Digital Commons Network^™

An Optical Adaptive Resonance Neural Network Utilizing Phase Conjugation, Donald C. Wunsch, D. J. Morris, T. P. Caudell, R. A. Falk

Electrical and Computer Engineering Faculty Research & Creative Works

An adaptive resonance (ART) device has been conceived that is fully optical in the input-output processing path. It is based on holographic information processing in a phase-conjugating crystal. This sets up an associative pattern retrieval in a resonating loop utilizing angle-multiplexed reference beams for pattern classification. A reset mechanism is used to reject any given beam, allowing an ART search strategy. The design is similar to an existing nonlearning optical associative memory, but it does allow learning and makes use of information the other device discards. This device is expected to offer higher information storage density than alternative ART implementations.

Fast Power Flow With Capability Of Corrective Control Using A Neural Network, Badrul H. Chowdhury, B. M. Wilamowski

Electrical and Computer Engineering Faculty Research & Creative Works

The authors present a number of different configurations of a neural network and identify a particular case which is most suitable for power flow analysis in real-time applications. The advantage of fast computation of the artificial neural network (ANN) is used for obtaining power flow solutions in real time. The inputs to the ANN are the real and reactive power generating and demand in the system, and the output data are the complex bus voltages. A few configurations of the neural network were experimented with, and the best results were achieved with a single-layer feedforward neural network with nonlinear feedback. …

New Emc Design Guidelines For Multilayer Printed Wiring Boards, Todd H. Hubing, Thomas Van Doren

Electrical and Computer Engineering Faculty Research & Creative Works

No abstract provided.

A New Method For Calculating The Economic Benefits Of Varying Degrees Of Power Factor Correction For Industrial Plant Loads, Mohammed Ishaque

Dissertations and Theses

A comparative study of the economic benefits that can be obtained from different degrees of power factor correction for medium and small scale industrial installations is shown. A new approach for precise calculation of kws and kvars required at different power factors is presented. These calculated values are used to find the return on investments for the capacitors needed for power factor correction. The developed method is easy to use, cost effective, accurate and will help electrical engineers with minimum knowledge of power systems to precisely determine the savings available by improving the power factor of an industrial load.

Error Detection And Correction In Compressed Data, Nadir Sezgin

Theses

Encoded data is very sensitive to the channel errors. Especially if the data is compressed by Arithmetic Encoding procedure, then the error propagation is very high. The error propagation in Arithmetic Coding is studied. Exploiting the high error propagation property when compressing data by Arithmetic encoding procedure, two different algorithms have been proposed for error detection and correction. Under certain conditions these algorithms detect and with a very high probability correct the errors introduced to the compressed data.

A Graphical Tool For The Simulation Of Timed Petri Nets, Javaid Aslam Siddiqi

Theses

A highly interactive graphical tool is developed for the drawing and simulation of discrete event systems involving time constraints. The Timed Petri Net Simulator tool (TPNS) is capable of simulating systems with deterministic and stochastic delays. Performance and utilization parameters are collected during simulation for behavior and analysis purposes. Simulation results are stored in different files (log, verify, utilize, and mark). Conflict resolution on the basis of priority or random selection of multiple transitions in conflict is implemented to handle complex modeling situations. A user friendly interface is developed for effective interaction between the tool and the user. This interface …

Simulation And Modeling Of Kinetics Of Silicon Oxidation In The Thin Oxide Regime, Tirthankar Dutta

Theses

Thermal oxidation of Silicon in dry O₂, in the thin regime(< 500A) is of vital importance to VLSI device engineers, because thin layers of SiO₂ are exclusively used as gate dielectric for high performance of MOS devices. There exist a number of models to explain the kinetics of oxidation in this thin regime. However there is considerable variance among them and the reported rate constants, which are supposed to describe the oxidation process. Rather than arriving at an alternative model, the present study aims at an extensive study and simulation of existing models of oxidation in dry oxygen, in the thin regime, with a recent set of experimental data and …

Design Of An Advanced Control Algorithm For A Nuclear Power Plant Feedwater Control System, Randall C. Ezzo

Theses

Electric power companies with nuclear power plants have been aggressively looking for ways to increase plant safety and availability by increasing the reliability and performance of the plant's control systems. The performance of the Oyster Creek Nuclear Station's feedwater control system can be significantly improved by replacing the proportional-integral compensator with a compensator based on modern control concepts. In response to a -10 inch reactor water level setpoint change, settling time is improved from 107 seconds with 3% overshoot to 34 seconds with no overshoot.

