Electrical and Computer Engineering Commons^™

Efficient Multiprocessor Scheduling By Mean Field Theory Neural Networks, Pei-Ken Yi

Theses

In this thesis, we develop an optimization method based on Mean Field Theory (MFT) Neural Networks to solve the Task Scheduling problem. The MFT algorithm combines characteristics of the Simulated Annealing (SA) algorithm and the Hopfield neural network. MFT exhibits rapid convergence and at the same time it preserves the solution quality afforded by SA. Since MFT has been successfully used to solve the Traveling Salesman Problem (TSP), a new modification to MFT is also presented which supports Task Scheduling problem. The temperature behavior of MFT during Task Scheduling is approximately analyzed and shown to possess a critical temperature (71,) …

Integrated Digital Speech System With Pc, Zhen Zhu

Theses

A PC-based real-time digital speech processing system is designed. Speech synthesis, analysis, storage and retrieval are implemented in the system with intelligent telephone functions. Speech data are stored on floppy disk and easy to retrieve. Though floppy disk is slow in speed, real-time processing is achieved with low bit rate coding. The computer gets speech data from a digital speech processing unit or send data to it through a synchronous seiial communication port. The analog input of digital speech processing unit can be selected from the telephone line or the microphone, and output of the digital speech processing unit can …

Uniformity Correction For Infrared Ir-Ccd Camera, Chien-Ming Leng

Theses

This thesis describes the results of a study of three types of uniformity correction schemes for an infrared camera. More specifically, a Datacube Image System was used to simulate the three uniformity correction schemes for a 320x244-element IR-CCD TV camera with PtSi Schottky-barrier detectors. The three uniformity correction schemes are: one-point uniformity correction with single background, one-point uniformity correction with multibackground and two-point uniformity correction. Datacube programs were developed for the three above uniformity correction schemes and some experimental results were obtained.

The Bandwidth Allocation Of Atm Through Genetic Algorithm, Hong Pan

Theses

In this thesis, we develop an optimization method based on Genetic Algorithm to achieve the bandwidth allocation of ATM network. In Broadband Integrated Service Digital Network (BISDN), the Asynchronous Transfer Model (ATM) is already adopted as the transfer standard by CCITT. ATM is a high-bandwidth, low-delay, fast-packet switching and multiplexing technique. Using ATM technique, we can flexibly reconfigure the network and reassign the bandwidth to meet the requirement of all types of services. As an effective optimization method, Genetic Algorithm(GA) is applied to implement the bandwidth allocation of ATM. GA is a class of search algorithm based on the mechanics …

Dual Token Ring Lan Performance, Pratap S. Pasupuleti

Theses

A simulation model to analyze the performance of a dual token ring LAN is developed. The performance measures throughput, average packet delay and maximum delay per packet are obtained. System performance is analyzed for different loads, number of stations, and packet lengths. The performances of token ring LANs with a conventional structure and a dual ring structure are compared for both uniform packet sizes and a case with mixed packet sizes. In the uniform packet size case, the dual ring delays were lower at equivalent throughput because of propagation delay effects. In the mixed packet size case the maximum delays …

Design, Fabrication, And Testing Of Temperature Compensated Magfet, Dilip K. Sampath

Theses

A simple model based on dc behaviour of MOS transistors operating in weak inversion is derived on the basis of previous publications. The bipolar like source to drain transfer characteristics of MOS transistor in weak inversion was used to implement a voltage source that is proportional to absolute temperature (PTAT); a cell that can be stacked to obtain the desired voltage. A CMOS temperature compensated current reference is implemented using PTAT cell with MOSFET as current defining element. A CMOS magnetic field sensor has been developed. The sensitivity of sensor is dependant on the bias current; incorporation of temperature independent …

An Efficient Annealing Algorithm For Global Optimization In Boltzmann Machines, Rajendra Sarasakrishna

Theses

This thesis proposes a new annealing algorithm for Boltzmann Machines. This algorithm uses an Exponential Formula for temperature scheduling that produces remarkably better solutions for global optimization. The superiority of the new algorithm is shown by computer simulations of several examples on Boltzmann Machine and its variants. This is also shown to have better properties than algorithms like Generalized Simulated Annealing which possess somewhat similar dynamics.

Fault Tolerant Clos Network, Preet Mohan S. Ahluwalia

Theses

Multistage interconnection networks, or MINs, provide paths between functional modules in multiprocessor systems. The MINs are usually segmented into several stages. Each stage connects inputs to appropriate links of the next stage so that the cumulative effect of all the stages satisfies input-output connection requirements.

This thesis deals with a fault tolerant Clos network. The fault tolerance technique involves addition of extra switches per stage to compensate for any switch failure The reliability analysis of both ordinary and fault tolerant Clos networks is presented. The optimal number of extra switches required to get the best reliability results has been analyzed.

