Wright State University CalendarsSearchDirectories
Info For:
Future Students Current Students Parents Alumni and Friends Faculty and Staff Government and Industry
Info About:
About the College Departments Pre-College Programs Academic Programs Research News and Events Contact Us! Visit Us!
CECS Web Index:
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
College of Engineering and Computer Science Home Dayton Area Graduate Studies Institute WSU Home

Research

Featured Research Topics

Advanced Nano Composites Dr. Sharmila Mukhopadhyay
Sponsors: AFRL (ML & PR), P&G, OBR, NSF

The X-Ray spectrometer is used to accomplish surface chemistry. X-Ray generated photo electrons allow exact chemical compositions of the surface of a material to be determined.
This research effort deals with the structure chemistry and properties of material bonds at the atomic level. Dr. Sharmila Mukhopadhyay is looking at the interface (or seam) of solids to better understand the forces that hold them together. One focus of her research involves developing multifunctional aerospace coatings that can be used across a broad spectrum of applications–from providing wearable fabrics with an effective shield against dangerous chemicals, to protecting components in microelectronic circuits, and developing corrosion resistant, lightweight airframes. Dr. Mukhopadhyay is also involved with high temperature superconductors. One goal is to develop superconducting wires that can make space-based electrical generators lighter and more efficient. Another goal is to investigate the use of superconductor-based antigravity and levitation in space systems.

Artificial Intelligence and Soft Computing

Artificial intelligence is an area of computer science focused on the development of computer systems that exhibit intelligent behavior in decision making, object recognition, planning, learning, and other applications that require intelligent assessment of complex information. Many AI systems incorporate learning systems such as simulated neural networks or evolutionary algorithms to improve performance with experience. One of the difficulties encountered in many of the domains considered in artificial intelligence is the need for decision making in the presence of uncertain and/or incomplete information. Soft computing refers to a number of formal systems that have been developed to represent and reason with uncertain information. These systems include fuzzy logic, rough set theory, Bayesian networks, and hidden Markov models. Current areas of research include:

  • Intelligent agents
  • Natural language understanding
  • Evolutionary algorithms and evolvable hardware
  • Autonomous robotic systems
  • Machine learning
  • Fuzzy and neural systems
  • Intelligent control systems

Faculty

Relevant Courses

CS: 609, 714, 767, 771, 772, 865; CEG: 619, 756, 757, 759


Assistive Technologies

"Disabilities can be very traumatic, leading to frustration and depression," according to the American Foundation for the Blind. The rate of unemployment among legally blind individuals of working age residing in the United States greatly exceeds the unemployment rate for individuals with no functional limitations. Clever devices and information technology engineering strategies can be developed to help people overcome barriers to pursue educational and professional opportunities that will allow them to become productive members of the society. Some of the ongoing research projects are:

  • Reading devices for the blind and visually impaired
  • Navigation devices for the blind
  • Multimodal forms of representation for virtual learning environments

Faculty

Relevant Courses

CS: 609, 670, 712, 714; CEG: 656, 756, 757


Automatic Target Recognition and Tracking Dr. Fred Garber
Sponsors: AFRL, DARPA

Pattern recognition is used to identify a military target vehicle (lower right).
In the realm of high technology, rapidly evolving military operations, one of the greatest challenges is achieving visibility over the battlefield and the location of enemy resources. This research effort deals with the military requirement to identify, evaluate, and track enemy targets on the battlefield and provide computer assessments of their disposition during and after attack. The focus is on achieving rapid and accurate pattern recognition of targets using machine processed tactical sensor data, then using the recognized pattern, identifying and tracking targets and their location.

