature of physiologic systems. Indirect means have to be employed for their description and exploration due to the unclear internal nature of the system. This study analyzed and developed several possible mathematical models using single or multidimensional nonlinear differential equations to approach the experimental circadian data. The numerical solutions of the models were obtained by computer simulation and the simulated and experimental acquired circadian data were analyzed in both the time and frequency domains. Phase plane plots, phase response curves and power spectrum analysis were employed to determine the nonlinearity of the system …

An Artificial Neural Network For Redundant Robot Inverse Kinematics Computation, Wibawa Utama

Theses

A redundant manipulator can be defined as a manipulator that has more degrees of freedom than necessary to determine the position and orientation of the end effector. Such a manipulator has dexterity, flexibility, and the ability to maneuver in presence of obstacles. One important and necessary step in utilizing a redundant robot is to relate the joint coordinates of the manipulator with the position and orientation of the end-effector. This specification is termed as the direct kinematics problem and can be written as x = f(q)
where x is a vector representing the position and orientation of the end-effector, q …

Multiprocessor System Design Tutor : Expert System Approach, Rakesh Kamdar

Theses

To increase computational bandwidth and system resilience, integration of several microprocessors in a single system becomes necessary. The overall throughput and efficiency of such a system is directly dependent on the hardware and software interconnection supported by the basic microprocessor chip. Sometimes it becomes difficult to put together all the information for design criteria and all the design related formulas.

The approach made here is to continuously update the hardware and software information in the database related to a given microprocessor. This information can be accessed at any time for efficient design solution. Intel 80386 and Motorola 68020 microprocessors are …

A Study Of The Effects Of Weighting Matrices In Power System State Estimation, Yanshi Wang

Theses

The effects of weighting matrices that may be used in the Weighted Least Squares (WLS) method when applied to the problem of Power System State Estimation (PSSE) are studied. These weighting matrices include the one usually used in the literature and the one used in the utility PSSE programs

The performances of these weighting matrices are compared using the Monte Carlo simulation technique.

To test the performances of the weighting matrices, a PSSE program is developed, different performance indices for state comparison are defined, and many cases are studied.

Results show that the weighting matrix usually used in literature, 1/σ², …

Design Of A Real Time Optical Sensor For Continuous Blood Cell Separation, Girish S. Yerande

Theses

For continuous measurement and display of oxygen saturation and RBC content in blood during cell separation a reflection type optical sensor is designed based on the theoretical model using the photon diffusion theory. The prototype sensor consists of two arrays of light emitting diodes, red and infrared, and photodiodes. The emission wavelength of LEDs are 660 nm and 800 nm. The 800 rim which is in the infrared region is very close to isobestic value for blood. The prototype sensor is mounted on the blood cell separator equipped with an antitwister mechanism. Each LED emits a specific wavelength of light …

Non-Parametric Dominant Point Detection Algorithms, Kuo-Wei Huang

Theses

Dominant point detection is one of the most important preprocessing steps for landmark-based shape recognition and point-based motion estimation. Two new methods for detecting dominant points are presented. Both methods do not require any input parameter and the dominant points obtained by these methods remain relatively the same even when the object curve is scaled or rotated.

In the first method, for each boundary point, a support region is assigned to the point based on its local properties. Each point is then smoothed by a Gaussian filter with a width proportional to its determined support region. A significance measure for …

Optimum Power System Planning, Youn Tae Kyong

Theses

This thesis presents analysis method of optimum power system planning. The Gauss-Seidel iteration method is used to calculate load flow data of power system. The real and reactive powers, voltage magnitude and phase angle at each bus were obtained using Gauss-Seidel iteration method for a given IEEE sample power system. The load flow calculations were repeated under 60 percentage, 70 percentage, 80 percentage, 90 percentage, 100 percentage loading condition.

The transformation of reference frames were performed to simplify given IEEE sample power system identification. The transformation of reference frame transforms real and reactive voltage, current and impedance variables(reference frame 1) …

Image Data Compression With Neural Networks, Jianjun Li

Theses

Three efficient image coding schemes, based orb neural networks, have been developed in this thesis: (1) Neural network vector quantization (NNVQ). The main advantage of this new technology is that it can accomplish the complex encoding process much faster than the previous algorithms. Its properties are studied and demonstrated by simulations. (2) A adaptive NNVQ for image sequence coding. Simulation experiments have been carried out with 4 x 4 blocks of pixels from an image sequence consisting of 40 frames. At 0.67 bits/pixel, this scheme achieves good image quality suitable for videoconferencing systems; (3) Neural network prediction. This new prediction …

Trellis Coded Tamed Frequency Modulation In Rayleigh Fading Channels, Kiran R. Patel

Theses

It is usually difficult to reduce the effects of channel impairements on transmitted information. A fading channel is one such media in which digital communications system operates. A signal is subject to distortion in amplitude and phase when propagating towards the receiver. This puts constraint in required SNR at the receiver, for given performance. Various modulation/coding and demodulation/decoding techniques are employed to lessen the effects of fading. Bit error rate is a performance parameter that sets the criteria for signal design, system architecture and performance.

