Investigations of a fundamental and/or applied nature, intended to develop design techniques, research techniques, initiative, and self-reliance. For the project option, after three credits, a written project report has to be completed and approved by the students graduate committee. For the thesis option, after six credits, a written thesis must be completed in accordance with the rules of the Graduate School and the College of Engineering. Admission to the course is based on a formal project/thesis proposal endorsed by the students graduate committee and submitted to the ECE Graduate Program Director.

Investigations of a fundamental and/or applied nature, intended to develop design techniques, research techniques, initiative, and self-reliance. A written thesis must be completed in accordance with the rules of the Graduate School and the College of Engineering. Admission to the course is based on a formal proposal endorsed by the students graduate committee and submitted to the ECE Graduate Program Director.

Research for and preparation of doctoral dissertation proposal. The dissertation proposal must provide a thorough survey of the research activities in the research topic area and it must present original and innovative research ideas and preliminary results as well as a defined research scope and directions. Ph.D. students must have passed this course before registering for doctoral dissertation research credits. Graded P/F

Ph.D. students who have completed course credit requirement but not yet passed qualifying exam may take the course with approval of faculty advisor.

An in depth exploration of the architecture and systems of state-of-the-art distributed computers. Students will develop an understanding of the requirements and design issues associated with high performance computing using networks of commodity computers, including the underlying networking technologies and issues and techniques associated with process scheduling and load balancing. Representative systems will be examined.

An in depth exploration of the issues and methodology in programming distributed computers. Students will develop an understanding of the programming languages and supporting programming environments associated with high performance computing on networks of commodity computers. Representative algorithms and applications will be examined.

A survey of issues and methodology in programming distributed computers. Students will develop an understanding of the hardware and software used in high performance computing based upon networks of commodity computers. Representative systems, algorithms and applications will be examined.

An in depth view of database management systems architecture and operations. The focus is on architectural and operational aspects of transactions and transaction processing. Topics include properties of data in a database, database management systems architecture, transaction properties, transaction processing, transaction and database recovery, concurrency control, locking protocols, storage management and the application of concepts within various database systems. The course includes a design project derived from topics covered.

An in depth exploration of the theory, architecture, implementation and design of state-of-the-art specialized data base systems. Students will develop an understanding of the requirements and design issues associated with emerging technologies applied to specialized database systems. Database systems to be studied will be selected based on present research interest of course faculty and students.

Basic theory and applications of wavelets and filter banks. Wavelet theory provides very general techniques that can be applied to many tasks in signal processing, e.g., multi-resolution analysis in computer vision, sub-band coding in speech and image compression, and wavelet series expansions in applied mathematics. The course is designed to enable participants to understand wavelet theory and to acquire a working knowledge of the techniques available in this signal processing area. In particular, a paramount goal is to enable each participant to develop a critical understanding of the advantages and limitations of wavelet analysis.

Advanced signal processing topics. Content may vary according to instructors preferences but typically includes selections from: two-dimensional signal processing, higher-order spectral analysis, chaotic signal processing, array signal processing, multirate signal processing, optimal filtering and linear prediction, time-frequency and time-scale signal analysis, smart antennas, and inverse problems (signal reconstruction). Applications are discussed in radar, sonar, acoustics, speech, communications, and image processing.

Basic theory of adaptive filter design and implementation including applications. Topics include optimal filters, adaptive linear combiners, performance measures, adaptive FIR filters, adaptive IIR filters, and nonlinear adaptive filters. Applications in adaptive signal processing include adaptive modeling and system identification, adaptive deconvolution and equalization, and adaptive interference canceling.

Signal processing and statistical techniques used in processing speech signals providing an understanding of how these techniques are used in the coding, synthesis and recognition of speech. Topics typically include the human vocal and auditory systems, characteristics of speech signals, lossless tube model of speech production, time and frequency domain representations of speech, time-frequency speech analysis methods, homomorphic speech processing, speech coding, speech synthesis, speech recognition, pitch detection and processing, and acoustic preprocessing for speech recognition.