Bioinformatics

Bioinformatics advances fundamental concepts in molecular biology, biochemistry, and computer science to help further understanding of basic DNA, genes, and protein structures it relates to mechanisms for drug development and treatment of diseases. Current research projects are:

  • Metabolomics and toxicology
  • Trends in molecular evolution
  • Automation of forensic DNA analysis
  • Indexing genomic databases
  • Stochastic reaction modeling
  • Search optimization
  • National model for bioinformatics education
  • Disease analysis

Faculty

Relevant Courses

CS: 671; BMS: 925, 765


Biomedical Imaging and Visualization

Biomedical imaging and visualization research has become a very active research field during the last two decades, offering unique solutions for a great variety of biological and biomedical problems. Analysis and visualization of medical images facilitates diagnosis and treatment planning. Visualization systems used as surgical navigation systems enable precise and minimally invasive surgery. Some of the research projects are:

  • Image registration in surgical navigation
  • Segmentation of MR and CT images for spinal surgery
  • Design of a surgical robot assistance for biopsy
  • Detection and visualization of brain shift during brain surgery
  • Automated endoscopic imaging
  • Bio-signatures of in-vivo cells

Faculty

Relevant Courses

CEG: 616, 658, 676, 677, 679, 724, 725, 726, 739, 777


Biometrics
  • 2D and 3D biometrics
    • Facial imaging
    • Finger prints
    • Iris and Retina scans
    • Gait and motion data
  • 2D and 3D feature extraction and analysis
  • Matching algorithms for comparison of features
  • Multimodal information fusion
  • Faculty: Goshtasby, Bourbakis
  • Funding sources: Pending

Cloud Computing

Cloud computing is a major step toward organizing all aspects of computation as a public utility service. It embraces concepts such as software as a service and platform as a service, including services for workflow facilities, application design and development, deployment and hosting services, data integration, and management of software. The cloud platform increases in importance as our industry makes the phase change from in-house data management to cloud-hosted data management to improve efficiency and focus on core businesses. However, like any new technology, there are formidable problems, from performance issues to security and privacy, from metadata management to massively parallel execution. Some of the active research projects are:

  • Cloud infrastructure for data management
  • Privacy and security in cloud data management
  • Cloud-based mining and learning algorithms
  • Cloud support for text mining and web search
  • Large-scale natural language modeling and translation
  • Parallel and distributed algorithms for bioinformatics
  • Performance evaluation and benchmarking

Faculty

Relevant Courses

CS: 619, 771, 772, 774; CEG: 505, 629, 759

For more information, visit the Cloud Collaboratory.


Command and Control of Remotely Piloted Vehicles Dr. S. Narayanan and Dr. Jennie Gallimore
Sponsors: AFRL, AFIT, DAGSI, ITRI, Industry

The multimodal control system provides an effective control interface for a remotely piloted vehicle.
The use of remotely piloted vehicles is increasing dramatically for applications involving reconnaissance, supression of enemy air defenses, and hazardous search and rescue missions. As the mission profiles of these vehicles become more sophisticated and demanding, the need for supervisory interfaces between human and machine increases. Drs. Narayanan and Gallimore are currently researching more effective ways to enhance human-system interaction involving remotely piloted vehicles. Their approach is to develop multimodal interfaces that can exploit graphical, auditory, and visual channels.

Communication Theory

Communication theory encompasses the processes of modifying, transmitting and receiving signals over distance. Research in the communication area includes the development of an all digital radar receiver, investigations in theoretical work on estimating the angle of arrival of single and multiple radar signals, the development of algorithms and theoretical bases for automatic target recognition and real-time spectral analysis. (Dr. F. Garber, Dr. A. Shaw)

(image info)

Computational Nonlinear Aeroelasticity for Multidisciplinary Analysis and Design Dr. Ramana Grandhi
Sponsors: AFRL, DAGSI, ITRI

Ensuring airframe structural integrity requires sophisticated analysis and design for flutter and limit cycle oscillation.