Here, a slow Rayleigh fading channel with three propagation paths possesing a scattering function of …

Multi-Sensor System For An Articulated Robot Hand, Narendra Sucahyo

Theses

The precision and dexterity of a manipulator hand critically depends on the proper extraction and processing of sensory information. This thesis describes how a sensor system equipped with fingertip sensors and strain gauges can provide proper feedback to drive the grasp kinematics of an articulated robot hand. The grasp kinematic that provide the manipulator hand with precision and dexterity is based on impedance control. The fingertip sensors and the strain gauges are constructed of piezoresistive materials for force transduction. Through optimization, based on the sensitivity and capability of the sensors, a synergism of force and contact information can be obtained …

A Study Of Signal Cancellation Effects In Adaptive Arrays, Cheng-Der Sun

Theses

An adaptive array can automatically place pattern nulls in directions from which undesired signals (interference, jamming, or clutter, etc.) arrive and can also provide gain for a desired signal. The purpose of this study was to study the effects of the adaptive array on the desired signal while suppressing the undesired signal.

This was done by studying the transfer characteristics of the adaptive array between its input and output. In this study, C.W. jammers along with the desired signal were considered and input and output relationship of the adaptive array were determined. Several basic types of adaptive arrays were studied …

Automatic Infrared Inspection Of Solder Joints On Mounted Circuit Boards, Ken Kekuan Zhou

Theses

This thesis reports the research work on automatic infrared and visual light inspections of solder joints on printed circuit boards (PCBs) utilizing an image processing system and statistical pattern recognition techniques.

A small scale image processing system, consisting of a SUN workstation and several DATACUBE modules, is set up for the research. Software packages for imaging, displaying and solder joint inspection are developed and ready for public usage.

The capability of infrared inspection for solder joints on PCBs is demonstrated. The fusing of infrared and visual light imagery for the inspection is shown to offer better results than that obtainable …

Eeg Mapping Using The Personal Computer, Zhuan-Zhang Pei

Theses

An approach for brain electrical activity mapping using a personal computer has been investigated to enhance the visibility of the EECs. An automatic procedure from data acquisition to the final mapping has been developed for on-line or off-line data processing, in which a table-look-up technique has been introduced to eliminate redundant processing. The power spectra in different EEC frequency bands for four electrodes are coded into 15 colors and mapped on a color monitor. The data correspondings to the areas between electrodes are interpolated by the surface spline method. The mapping speed and smoothness has been optimized. An experiment has …

An Implementation Of Mht Algorithm For Adaptive Transform Coding Of Speech, Peter Battenschlag

Theses

The Modified Hermite Transform (MHT), a suboptimum block transform scheme, is implemented on a TMS320C25 signal processing system with the objective of reducing the redundancy in speech signals and transmit a quantized vector representation at bitrates as less as 0.5 bits per sample.

The generation of the Hermite polynomials with a binomial and several first order IIR filter structures is presented focusing on the fact that no muliplications are required which simplifies the MHT algorithm considerably.

An assembler program performs forward and inverse transformation together with a vector quantization using adaptive bitallocation based on short-time statistics for the transform coefficients. …

A Theoretical Model For Radio Attenuation Inside Buildings, Kyu-Sung Kim

Theses

Research to expand the capability of indoor, portable, radio frequency links for voice and data transmission is being pursued worldwide. Sophisticated communication systems using digital portable radio telephones have been proposed. To accomplish this goal, it is necessary to ascertain the attenuation of electromagnetic signal propagation inside buildings. Because modeling of the indoor environment at radio frequencies is very complicated, most publications document experimental results. In this study,. a theoretical model is proposed for predicting path loss inside buildings. The theory involves a waveguide model of the indoor environment which permits a rigorous modal solution similar to that developed by …

Temperature Dependent Current-Voltage Characteristics In Thin Sio2 Films, Jin Zhao

Theses

An analysis of the Fowler-Nordheim tunneling (FNT) theory and its application to temperature dependent current-voltage characteristics, of very thin films of SiO₂ on silicon, is presented. The final results are believed to provide the most complete examination of FN emission theory and predict the breakdown electric field in thin SiO₂ films. The role of the roughness, at the Si-SiO₂ interface, in determining the FNT current and breakdown mechanisms in metal-oxide-semiconductor (MOS) structures are also discussed.