Fast Packet Switching For B-Isdn With Neural Net Control, Ajit K. Chaudhuri

Theses

The concept of integrated network where both the voice packet and data packet are delt with, started taking shape since 1970. Later, Users? demand for communication of data, voice, file, facsimile, image, videotext, videophone, videomovie has also been felt. All these services need bit rate of transmission in the order of 10, even 100 Mbits. An economical way to transmit this higher bit rate, with flexibility of bandwidth, is to treat voice/video in the form of packets and transmit those packets in real time. Studies of this kind of communication is carried out in Broadband ISDN ( BISDN). At this …

N-Cross Coupled Phase Locked Loops, Joseph C. Aboukhalil

Theses

Interference cancelation has been mainly performed using arrays, in an important issue in communication both array and spatial cancelation and time domain approach were used. In array applications, depth of cancelation may dependent on the interferences to background power ratio and their spatial distributions. Also, array cancelers are inadequate when the signals are less than an array beamwidth apart. The latter was dealt with using supperesolution and eigenanalysis methods. Different time domain methods were used in the past. Among them what is termed Cross-Coupled PEL. It is an arrangement of two PLL each one is locked on and tracking one …

Hybrid Wafer Scale Microcircuit Integration -- [ 4989063 ], Edward S. Kolesar Jr.

AFIT Patents

A wafer scale integration arrangement wherein integrated circuit die of varying size, fabrication processes, and function are commonly mounted in the same host wafer using a filled epoxy material of special characteristics. The mounting epoxy material also serves as a substrate for the die interconnecting conductors in regions adjacent the mounted die. The described assembly also includes a newly available photosensitive polyimide material as a planarization and passivation covering for the die and host wafer and as a mounting surface for an interconnecting metal conductor array. Multiple levels of interconnection metal. Fabrication processes for the die to host wafer attachment …

Forming Fine Particles, Dennis R. Alexander

Department of Electrical and Computer Engineering: Faculty Publications

To alter feedstock material, the material is exposed to laser radiation applied at a selected angle of incidence, intensity and wavelength related to the refractive index of the feedstock material. Fine uniform particles may be formed through vapor explosion and/or plasma formation and used by this method to coat surfaces, such as with paint or adhesive or to suply uniform small particles to a heat engine. Moreover, moving materials such as a column of liquid may be subjected to high internal pressure and temperature for creating physical and chemical changes.

What To Look For And Why When Buying A Telecommunications Switch

ACUTA: Other Publications

There are three characteristics of institutions of higher education that have a strong

impact on telecommunications:

. Campus Environment

. Multiple Service Populations

. Limited Budgets

Each characteristic translates into requirements for an effective telecommunication

system.

To Serve the Campus Environment

A campus is usually the first image associated with the word "college" or "university." Typically, institutions of higher education are housed in several buildings which may be fairly close together, or spread over some distance, perhaps reaching across a city or even a state.

To serve the campus user, a system must provide feature transparency, regardless of location. Small …

Properties Of Ruthenium Oxide Coatings, Roger M. Hawk, Kamesh V. Gadepally, David N. Patangia

Journal of the Arkansas Academy of Science

Ruthenium oxide coatings have been deposited on titanium substrates using a flood coating process. These films were heat treated for varying times and temperatures. The resulting films subsequently were characterized by performing resistivity and SEM analyses. Resistivity of the ruthenium oxide coating was found to be extremely dependent upon the firing temperature. Effect of the process conditions and formulations of the coatings on the morphology with respect to their electrical characteristics is presented. Capacitors were fabricated using plates coated with ruthenium oxide coatings. Capacitance versus heat treatment temperatures are discussed and at one firing temperature (480'C), the capacitance was 50 …

An Optoelectronic Adaptive Resonance Unit, Donald C. Wunsch, T. P. Caudell, R. A. Falk, C. David Capps

Electrical and Computer Engineering Faculty Research & Creative Works

The authors demonstrate a hardware implementation of the adaptive resonance theory ART 1 neural network architecture. The optoelectronic ART1 unit, is a novel application of an old device. This device-the 4-f or Van der Lugt correlator-has historically been used as a fast pattern classifier. Usually the correlation operation is employed as a matched filter, so that a maximum correlation peak corresponds to a well-matched pattern. The device described also uses the large peaks, but takes specific advantage of the fact that a zero-shift correlation is mathematically equivalent to a two-dimensional inner product. The authors describe a promising method for emulating …

Electric Field Induced Emission As A Diagnostic Tool For Measurement Of Local Electric Field Strengths, A. N. Dharamsi, K. H. Schoenbach