One of the key elements in building airframes is developing materials that provide the needed structural integrity. This research focuses on dealing with the problem of airframe integrity by optimizing structural design. The goal is to maximize performance while minimizing weight. This research specifically looks at hypersonic vehicle design using multi-physics modeling of flow-behavior with high fidelity simulations. It integrates aerodynamics, structures, and controls with reduced-order systems for computationally efficient preliminary designs.

Control Theory

Control theory involves the modeling of electrical circuits that use the difference between nominal and sensed data to control a mechanical device or process. Current research includes the study of more efficient algorithms for designing robust multivariable control systems. Research also is being done in the areas of fuzzy logic with biomedical applications. (Dr. L. Hong, Dr. P. Misra, Dr. K. Rattan)


Databases and Data Mining

Data mining is the process of extracting useful knowledge from a database. Data mining facilitates the characterization, classification, clustering, and searching of different databases, including text data, image and video data, and bioinformatics data for various applications. Text, multimedia, and bioinformatics databases are very large and so parallel/distributed data mining is essential for scalable performance. Active projects in this area are:

  • Parallel/distributed data mining
  • Text/image clustering and categorization
  • Metadata for timelining events
  • XML database
  • Data warehousing
  • Biological/medical data mining

Faculty

Relevant Courses

CS: 605, 701, 705, 707, 711, 801, 805


One of several of WSU's parallel data mining systems

DSP

  (image info)

Digital signal processing involves the application of digital technology to the processing of analog signals. This process enables programming flexibility and the use of sophisticated programming algorithms, improves accuracy, and stores and transports data. Investigations include developing algorithms to acquire 3-D representations of objects through optical scanning and processing images for designing and fitting prosthetic devices. Other research topics include the conversion of low resolution imaging sources such as video, graphical data, and still images into high resolution data for use in color and graphics printers, and large screen displays; the application of color image and video compression techniques to efficiently code color images and video data; and new approaches to real-time multidimensional/multiresolutional estimation. (Dr. K. Xue)


Electromagnetics

The study of electromagnetic fields is concerned with the effects of forces created by an electric charge moving from one position to another. Research in the electromagnetics area includes the development of computer models for the analysis of electromagnetic fields created in microwave and millimeter wave integrated circuits, the analysis of conformal antennas on curved surfaces, and computational methods for the prediction and control of electromagnetic interference in electronic products. (Dr. P. Misra)


Enhanced Use of Medical Images Dr. Ardeshir Goshtasby
Sponsors: ITRI, Kettering Medical Center

Sophisticated pattern recognition techniques are being developed to help in the diagnosis of tomographic images, and even assist in treatment.
The use and scope of medical diagnostic imaging has progressed to the point where more effective ways are needed to fully exploit the value of advanced imaging products. Dr. Goshtasby is developing a method for registering multimodal volume images. The method is currently optimized to register brain images, but it can be adjusted to register whole-body images. In fact, it has been tested with computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomographic (PET) images with very encouraging results, and is being used now in the operating room. The ultimate goal is to provide more effective machine assisted diagnostics and, in some cases, treatment.

High Cycle Fatigue for Turbine Engine Blades Dr. Mitch Wolff and Dr. Joseph Slater
Sponsors: AFRL, DAGSI

A Shock Tube generates a supersonic shock waves to calibrate the pressure sensors required for advanced jet engine research.
This research focuses on one of the leading causes of engine failure in USAF aircraft. Given the cost and time of development of new weapon systems, the Air Force has long been a leader in extending the service life of existing airframes through various programs that deal with structural integrity, propulsion, and avionics. This particular experimental and computational research looks at engine failure in high performance aircraft with the goal of understanding specific failure modes.