An Effective Cube Comparison Method For Discrete Spectral Transformations Of Logic Functions, Ingo SchäFer

Dissertations and Theses

Spectral methods have been used for many applications in digital logic design, digital signal processing and telecommunications. In digital logic design they are implemented for testing of logical networks, multiplexer-based logic synthesis, signal processing, image processing and pattern analysis. New developments of more efficient algorithms for spectral transformations (Rademacher-Walsh, Generalized Reed-Muller, Adding, Arithmetic, multiple-valued Walsh and multiple-valued Generalized Reed- Muller) their implementation and applications will be described.

Spontaneous Coherent Pulsations In Standing-Wave Laser Oscillators: Simplified Models, Lee W. Casperson, Manuela Fe Tarroja

Electrical and Computer Engineering Faculty Publications and Presentations

A series of simplifications and approximations is introduced into a recently described model for spontaneous pulsations in standing-wave laser oscillators. To the extent that these simplifications are valid they can lead to significant reductions in computation time and sometimes also to a better understanding of the relative importance of various physical effects. Of special interest is the number of spatial harmonics required to represent adequately the effects of longitudinal spatial hole burning. Other approximations investigated include neglect of spectral cross relaxation, neglect of multiple-energy-level equations, neglect of electric-field derivatives, and neglect of polarization derivatives. In addition, a detailed discussion is …

Analysis And Development Of A Switched Capacitor Power Supply, John Richard Hackworth

Electrical & Computer Engineering Theses & Dissertations

The combination of a switched capacitor network and switching power supply techniques can result in a transformer less regulated DC power supply that is small, relatively simple, and efficient. In addition, the output of such a system could be operated over a wide range of common mode voltages. Multiple switched capacitor supplies could be stacked to achieve voltage multiplication.

Although the output of a system of this type is a continuous function, the internal operation must be analyzed in finite time elements. This numerical analysis will be shown in both a mathematical approach and a computer model.

Effects Of Bandwidth‐Limiting Tuning Elements In Synchronously Pumped Mode‐Locked Lasers, Bahram Zandi, Lee W. Casperson, Duncan Leo Macfarlane

Electrical and Computer Engineering Faculty Publications and Presentations

A description of bandwidth‐limiting tuning filters is introduced into a semiclassical model for synchronously pumped mode‐locked dye lasers. The finite phase memory of the molecular wave functions is included as are the isotropic molecular distribution and the finite vibrational relaxation times. The new set of equations has been solved numerically using the best available values for the various parameters. The results have been compared with experimental data obtained using a rhodamine 6G dye laser, which is synchronously pumped using an acousto‐optically mode‐locked argon laser. Tuning element effects have been studied using two‐ and three‐plate birefringent filters and a tuning wedge, …

Adaptive Fir Digital Filter Design And Simulation In The Noise Reduction System, Jack Jiabing Pang

Theses

The adaptive FIR digital filter is designed to suppress the noise in the tape recording and audio communication systems. The cutoff frequency of the digital FIR filters can be varied adaptively depending on the energy level of the input signal. In audio communication, especially, in tape recording systems, the noise obviously gets much bigger, when the frequency band of the system increases. The noise reduction system makes use of a phenomenon called masking, where the big signal-to-noise ratio makes human ears not sensitive to the noise. In a constant interval time, the system detects the signal level and compares the …

Sputter Deposition And Characterization Of High Tc Superconducting Thin Films, Vipulkumar Patel

Theses

Preparation of thin superconducting films by direct RF sputter deposition on Si, MgO, SrTiO₃ and quartz substrates and silver and Sl3N4 buffer layers (deposited on silicon substrates) was studied. To study resistance versus temperature characteristics of thin films and bulk materials, a computerized measurement system was developed. 1-1.5 pm thick YBaCuO films deposited on Si₃N₄/Si were found to be insulating with room temperature resistance higher than 236KΩ, while films with Ag as a buffer and/or a passivating layer exhibited metallic behavior with room temperature resistances lower than 700mΩ. The area directly under the 2.25 inches …