Computer Engineering Commons^™

Expanding Extracurricular Learning Opportunities Through International Engineering Student And Faculty Exchange, Mark Polczynski

Electrical and Computer Engineering Faculty Research and Publications

No abstract provided.

Multi-Model Kalman Filtering For Adaptive Nonuniformity: Correction In Infrared Sensors, Jorge E. Pezoa, Majeed M. Hayat, Sergio N. Torres, Md. Saifur Rahman

Electrical and Computer Engineering Faculty Research and Publications

This paper presents an adaptive technique for the estimation of nonuniformity parameters of infrared focal-plane arrays that is robust with respect to changes and uncertainties in scene and sensor characteristics. The proposed algorithm is based on using a bank of Kalman filters in parallel. Each filter independently estimates state variables comprising the gain and the bias matrices of the sensor, according to its own dynamical-model parameters, which underly the statistics of the scene and the nonuniformity as well as the temporal drift in the nonuniformity. The supervising component of the algorithm then generates the final estimates of the state variables …

Time-Domain Isolated Phoneme Classification Using Reconstructed Phase Spaces, Michael T. Johnson, Richard J. Povinelli, Andrew C. Lindgren, Jinjin Ye, Xiaolin Liu, Kevin M Indrebo

Electrical and Computer Engineering Faculty Research and Publications

This paper introduces a novel time-domain approach to modeling and classifying speech phoneme waveforms. The approach is based on statistical models of reconstructed phase spaces, which offer significant theoretical benefits as representations that are known to be topologically equivalent to the state dynamics of the underlying production system. The lag and dimension parameters of the reconstruction process for speech are examined in detail, comparing common estimation heuristics for these parameters with corresponding maximum likelihood recognition accuracy over the TIMIT data set. Overall accuracies are compared with a Mel-frequency cepstral baseline system across five different phonetic classes within TIMIT, and a …

Optimal Flux Weakening In Surface Pm Machines Using Fractional-Slot Concentrated Windings, Ayman M. El-Refaie, Thomas M. Jahns

Electrical and Computer Engineering Faculty Research and Publications

A design approach is presented for achieving optimal flux-weakening operation in surface permanent-magnet (SPM) synchronous machines by properly designing the machine's stator windings using concentrated, fractional-slot stator windings. This technique makes it possible to significantly increase the machine inductance in order to achieve the critical condition for providing wide speed ranges of constant-power operation. The conditions for optimal flux weakening can be achieved while simultaneously delivering sinusoidal line-to-line back-electromotive-force waveforms and low cogging torque. A closed-form analytical model is described that can be used to design SPM machines to achieve optimal flux-weakening conditions. This technique is applied to design a …

Capacity And Complexity Of Hmm Duration Modeling Techniques, Michael T. Johnson

Electrical and Computer Engineering Faculty Research and Publications

The ability of a standard hidden Markov model (HMM) or expanded state HMM (ESHMM) to accurately model duration distributions of phonemes is compared with specific duration-focused approaches such as semi-Markov models or variable transition probabilities. It is demonstrated that either a three-state ESHMM or a standard HMM with an increased number of states is capable of closely matching both Gamma distributions and duration distributions of phonemes from the TIMIT corpus, as measured by Bhattacharyya distance to the true distributions. Standard HMMs are easily implemented with off-the-shelf tools, whereas duration models require substantial algorithmic development and have higher computational costs when …

Effects Of Load Magnitude On Diagnosing Broken Bar Faults In Induction Motors Using The Pendulous Oscillation Of The Rotor Magnetic Field Orientation, Behrooz Mirafzal, Nabeel Demerdash

Electrical and Computer Engineering Faculty Research and Publications

The effects of load level on the ability to diagnose broken bar faults in squirrel-cage induction motors are studied in this paper. The pendulous oscillation of the rotor magnetic field orientation is implemented as a fault signature for rotor fault diagnostic purposes at steady-state operations. Moreover, the effects of load level on the low-side band component of the stator current spectrum, and associated diagnostic difficulties in this approach particularly in the presence of motor operation from pulsewidth-modulation drives, are reported as well. These investigations were performed through testing 2-hp and 5-hp induction motors over a wide range of load levels …