Computer Aided Design Of A Broadband Microwave Power Divider, Jaideep Mukherjee

Theses

The purpose of this thesis to provide alternative design procedure for a power divider that has broadband characteristics. The Wilkinson power divider consisting of cascaded quarter wave sections is known to increase the bandwidth while maintaining high isolation and good match conditions at individual ports. Resistors placed between the quarter wave sections served to increase isolation. The values of required components were determined using odd and even mode analysis. In this work, the approximate theory developed previously is used as a preliminary design model, and it is extended further by considering optimization techniques and coupling between adjacent arms. This novel …

Extension And Applications Of Knitting Technique For Petri Net Synthesis, Yanzhong Zhang

Theses

Petri nets have been used for modeling and analyzing various systems [1,2,3,4,8]. One approach to Petri net synthesis is a knitting technique [4,5, 6,7]. It was developed for modeling concurrent systems. The Petri net model is incrementally expanded by adding paths of nodes according to rules. To extend the knitting technique to cover synchronized systems, a set of new rules based on synchronic distance concept is presented. The rules penult synchronizing M exclusive processes (M 2) to a synchronic distance of any positive integer N. They can be applied to both exclusive processes and independent processes. The resulting Petri net …

A Color Petri Net Simulation Tool, Sanjaykumar V. Desai

Theses

A user friendly, modular and highly interactive graphical tool for drawing, editing and simulating color Petri nets is developed. The tool is designed to model concurrent and interlocked systems, such as flexible manufacturing systems, communication protocols and computer systems. A Color Petri Net(CPN) has a more concise graphical system representation than a standard Petri net, because net entities in CPN carry information. The CPNS(Color Petri Net Simulation) tool simulates a net having conflicts, reveals deadlocks, and creates a log file with the state history of the system yielding performance measures.

An Implementation Of Bionomial-Qmf For Subband Coding Of Speech, Feng Chen

Theses

A real-time subband speech codec is implemented by the author on TMS320C25 digital signal processor based hardware for speech compression down to 0.25 bit per sample.

The Binomial Quadrature Mirror Filters (B-QMFs) are employed as the basic block for the hierarchical, dynamic tree, analysis/synthesis filter bank. After the analysis, available bits are optimally allocated among subbands. Band statistics are updated to overcome the nonstationarities of speech signal.

The energy compaction performance of this codec is compared with different subband and DCT in simulations for speech signals and AR(1) source models.

It is shown that the speech codec implemented provides an …

Vision-Guided Servoing Robot Arm System, Shing Lee

Theses

This thesis investigate the problem of tracking 2-D circle parts with an eye-in-hand robot arm system. A vision module capable of measuring the motion of a circle is developed. Two algorithms based on the conic equation representation of circle and the symmetric property are developed. The first algorithm approximates the image of the circle as an ellipse. The second algorithm use the arc radius estimator to extract two image points of the circle located at the center of the largest arc. These image points are then used to solve for the inverse perspective transformation. An inverse kinematics solution based on …

Spectral And Spatial Quantum Efficiency Of Algaas/Gaas And Ingaas/Inp Pin Photodiodes, Steven Alan Tabor

Dissertations and Theses

This thesis reports a novel system capable of testing both the spectral responsivity and the spatial quantum efficiency uniformity of heterostructure photodiodes using optical fiber coupled radiation. Testing was performed to confirm device specifications. This study undertakes to quantify the spectral bandwidth of an AlGaAs I GaAs double heterostructure photodiode and two InGaAs I InP double heterostructure PIN photodiodes at D.C., through the use of spatial scanning. The spatial scanning was done using lasers at 670 nm, 780 nm, 848 nm, 1300 nm, and 1550 nm, coupled through singlemode optical fiber. The AlGaAs I GaAs material system covers the 600 …

Composite Damage Assessment Employing An Optical Neural Network Processor And An Embedded Fiberoptic Sensor Array, Barry G. Grossman, Xing Gao, Michael H. Thursby

Electrical Engineering and Computer Science Faculty Publications

This paper discusses a novel approach for composite damage assessment with potential for DoD, NASA, and commercial applications. We have analyzed and modeled a two dimensional composite damage assessment system for real-time monitoring and determination of damage location in a composite structure. The system combines two techniques: a fiberoptic strain sensor array and an optical neural network processor. A two dimensional fiberoptic sensor array embedded in the composite structure during the manufacturing process can be used to detect changes in the mechanical strain distribution caused by subsequent damage to the structure. The optical processor, a pre-trained Kohonen neural network, has …

Investigation Of The Effects Of Aeroelastic Deformations On The Radar Cross Section Of Aircraft, Samuel D. Mckenzie

Theses and Dissertations

The effects of aeroelastic deformations on the radar cross section (RCS) of a T-38 trainer jet and a C-5A transport aircraft are examined and characterized. Realistic representations of structural wing deformations are obtained from a mechanical/computer aided design software package called NASTRAN . NASTRAN is used to evaluate the structural parameters of the aircraft as well as the restraints and loads associated with realistic flight conditions. Geometries for both the non-deformed and deformed airframes are obtained from the NASTRAN models and translated into RCS models. The RCS is analyzed using a numerical modeling code called the Radar Cross Section - …