Advanced topics in computer networks. Topics include: network management systems and architectures; network management protocols and standards; management of information bases. Examples are drawn primarily from the Internet (e.g., SNMP).

Fundamental aspects of information theory. Topics covered include discrete and differential entropy, discrete source and channel model, information rate, mutual information and channel capacity, coding theorems for sources and channels, the data processing theorem, encoding and decoding of data for transmission over noisy channels, rate distortion theory, maximum entropy distributions and entropy estimation techniques for unknown sources. Several applications of information theory are included.

Fundamentals of detection theory. Topics include Bayes and Neyman-Pearson tests, composite hypothesis testing, nonparametric test, detection of known signals in Gaussian noise, detection of signals with random parameters in noise, multiple pulse detection of signals, generalized likelihood ratio test, Bayes and maximum likelihood estimators, space-time processing, application to radar and sonar.

An introduction to the theory and applications of pattern recognition. Topics include descriptions of patterns, problem formulation, linear and nonlinear classification theories, representation of patterns, feature selection, supervised and unsupervised training, nonparametric methods in pattern recognition, cluster and mode-seeking techniques, recursive algorithms using stochastic approximation, sequential pattern recognition, design of computer recognition experiments, linguistic approach to pattern recognition.

Time-varying signal processing methods. The course covers many of the prevalent techniques that have been developed over the years for time-frequency signal analysis and addresses the characteristics and properties of time-frequency representations in Cohens fixed kernel class, e.g., the spectrogram and the Wigner distribution. The course covers many time-frequency representations and addresses their performance tradeoffs in applications. Gradually, the student learns about the terms that are pertinent to the field and develops an understanding for the state-of-the-art of this area of signal processing.

Principles and design of sonar systems. Topics include: complex array and element apertures (weighting) functions, and beam shaping; linear, planar, and volumetric arrays; directivity and beam-forming; operating and installation of sonar systems; improving signal-to-noise ratios; wave vector spectrum filtering.

Theory of neural networks. Topics include learning models, single and multilayer perceptrons, LMS algorithm, back propagation algorithms, radial basis function networks, Hopfield networks and Boltzman machine, self-organizing systems including Hebbian learning, Kohonen feature map algorithm, temporal processing neural networks, biological neural networks, and VLSI implementation.

Fundamentals of digital image processing. Topics include human vision models, 2-D sampling and quantization, image transforms, image enhancements, color image processing, image restoration, image and video compression, image segmentation by thresholding and region analysis, texture analysis, boundary descriptions, morphological methods, image processing system architecture.

Common mathematical frameworks in the processing of geophysical, radar, and speech signals are introduced, followed by a study of individual source mechanisms and transmission media. Specific digital filtering, deconvolution, spectral analysis and interference or clutter rejection techniques are discussed. Case studies for effective processing of seismic, radar and speech signals are also included.

Conditioning and labeling, the facet model, texture models, image segmentation and arc extraction, 3-D shape representation and shape recovery, surface reflection mechanism, shape from shading, range image analysis, stereo vision, 2-D and 3-D motion analysis, non-rigid body motion analysis, relational matching, 3-D object recognition, fundamentals of robot vision, architecture of computer vision systems.

Nonlinear acoustic fields and parametric sources. Topics include nonlinear acoustics of fluids, turbulence, underwater explosions as sources of sound, parametric acoustic arrays, finite-amplitude effects, acoustic cavitation and streaming.

An advanced course covering fundamental principles, design, and operation of transducers for the reception and generation of underwater sound using energy analysis methods. Topics include: theory of simple radiators and receivers, electromechanical circuit analogies, impedance functions and equivalent circuits; piezoelectricity; reciprocity; acoustic properties of transducer materials; acoustic motion sensors; pressure gradient sensor designs, and diffractions constant.