Nanoindentation Technique For Characterizing Cantilever Beam Style Rf Microelectromechanical Systems (Mems) Switches, Hyukjae Lee, Ronald A. Coutu Jr., Shankar Mall, Paul E. Kladitis

Electrical and Computer Engineering Faculty Research and Publications

A nanoindentation technique was used to mechanically actuate a radio frequency micro-switch along with the measurement of contact resistance to investigate its applicability to characterize deflection and contact resistance behaviors of micro-sized cantilever beam switches. The resulting load–displacement relationship showed a discontinuity in slope when the micro-switch closed. The measured spring constants reasonably agreed with theoretical values obtained from the simple beam models. The change in contact resistance during test clearly indicated micro-switch closure but it did not coincide exactly with the physical contact between two electric contacts due to a resistive contaminated film.

Generalized Phase Space Projection For Nonlinear Noise Reduction, Michael T. Johnson, Richard J. Povinelli

Electrical and Computer Engineering Faculty Research and Publications

Improved phase space projection methods, adapted from related work in the linear signal processing field based on subspace decomposition, are presented for application to the problem of additive noise reduction in the context of phase space analysis. These methods improve upon existing methods such as Broomhead–King singular spectrum analysis projection by minimizing overall signal distortion subject to constraints on the residual error, rather than using a direct least-squares fit. This results in a range of weighted projections which estimate and compensate for the portion of the principal component's singular values corresponding to noise rather than signal energy, and which include …

Automatic Classification And Speaker Identification Of African Elephant (Loxodonta Africana) Vocalizations, Patrick J. Clemins, Michael T. Johnson, Kirsten Leong, Anne Savage

Electrical and Computer Engineering Faculty Research and Publications

A hidden Markov model (HMM) system is presented for automatically classifying African elephant vocalizations. The development of the system is motivated by successful models from human speech analysis and recognition. Classification features include frequency-shifted Mel-frequency cepstral coefficients (MFCCs) and log energy, spectrally motivated features which are commonly used in human speech processing. Experiments, including vocalization type classification and speaker identification, are performed on vocalizations collected from captive elephants in a naturalistic environment. The system classified vocalizations with accuracies of 94.3% and 82.5% for type classification and speaker identification classification experiments, respectively. Classification accuracy, statistical significance tests on the model parameters, …

Gain-Bandwidth Product Optimization Of Heterostructure Avalanche Photodiodes, Oh-Hyun Kwon, Majeed M. Hayat, Joe C. Campbell, Bahaa E.A. Saleh, Malvin Carl Teich

Electrical and Computer Engineering Faculty Research and Publications

A generalized history-dependent recurrence theory for the time-response analysis is derived for avalanche photodiodes with multilayer, heterojunction multiplication regions. The heterojunction multiplication region considered consists of two layers: a high-bandgap Al/sub 0.6/Ga/sub 0.4/As energy-buildup layer, which serves to heat up the primary electrons, and a GaAs layer, which serves as the primary avalanching layer. The model is used to optimize the gain-bandwidth product (GBP) by appropriate selection of the width of the energy-buildup layer for a given width of the avalanching layer. The enhanced GBP is a direct consequence of the heating of primary electrons in the energy-buildup layer, which …

The Effect Of Time Delays On The Stability Of Load Balancing Algorithms For Parallel Computations, John Chiasson, Zhong Tang, Jean Ghanem, Chaouki T. Abdallah, J. Douglas Birdwell, Majeed M. Hayat, Henry Jérez

Electrical and Computer Engineering Faculty Research and Publications

A deterministic dynamic nonlinear time-delay system is developed to model load balancing in a cluster of computer nodes used for parallel computations. The model is shown to be self consistent in that the queue lengths cannot go negative and the total number of tasks in all the queues and the network are conserved (i.e., load balancing can neither create nor lose tasks). Further, it is shown that using the proposed load balancing algorithms, the system is stable in the sense of Lyapunov. Experimental results are presented and compared with the predicted results from the analytical model. In particular, simulations of …