Human Performance in Aviation Dr. Jennie Gallimore
Sponsors: AFRL, NAMRL, DAGSI
A special head-mounted camera tracks eye movements (lower center display) during flight simulation.
A significant number of aircraft accidents today are caused by pilot error. In fact, every day five accidents occur, and in Commercial Aviation, 75% are caused by human error. In General Aviation, the numbers are much higher. Unfortunately, accident investigations often do not determine specifically why the pilot made the error that caused the accident, but only that the error occurred. Dr. Gallimore, in her Aerospace and Virtual Environments Lab, is focusing on pilot error in aviation in terms of the specific human factors deficiencies that lead to aircraft accidents. Her research deals with such things as spacial awareness and disorientation, and visual and vestibular illusions.

Intelligent Agents and Robotics
  • Distance learning engineering practica with autonomous robots
  • Locomotion control of legged and winged robots
  • Autonomous Negotiating Teams
  • Space Map Generation

Medical Imaging
  • Image registration for colorectal cancer surgery
  • Segmentation and registration of MR and CT images for spinal surgery
  • Design of a surgical robot assistance for biopsy
  • Detection and visualization of
    • brain shift during surgery
    • blood vessels
  • Endoscopic Imaging

Multimedia Systems and Applications

Multimedia systems offer synergistic and integrated solutions to a great variety of applications related to multi-modality data, such as automatic target recognition, surveillance, tracking human behavior, etc. Some of the current projects are:

  • Object recognition in digital images and video
  • Multimedia content classification and indexing
  • Integrated search and retrieval in multimedia repositories
  • Background elimination in live video
  • Modeling and visualization
  • Biometrics and cyber security
  • Network and security visualization

Faculty

Relevant Courses

CS: 607, 707; CEG: 724, 725, 726


SCAN: Multimodal Information Accessing and Processing Dr. Nikolaos G. Bourbakis
Sponsors: AIIS, AFRL, ITRI

Information can be encrypted and hidden in many ways. Above, the woman's image is hidden in the noise field (upper right).
As society becomes more critically reliant on computing systems and networks to store, process and transfer information, the risk for security compromise and the destruction and modification of data increases. This research responds to this increased risk. Dr. Bourbakis is now focusing on the need to protect data in today's digital world using the SCAN methodology, and at the same time, detecting hidden messages in suspicious information using his Pixel Flow Functions developed in 1987. Valuable information is now transferred electronically by very large volumes in various forms and is vulnerable to undesirable attacks. SCAN is a tool-methodology developed in 1986 that provides security and assurance for information processing tasks. SCAN does this by using lossless compression, encryption, and information hiding on still images and video. For still images, SCAN has had an equivalent performance with JPEG-2000, while its encryption part provides a complexity of 10 75000 in years for a parallel computing machine. SCAN also provides undetectable infofor information processing tasks. SCAN does this by Information can be encrypted and hidden in many ways.

Networking & Mobile Computing
  • Ultra high-speed optical networking, WDM, OBS
  • Network protection, restoration, availability assurance
  • Wireless sensor actuator networks for detection, measurement, monitoring, tracking, visualization
  • Music tele-presence
  • Cross-layer design for QoS and high quality multimedia service in mobile wireless networks
  • Composable simulation design for biological network discovery and application

Power Electronics


(Image by David Perreault.)

The study of power electronics includes the use of solid state devices to replace traditional transformers and tubes in power supplies. Research is currently being conducted on the comparative merits of high frequency dc/dc converters. The goals are to find systems with high energy conversion efficiencies that are compatible with existing inverters and rectifiers. Investigations include the characterization and modeling of the static and dynamic operating conditions of a new composite power device called a MOS-controlled thyristor. Additionally, research is being conducted to design and prototype a backup power system for aircraft for the U. S. Air Force. (Dr. M. Kazimierczuk)


RFID Applications
  • Interface RFID, user inputs, and databases
  • Incorporate RFID in Wireless sensor networks
  • Develop target and reader localization algorithms

Robotics

The field of robotics involves the study of multifunctional manipulators designed to move material, parts, tools or specialized devices for the performance of a variety of tasks. Current research in robotics involves developing sensor-based intelligent robotic systems. This includes such diverse topics as planning the assembly of optimally manufactured parts, developing a flexible robot arm, modeling profiles of manufactured objects using the spline function, recognizing patterns through monocular and binocular vision systems and integrating data from multiple sensor sources.(Dr. K. Rattan,Dr. L. Hong)