Bioelectrics Publications

The phenomenon of electric field induced (EFI) emission is examined in several diatomic and polyatomic molecules. The possibility of using this phenomenon as a diagnostic tool to measure, nonintrusively, the strength and direction of local electric fields in plasmas is discussed. An estimate of the EFI signal emitted in a typical application plasma is given. This yields a lower bound on the detector sensitivity necessary to exploit EFI emission in practical applications. It is concluded that, at present, the EFI signal could be measured by some very sensitive infrared detection schemes available. Current progress in infrared detector technology, if maintained, …

Self-Pulsation Operating Regime For The Absorber Of A Twin Section Laser Diode, Gerald Farrell, P. Phelan, Joe Hegarty

Articles

The voltage-current characteristic of the absorber of a twin section laser diode is investigated as a function of the gain section current. For self pulsation to occur the absorber must be operated within a specific region of the voltage-current characteristics. This region only exists for absorber voltage current characteristics which contain an S-shaped negative resistance.

Analysis Of A Coordination Framework For Mapping Coarse-Grain Applications To Distributed Systems, Linda Ruth Schaefer

Dissertations and Theses

A paradigm is presented for the parallelization of coarse-grain engineering and scientific applications. The coordination framework provides structure and an organizational strategy for a parallel solution in a distributed environment. Three categories of primitives which define the coordination framework are presented: structural, transformational. and operational. The prototype of the paradigm presented in this thesis is the first step towards a programming development tool. This tool will allow non-specialist programmers to parallelize existing sequential solutions through the distribution, synchronization and collection of tasks. The distributed control, multidimensional pipeline characteristics of the paradigm provide advantages which include load balancing through the use …

A Practical Parallel Algorithm For The Minimization Of KröNecker Reed-Muller Expansions, Paul John Gilliam

Dissertations and Theses

A number of recent developments has increased the desirability of using exclusive OR (XOR) gates in the synthesis of switching functions. This has, in turn, led naturally to an increased interest in algorithms for the minimization of Exclusive-Or Sum of Products (ESOP) forms. Although this is an active area of research, it is not nearly as developed as the traditional Sum of Products forms. Computer programs to find minimum ESOPs are not readily available and those that do exist are impractical to use as investigative tools because they are too slow and/or require too much memory. A practical tool would …

A Commodity Trading Model Based On A Neural Network-Expert System Hybrid, K. Bergerson, Donald C. Wunsch

Electrical and Computer Engineering Faculty Research & Creative Works

Demonstrates a system that combines a neural network approach with an expert system to provide superior performance compared to either approach alone. Learning capability is provided in a software-based approach to commodity trading systems. The authors used the backpropagation network with some parameters selected experimentally. They used a human expert to implicitly define patterns, using hindsight, that an intelligent system might have been able to use for an accurate prediction. Desired outputs were found by a combination of observing the behavior of technical indices that normally precede a certain kind of market behavior, and by observing the actual market behavior …

Adaptive Gain Control And Contrast Improvement For Medical Diagnostic Ultrasound B-Mode Imaging System Using Charge-Couples Devices, Alan W. Bryant, Thomas J. Herrick

Electrical and Computer Engineering Faculty Research & Creative Works

A B-mode ultrasound imaging system operates by transmitting a high-frequency sound pulse and measuring the strength and time of flight of the reflected signal. The ultrasound signal may be attenuated as much as 60 dB in its round trip to a target and must be normalized so that contrast is not lost in the far field. Manual controls for adjusting receiver gain versus depth are common in many clinical uses but are impractical in most surgical applications. The goal of this project was to design a circuit to automatically adjust the gain. A two-dimensional low-pass filter is used to approximate …

Design Of Robust Controllers For Gas Turbine Engines, D. E. Moellenhoff, Vittal S. Rao, C. A. Skarvan

Electrical and Computer Engineering Faculty Research & Creative Works

This paper describes robust controller design methodologies for gas turbine engines. A linear state variable model for the engine is derived using partial derivatives. The Linear Quadratic Gaussian with Loop Transfer Recovery (LQG/LTR) and the Parameter Robust Linear Quadratic Gaussian (PRLQG) robust controller design methodologies have been used to design a controller for gas turbine engines. A new method is proposed by combining the features of LQG/LTR and PRLQG methods and yields good robustness properties with respect to both unstructured uncertainties in the frequency domain and structured parameter variations in the time domain. The new procedure is illustrated with the …

An Industrial Application To Neural Networks To Reusable Design, Donald C. Wunsch, R. Escobedo, T. P. Caudell, S. D. G. Smith, G. C. Johnson