Exponential Time Response In Analogue And Geiger Mode Avalanche Photodiodes, C. Groves, C. H. Tan, J. P.R. David, G. J. Rees, Majeed M. Hayat

Electrical and Computer Engineering Faculty Research and Publications

The mean avalanche current impulse response in an avalanche photodiode exhibits an initial transient and then grows or decays, above or below breakdown, at exponential rates which depend only on the probability distributions of the electron and hole ionization events. The process continues while the electric field profile remains unchanged by the applied bias or the evolving space charge. Below breakdown the distribution in the avalanche duration also exhibits an initial transient and then decays exponentially at the same rate as the mean current. Below breakdown the standard deviation in current decays exponentially at one half of the rate of …

Information-Theoretic Criterion For The Performance Of Single-Photon Avalanche Photodiodes, David A. Ramirez, Majeed M. Hayat, Sergio N. Torres, Bahaa E.A. Saleh, Malvin Carl Teich

Electrical and Computer Engineering Faculty Research and Publications

A channel-capacity metric is introduced for assessing the performance of single-photon avalanche photodiodes (SPADs) when used as detectors in laser communication systems. This metric is employed to theoretically optimize, with respect to the device structure and operating voltage, the performance of SPADs with simple InP or In/sub 0.52/Al/sub 0.48/As-InP heterojunction multiplication regions. As the multiplication-region width increases, an increase is predicted in both the peak and the full-width at half-maximum of the channel capacity curve versus the normalized excess voltage. Calculations also show the existence of an optimal In/sub 0.52/Al/sub 0.48/As-InP heterojunction multiplication region that maximizes the peak channel capacity …

Theoretical Investigation Of Quantum-Dot Avalanche Photodiodes For Mid-Infrared Applications, Sanjay Krishna, Oh-Hyun Kwon, Majeed M. Hayat

Electrical and Computer Engineering Faculty Research and Publications

A novel midinfrared sensor, called the quantum-dot avalanche photodiode (QDAP), is proposed which is expected to have improved signal-to-noise ratio (SNR) in the presence of Johnson noise over its quantum-dot (QD) counterpart. In the QDAP, an intersubband QD detector is coupled with a thin, low-noise GaAs avalanche layer through a tunnel barrier. The avalanche layer provides the necessary photocurrent gain required to overcome Johnson noise and nearly achieve the dark-current-limited SNR of the QD detector. In the proposed three-terminal device, the applied biases of the QD-detector and the avalanche-photodiode sections of the QDAP are controlled separately. This feature permits the …

Condition Monitoring Of Squirrel-Cage Induction Motors Fed By Pwm-Based Drives Using A Parameter Estimation Approach, Behrooz Mirafzal, F. Fateh, Chia-Chou Yeh, Richard J. Povinelli, Nabeel Demerdash

Electrical and Computer Engineering Faculty Research and Publications

Abstract:

A rotor condition monitoring technique is presented in this paper based on a parameter estimation approach. In this technique, the stator currents, voltages and motor speed are used as the input signals, where the outputs will be the rotor's inductance, resistance and consequently rotor time constant. This approach is verified by simulation of two different induction motor cases. These simulations are buttressed by experimental data obtained for a 2-hp induction motor in the case of healthy as well as one, three and five rotor bar breakages. In these tests, the induction motor was energized from a PWM-based drive, in …

Micro-Switches With Sputtered Au, Aupd, Au-On-Aupt, And Auptcu Alloy Electric Contacts, Ronald A. Coutu Jr., P. E. Kladitis, R. Cortez, R. E. Strawser, Robert L. Crane