Incorporation Of Differential Global Positioning System Measurements Using An Extended Kalman Filter For Improved Reference System Performance, William J. Negast

Theses and Dissertations

To quantify the performance abilities of existing or proposed navigation systems, the U.S. Air Force has, for the last several years, compared the performance of the system under test to the performance of a baseline navigation system known as the Completely Integrated Reference Instrumentation System (CIRIS). CIRIS obtains a highly accurate navigation solution by combining information from three major subsystems: inertial navigation system (INS) information, barometric altitude information, and range and range-rate information from ground transponders which have been precisely surveyed. Although the navigation solution produced by CIRIS is highly accurate, it will soon be inadequate as the standard against …

Function Prediction Using Recurrent Neural Networks, Randall L. Lindsey

Theses and Dissertations

A fully recurrent neural network was applied to the function prediction problem. The real-time recurrent learning (RTRL) algorithm was modified and tested for use as a viable function predictor. The modification gave the algorithm a variable learning rate and a linear/sigmoidal output selection. Verifying the networks ability to temporally learn both the classic exclusive-OR (XOR) problem and the internal state problem, the network was then used to simulate the frequency response of a second order IIR lowpass Butterworth filter. The recurrent network was then applied to two problems: head position tracking, and voice date reconstruction. The accuracy at which the …

Simulation Of Finite-Precision Effects In Digital Filters, Perry L. Choate

Theses and Dissertations

This thesis develops and demonstrates software that models finite- precision effects in digital filters. Coefficient quantization and internally generated roundoff noise effects are simulated by rounding coefficients and multiplier output values to (B+1) bits. Outputs include magnitude and phase responses that may be plotted to show results of coefficient quantization. The output noise power due to roundoff is calculated. Three types of digital filter structures are examined. The first type is direct form I implementation for FIR and IIR structures, second is direct form II, and third is cascading the canonic direct form II second-order sections for higher order filter …

Application Of Nonlinear Qft To Flight Control Design For High Angle Of Attack Maneuvers With Thrust Vectoring, Steven J. Rasmussen

Theses and Dissertations

Nonlinear Quantitative Feedback Theory (QFT) is applied to the design of an flight Control system (FCS) to control maneuvers of an agile aircraft. To do this, maneuvers are chosen that are desirable during within visual range combat. Plant transfer functions are developed through the use of system identification, using the input/output time histories of the maneuvers. During the system identification process the q sub ij(o) are found directly and are constrained to be minimum phase. After identifying the plants, the cascaded multiple-input multiple-output QFT technique is used to design the FCS. During the design of the FCS, stability boundaries are …

A Rigorous Utd Analysis Of Electromagnetic Scattering From Resistive Strips And Resistive-Loaded Conducting Strips, Douglas H. Casanova

Theses and Dissertations

This thesis investigates electromagnetic scattering from resistive strips and resistive-loaded conducting strips using a rigorous UTD formulation. The UTD diffraction coefficients are based on the Wiener-Hopf technique and Jones' method. Scattering predictions are performed for constant resistive strips, constant resistive-loaded conducting strips, tapered resistive strips, and tapered resistive-loaded conducting strips. All strip geometries have a total width of 4 lambda. Predictions are compared to method of moments (MM) and measurements to determine the validity of the UTD prediction. Overall, there is good agreement. For TMz polarization the only deviations that occur with MM are near edge-on when the degree of …

Binaural Sound Localization Using Neural Networks, Rushby C. Craig

Theses and Dissertations

The purpose of this study was to investigate the use of Artificial Neural Networks to localize sound sources from simulated, human binaural signals. Only sound sources originating from a circle on the horizontal plane were considered. Experiments were performed to examine the ability of the networks to localize using three-different feature sets. The feature sets used were: time-samples of the signals, mena Fast Fourier Transform magnitude and cross correlation data, and auto-correlation and cross correlation data. The two different types of sound source signals considered were tones and gaussian noise. The feature set which yielded the best results in terms …

A Time-Dependent Adaptive Filter For Cochannel Interference Reduction, Matthew H. Foster

Theses and Dissertations

This thesis presents a Time Dependent Adaptive Filter (TDAF) which exploits the cyclostationarity of digitally modulated communications signals and seeks to improve the Signal to Interference Ratio (SIR) and Signal to Noise Ratio (SNR) of such signals. The TDAF is embedded in a computer simulation of a simple communication system consisting of a data source, data formatter, pulse shaping filter, BPSK modulator, and demodulator. In the simulation the TDAF and a Time Independent Adaptive Filter (TIAF) attempt to extract the Signal of Interest (SOI) from noise or interference. the criteria of Mean Squared Error (MSE) is used as the primary …