Electrical and Computer Engineering Faculty Research & Creative Works

Summary form only given, as follows. The feasibility of training an adaptive resonance theory (ART-1) network to first cluster aircraft parts into families, and then to recall the most similar family when presented a new part has been demonstrated, ART-1 networks were used to adaptively group similar input vectors. The inputs to the network were generated directly from computer-aided designs of the parts and consist of binary vectors which represent bit maps of the features of the parts. This application, referred to as group technology, is of large practical value to industry, making it possible to avoid duplication of design …

Applying Artificial Intelligence To The Identification Of Variegated Coloring In Skin Tumors, Scott E. Umbaugh, Randy Hays Moss, William V. Stoecker

Electrical and Computer Engineering Faculty Research & Creative Works

The importance of color information for the automatic diagnosis of skin tumors by computer vision is demonstrated. The utility of the relative color concept is proved by the results in identifying variegated coloring. A feature file paradigm is shown to provide an effective methodology for the independent development of software modules for expert system/computer vision research. An automatic induction tool is used effectively to generate rules for identifying variegated coloring. Variegated coloring can be identified at rates as high as 92% when using the automatic induction technique in conjunction with the color segmentation method

A Hybrid Optoelectronic Art-1 Neural Processor, Donald C. Wunsch, T. P. Caudell

Electrical and Computer Engineering Faculty Research & Creative Works

Summary form only given. A new implementation of ART-1 (adaptive resonance theory) has been proposed that efficiently combines optical and electronic devices. All parallel computations are performed by the optics, while serial operations are performed in electronics.

A Laboratory Investigation Of Electro-Optic Kerr Effect For Detection Of Electric Transmission Line Faults, Badrul H. Chowdhury, E. Grigsby, T. J. Englert

Electrical and Computer Engineering Faculty Research & Creative Works

A prototype Kerr cell has been constructed and tested for detecting and identifying faults by monitoring high voltages such as are found in electric power delivery systems. Simulated faults were generated under laboratory conditions and monitored by the Kerr cell. Preliminary analysis was done using analog-to-digital conversion of the detected waveforms with a single board microprocessor serially interfaced with a personal computer. The occurrence of faults is readily observed and results indicate that identification of fault types can be accomplished within less than one cycle of a standard sixty-cycle-per-second delivery system. With a dedicated analysis system such a technique may …

A Neural Architecture For Unsupervised Learning With Shift, Scale And Rotation Invariance, Efficient Software Simulation Heuristics, And Optoelectronic Implementation, Donald C. Wunsch, D. S. Newman, T. P. Caudell, R. A. Falk, C. David Capps

Electrical and Computer Engineering Faculty Research & Creative Works

A simple modification of the adaptive resonance theory (ART) neural network allows shift, scale and rotation invariant learning. The authors point out that this can be accomplished as a neural architecture by modifying the standard ART with hardwired interconnects that perform a Fourier-Mellin transform, and show how to modify the heuristics for efficient simulation of ART architectures to accomplish the additional innovation. Finally, they discuss the implementation of this in optoelectronic hardware, using a modification of the Van der Lugt optical correlator

Security Assessment Using Neural Computing, Badrul H. Chowdhury, B. M. Wilamowski

Electrical and Computer Engineering Faculty Research & Creative Works

The advantage of fast computation capability of an artificial neural network (ANN) is used to introduce an iterative scheme for security assessment of power systems. Two related approaches are shown which demonstratedly work satisfactorily. The idea of feedback in a single-layer feedforward neural network is experimented yielding higher accuracy. The ANN is trained by using a set of data obtained from off-line analysis of the power network. After training, an approximate solution for a given condition may be found almost immediately. The approximate solution obtained is judged adequate for assessing the security of the power system. A case study is …

Error Control Coding For Meteor Burst Communications, Thomas J. Frederick

Retrospective Theses and Dissertations

The phenomenon of meteor burst communications is described and its mathematical model is presented. Protocols for meteors burst systems are outlined and examples are given of current use. Recent research in the area of coding performance in meteor burst systems is pointed out. The performance of several error control coding schemes for a meteor burst channel, including two new adaptive algorithms, are evaluated via analysis and simulation. These coding strategies are compared using the probability of successful transmission of a fixed size packet through a single burst as a performance measure. The coding methods are compared via simulation for several …

Surface Acoustic Wave Resonator Parameter Extraction And Oscillator Design, Mark Stephen Cavin

Retrospective Theses and Dissertations

Surface Acoustic Wave (SAW) resonators are frequently used in UHF oscillator designs. SAW resonators are available in the 100 MHz - 1 GHz frequency range which eliminates the need for frequency multiplication necessary with Bulk Acoustic Wave (BAW) devices. As operating frequencies increase, circuit parasitics play a larger role in device performance. One important source of parasitics is the device package and associated interconnects. To model a SAW resonator effectively, these parasitics need to be taken into account Resonator device models and their associated S-parameters can then be used in the design of oscillators. This thesis will discuss the design …