Electrical and Computer Engineering Faculty Research and Publications

This work is the first to report on a new analytic model for predicting micro-contact resistance and the design, fabrication, and testing of microelectromechanical systems (MEMS) metal contact switches with sputtered bi-metallic (i.e. gold (Au)-on-Au-platinum (Pt), (Au-on-Au-(6%)Pt)), binary alloy (i.e. Au-palladium (Pd), (Au-(2%)Pd)), and tertiary alloy (i.e. Au-Pt-copper (Cu), (Au-(5%)Pt-(0.5%)Cu)) electric contacts. The micro-switches with bi-metallic and binary alloy contacts resulted in contact resistance between 1-2 /spl Omega/ and, when compared to micro-switches with sputtered Au electric contacts, exhibited a 3.3 and 2.6 times increase in switching lifetime, respectively. The tertiary alloy exhibited a 6.5 times increase in switch lifetime …

A Comparison Of Micro-Switch Analytic, Finite Element, And Experimental Results, Ronald A. Coutu Jr., P. E. Kladitis, Lavern A. Starman, J. R. Reid

Electrical and Computer Engineering Faculty Research and Publications

Electrostatically actuated, metal contact, micro-switches depend on having adequate contact force to achieve desired, low contact resistance. In this study, higher contact forces resulted from overdriving cantilever beam style switches, after pull-in or initial contact, until the beam collapsed onto the drive or actuation electrode. The difference between initial contact and beam collapse was defined as the useful contact force range. Micro-switch pull-in voltage, collapse voltage, and contact force predictions, modeled analytically and with the CoventorWare finite element software package, were compared to experimental results. Contact resistance was modeled analytically using Maxwellian spreading resistance theory. Contact resistance and contact …

Selecting Metal Alloy Electric Contact Materials For Mems Switches, Ronald A. Coutu Jr., Paul E. Kladitis, Kevin D. Leedy, Robert L. Crane

Electrical and Computer Engineering Faculty Research and Publications

This paper presents a method for selecting metal alloys as the electric contact materials for microelectromechanical systems (MEMS) metal contact switches. This procedure consists of reviewing macro-switch lessons learned, utilizing equilibrium binary alloy phase diagrams, obtaining thin film material properties and, based on a suitable model, predicting contact resistance performance. After determining a candidate alloy material, MEMS switches were designed, fabricated and tested to validate the alloy selection methodology. Minimum average contact resistance values of 1.17 and 1.87 Ω were measured for micro-switches with gold (Au) and gold–platinum (Au–(6.3%)Pt) alloy electric contacts, respectively. In addition, 'hot-switched' life cycle test results …

Time Series Classification Using Gaussian Mixture Models Of Reconstructed Phase Spaces, Richard J. Povinelli, Michael T. Johnson, Andrew C. Lindgren, Jinjin Ye

Electrical and Computer Engineering Faculty Research and Publications

A new signal classification approach is presented that is based upon modeling the dynamics of a system as they are captured in a reconstructed phase space. The modeling is done using full covariance Gaussian mixture models of time domain signatures, in contrast with current and previous work in signal classification that is typically focused on either linear systems analysis using frequency content or simple nonlinear machine learning models such as artificial neural networks. The proposed approach has strong theoretical foundations based on dynamical systems and topological theorems, resulting in a signal reconstruction, which is asymptotically guaranteed to be a complete …

Application Of Bi-State Magnetic Material To Automotive Offset-Coupled Ipm Starter/Alternator Machine, Ayman M. El-Refaie, Russell Manzke, Thomas M. Jahns

Electrical and Computer Engineering Faculty Research and Publications

This paper investigates a new approach to designing high-speed interior permanent-magnet (IPM) synchronous machines using a bi-state soft magnetic material. The bi-state material can have its normally high magnetic permeability permanently reduced in localized regions to that of air by means of heat treatment. This new work significantly expands a previous investigation by considering offset-coupled IPM machines that make it possible to significantly increase the rotor speed while retaining all of the other specifications of the 6-kW starter/alternator application. Lumped-parameter models, Monte Carlo optimization, and both electromagnetic and structural finite-element analysis are used to develop new offset-coupled IPM machine designs …