A Rigorous Uniform Diffraction Analysis Of The Electromagnetic Scattering From Impedance Edges And Junctions, Joseph C. Fortney

Theses and Dissertations

This paper investigates the scattering from impedance strips and impedance-loaded conducting strips. The impedance strips are analyzed using Senior's impedance half plane formulation. Once the primary diffraction from the impedance half plane is presented, it is used to develop multiple diffraction mechanisms on an impedance strip. The scattering from impedance-loaded strips are analyzed using Maliuzhinets' impedance wedge formulation. The primary diffraction mechanism from an impedance wedge is used to develop the multiple diffractions on an impedance double wedge. The multiple diffractions on both types of strips are developed using the Extended Spectral Ray Method. Sample calculations are made for impedance …

Gabor Segmentation Of High Resolution Synthetic Aperture Radar Imagery, Michael A. Hazlett

Theses and Dissertations

This thesis investigates the use of Gabor filters and a radial basis function (RBF) network for segmentation of high resolution (1 foot by 1 foot) synthetic aperture radar (SAR) imagery. Processing involved correlation between the SAR imagery and Gabor functions. Two methods for selecting the optimal Gabor filters are presented. This research used complex Gabor functions and operated on single polarization HH complex data. Following the selection f the proper Gabor filters, correlation coefficients for each image were calculated and used as features for the RBF network. Provided are results demonstrating how Gabor processing and a RBF network provide image …

Wave Propagation In A Randomly Inhomogeneous Medium - A Study Of The Problem, Kyle Hunter

Theses and Dissertations

The problem of calculating the scintillation index of an atmospherically propagating spherical wave is examined. The fourth statistical moment of the complex field is obtained by using Feynman (path) integral techniques applied to the stochastic parabolic equation. The general trajectory of each Feynman integral is approximated by a truncated Fourier-sine series and the infinite-fold integration of the Feynman integral is reduced to a three-fold Riemann integral which is shown to match results derived under different assumptions. This thesis is highly tutorial.

Comparison Of Afitpac Versus Nos And A Packet Radio Network Design, Thomas A. Iannelli

Theses and Dissertations

This thesis encompasses two themes: (1) A comparison of the Network Operating System (NOS) software package, developed by Phil Karn and others, and AFITPAC, written by Marsh and Geier at the Air Force Institute of Technology. The comparison is based on the Air Force Logistics Command's requirements for the development of a Packet Radio Network program. The results of the comparison are a recommendation that use of the NOS package be pursued because it can meet all twelve of the requirements and compiles with five of the Military Standard Protocols. (2) The design of a personal computer Packet Radio Network …

An Investigation Of A Design For A Finite-Difference Time Domain (Fdtd) Hardware Accelerator, James R. Marek

Theses and Dissertations

This study investigated increasing the speed of Finite-Difference Time Domain (FDTD) cell calculations through a special purpose architecture using Very Large Scale Integration (VLSI). These FDTD cell equations model inhomogeneous, isotropic, lossy magnetic and dielectric materials. Special attention was given to simplicity and performance, using the fastest components generally available in AFIT VLSI programs, while attempting to minimize component count. A VHSIC Hardware Description Language simulation of the proposed chip established design feasibility and provided performance estimates: 350 ns to generate the first cell value, 200 ns thereafter (30 MFLOPS maximum double-precision). This study also implemented boundary conditions in hardware …

Analysis Of Visual Illusions Using Multiresolution Wavelet Decomposition Based Models, John S. Laing

Theses and Dissertations

This thesis provides alternatives to the explanation that spatial filtering is responsible for the perception of illusory contours in the Kanisza Triangle illusion. Specifically, we use a Multiresolution Wavelet Decomposition to divide an image into spatial-frequency bands that are used as inputs to three biologically motivated models. The thesis includes a brief tutorial of Wavelet theory and an in-depth explanation of our implementation of recently published algorithms for Multiresolution Wavelet Analysis. The first model is based on the saccadic movements of the human eye. It demonstrates the importance of the high spatial-frequency content of an image in the formulation of …

Design And Testing Of A Lightweight, Planar Microwave Lens, Paul M. Proudfoot

Theses and Dissertations

This thesis documents the design, fabrication, and testing of a lightweight, planar microwave lens. The lens consists of two planar arrays of lightweight printed circuit antennas interconnected by microstrip lines whose length varies as a function of position on the lens. Microwave energy is coupled from one face of the lens to the other by means of a slot etched in the ground plane. At 8.0 GHz the lens is ten wavelengths across, is less than 1/8 inch thick, and weighs less than 4 pounds. The lens can scan a beam over a ± 30 degree region in both the …