Semantic Web

Today the World Wide Web contains tremendous information in the form of HTML files that are accessible only by interactive means. The Semantic Web initiative, a standards-based web of data and inference rules that enables machine processing, aims to make web pages understandable to machines. Current research projects include:

  • Computer assisted document interpretation tools
  • Information extraction from semi-structured documents
  • Metadata for timelining events
  • Semantic Web knowledge representation
  • Semantic sensor web

Faculty

Relevant Courses

CS: 707, 790, 875


Sensor Fusion Dr. Lang Hong
Sponsors: AFRL, DARPA

To provide adequate visibility of a complex battlefield, inputs from multiple sensor platforms must be effectively integrated to compensate for potential degredation of any single sensor.
Doing automatic target recognition and tracking (see above) on the battlefield often involves dealing with multiple targets in various locations. Variables such as foliage, weather, range, and terrain can often degrade a particular sensor. In this research, Dr. Hong is focusing on fusing together multiple sensors from a variety of sources in a coherent and integrated way. The goal is to provide enhanced target identification and assessment in situations where individual sensors may be degraded. The use of sensor fusion can substantially reduce the impact of many variables on our ability to identify, track, and assess targets.

Severe Plastic Deformation of Aluminum Dr. Raghavan Srinivasan
Sponsor: DOE

The transmission electron microscope allows detailed imaging of the molecular structures of plastically deformed aluminum.
This project focuses on developing a new process to accomplish severe plastic deformation of aluminum. The goal is to create microstructures that make aluminum much stronger, easier to produce, and less expensive to forge. Dr. Raghavan Srinivasan's focus is on using this new process to create nano structures that can result in applications such as lighter, stronger, and less expensive airframes. This particular process, which is being developed for the Department of Energy under its Aluminum Initiative, will have much broader applications. Not only will it enable the production of millions of pounds of aluminum each year, it will also be applicable to the processing of other materials.

VLSI

Very Large Scale Integrated (VLSI) circuits are microelectronic chips that contain tens to hundreds of thousands of transistors. These chips are used in a variety of applications including computers, automobiles, airplanes, and communication and robotic systems to perform signal and data processing functions. Current research in this area includes new design techniques and computer-aided design tools for incorporating a built-in self-test feature into VLSI circuits. In addition, custom VLSI circuits are being designed, fabricated, and tested for specific signal and data processing applications, such as oversampled analog-to-digital converters, parallel processors using residue arithmetic, and adaptive filters. (Dr. H. ChenDr. M. EmmertDr. S. RenDr. R. Siferd)


Wireless and Internet Security
  • Security Enhanced Protocols (RFCs)
  • Retrofitting Security into Implementations
  • Security Analysis of FreeRADIUS
  • Hardening the Firewalls
  • Lightweight secure communication protocols
  • Wireless sensor actuator networks
  • SCAN encryption-compression-hiding

Wireless, Networking, and Security

Wireless communication and networking have revolutionized the way people communicate. Currently, there are more than two billion cellular telephone subscribers worldwide. Wireless local area networks have become a necessity in many parts of the globe. With new wireless enabled applications being proposed every day, such as wireless sensor networks, telemedicine, music telepresence, and intelligent web, the potential of this discipline is just being unleashed. Recent research projects are:

  • Ultra-high speed optical network
  • Wireless sensor network
  • Music telepresence
  • Cognitive radio and dynamic spectrum access
  • Secure protocol and secure processors authentication
  • Cyber-physical systems
  • Network coding

Faculty

Relevant Courses

CEG: 602, 603, 604, 636, 802


3640 Colonel Glenn Highway - Dayton, Ohio - 45435
Wright State University