Induction Machine Broken-Bar Fault Diagnosis Using The Rotor Magnetic Field Space-Vector Orientation, Behrooz Mirafzal, Nabeel Demerdash

Electrical and Computer Engineering Faculty Research and Publications

A new technique based on rotor magnetic field space vector orientation is presented to diagnose broken-bar faults in induction machines operating at steady state. In this technique, stator currents and voltages are used as inputs to compute and subsequently observe the rotor magnetic field orientation, which has a more significant "swing-like" pendulous oscillation in case of broken-bar faults than in healthy operation. It will be shown here that the range of this "pendulous oscillation" is a function of the number of broken bars. Also in this technique, it was found that an inter-turn shorted stator-winding fault, which exhibits similar pendulous …

Optimal Calibration Of Pet Crystal Position Maps Using Gaussian Mixture Models, Kelly A. Stronger, Michael T. Johnson

Electrical and Computer Engineering Faculty Research and Publications

A method is developed for estimating optimal PET gamma-ray detector crystal position maps, for arbitrary crystal configurations, based on a binomial distribution model for scintillation photon arrival. The approach is based on maximum likelihood estimation of Gaussian mixture model parameters using crystal position histogram data, with determination of the position map taken from the posterior probability boundaries between mixtures. This leads to minimum probability of error crystal identification under the assumed model.

Effect Of Stochastic Dead Space On Noise In Avalanche Photodiodes, Oh-Hyun Kwon, Majeed M. Hayat, Joe C. Campbell, Bahaa E.A. Saleh, Malvin Carl Teich

Electrical and Computer Engineering Faculty Research and Publications

A stochastic dead-space model for impact ionization is developed and used to study the effect of the soft nature of the ionization capability of carriers on the excess noise factor of avalanche photodiodes. The proposed model is based on the rationale that the gradual, or soft, transition in the probability density function (PDF) for the distance from birth to impact ionization can be viewed as that resulting from uncertainty in the dead space itself. The resulting soft PDF, which is parameterized by a tunable softness parameter, is used to establish the limitations of the existing hard-threshold ionization models in ultrathin …

Optimized Breakdown Probabilities In Al/Sub 0.6/Ga/Sub 0.4/As-Gaas Heterojunction Avalanche Photodiodes, Oh-Hyun Kwon, Majeed M. Hayat, Joe C. Campbell, Bahaa E.A. Saleh, Malvin Carl Teich

Electrical and Computer Engineering Faculty Research and Publications

Recently, it has been shown that the noise characteristics of heterojunction Al/sub 0.6/Ga/sub 0.4/As-GaAs avalanche photodiodes (APDs) can be optimized by proper selection of the width of the Al/sub 0.6/Ga/sub 0.4/As layer. Similar trends have also been shown theoretically for the bandwidth characteristics. The resulting noise reduction and potential bandwidth enhancement have been attributed to the fact that the high bandgap Al/sub 0.6/Ga/sub 0.4/As layer serves to energize the injected electrons, thereby minimizing their first dead space in the GaAs layer. We show theoretically that the same optimized structures yield optimal breakdown-probability characteristics when the APD is operated in Geiger …

The Effect Of Pruning And Compression On Graphical Representations Of The Output Of A Speech Recognizer, Yang Liu, Mary P. Harper, Michael T. Johnson, Leah H. Jamieson

Electrical and Computer Engineering Faculty Research and Publications

Large vocabulary continuous speech recognition can benefit from an efficient data structure for representing a large number of acoustic hypotheses compactly. Word graphs or lattices have been chosen as such an efficient interface between acoustic recognition engines and subsequent language processing modules. This paper first investigates the effect of pruning during acoustic decoding on the quality of word lattices and shows that by combining different pruning options (at the model level and word level), we can obtain word lattices with comparable accuracy to the original lattices and a manageable size. In order to use the word lattices as the input …

Diagnostics Of Eccentricities And Bar/End-Ring Connector Breakages In Polyphase Induction Motors Through A Combination Of Time-Series Data Mining And Time-Stepping Coupled Fe-State Space Techniques, John F. Bangura, Richard J. Povinelli, Nabeel Demerdash, Ronald H. Brown

Electrical and Computer Engineering Faculty Research and Publications

This paper develops the foundations of a technique for detection and categorization of dynamic/static eccentricities and bar/end-ring connector breakages in squirrel-cage induction motors that is not based on the traditional Fourier transform frequency-domain spectral analysis concepts. Hence, this approach can distinguish between the "fault signatures" of each of the following faults: eccentricities, broken bars, and broken end-ring connectors in such induction motors. Furthermore, the techniques presented here can extensively and economically predict and characterize faults from the induction machine adjustable-speed drive design data without the need to have had actual fault data from field experience. This is done through the …

A New Temporal Pattern Identification Method For Characterization And Prediction Of Complex Time Series Events, Richard J. Povinelli, Xin Feng

Electrical and Computer Engineering Faculty Research and Publications

A new method for analyzing time series data is introduced in this paper. Inspired by data mining, the new method employs time-delayed embedding and identifies temporal patterns in the resulting phase spaces. An optimization method is applied to search the phase spaces for optimal heterogeneous temporal pattern clusters that reveal hidden temporal patterns, which are characteristic and predictive of time series events. The fundamental concepts and framework of the method are explained in detail. The method is then applied to the characterization and prediction, with a high degree of accuracy, of the release of metal droplets from a welder. The …

Breakdown Probabilities For Thin Heterostructure Avalanche Photodiodes, Majeed M. Hayat, Ünal Sako ̆Glu, Oh-Hyun Kwon, Shuling Wang, Joe C. Campbell, Bahaa E.A. Saleh, Malvin Carl Teich

Electrical and Computer Engineering Faculty Research and Publications

The recurrence theory for the breakdown probability in avalanche photodiodes (APDs) is generalized to heterostructure APDs that may have multiple multiplication layers. The generalization addresses layer-boundary effects such as the initial energy of injected carriers as well as the layer-dependent profile of the dead space in the multiplication region. Reducing the width of the multiplication layer serves to both downshift and sharpen the breakdown probability curve as a function of the applied reverse-bias voltage. In structures where the injected carriers have an initial energy that is comparable to the ionization threshold energy, the transition from linear mode to Geiger-mode is …

Scene-Based Nonuniformity Correction For Focal Plane Arrays By The Method Of The Inverse Covariance Form, Sergio N. Torres, Jorge E. Pezoa, Majeed M. Hayat

Electrical and Computer Engineering Faculty Research and Publications

What is to our knowledge a new scene-based algorithm for nonuniformity correction in infrared focal-plane array sensors has been developed. The technique is based on the inverse covariance form of the Kalman filter (KF), which has been reported previously and used in estimating the gain and bias of each detector in the array from scene data. The gain and the bias of each detector in the focal-plane array are assumed constant within a given sequence of frames, corresponding to a certain time and operational conditions, but they are allowed to randomly drift from one sequence to another following a discrete-time …

Optimal Excess Noise Reduction In Thin Heterojunction Al0.6Ga0.4As-Gaas Avalanche Photodiodes, Oh-Hyun Kwon, Majeed M. Hayat, Shuling Wang, Joe C. Campbell, Archie L. Holmes, Yi Pan, Bahaa E.A. Saleh, Malvin Carl Teich

Electrical and Computer Engineering Faculty Research and Publications

It has been recently found that the initial-energy effect, which is associated with the finite initial energy of carriers entering the multiplication region of an avalanche photodiode (APD), can be tailored to reduce the excess noise well beyond the previously known limits for thin APDs. However, the control of the initial energy of injected carriers can be difficult in practice for an APD with a single multiplication layer. In this paper, the dead-space multiplication recurrence theory is used to show that the low noise characteristics associated with the initial-energy effect can be achieved by utilizing a two-layer multiplication region. As …