The MERIT Summer Research Program
 
 
 
 
 
 
 
 
 
 
  The A. James Clark School of Engineering
  University of Maryland Home

RITE Project Descriptions: Summer 2007


1. Automatic Speaker Recognition
Prof. C. Espy-Wilson

The need for robust speaker identification and speaker verification systems has become even more critical in recent years given the nation's security issues. Research in the department in this areas focuses on the development of digital signal processing algorithms that capture speaker-specific information and the development of acoustic models that can deal with limited training data. Typical subprojects that are suitable for undergraduates include the development of algorithms and speaker-specific parameters for speaker identification, integration of algorithms into the speaker verification system, evaluation and comparison of different algorithms for speaker identification.


2. Cell Clinics for Bioelectric Interface with Single Cells
Prof. P. Abshire

Interfacing electronics to biological systems leads to the possibility of creating devices capable of being used as biosensors, environmental monitors, and hybrid bioelectronic computational engines. The potential applications in healthcare, security, and scientific research are numerous. Students will contribute to ongoing research in the Integrated Biomorphic Information Systems Laboratory to develop such bioelectronic and biophotonic interfaces to single cells. These interfaces consist of microelectronic circuits integrated together with micromechanical structures for cell manipulation and capture. Subprojects include circuit design and simulation, characterization of integrated circuits, comparison of integrated measurements with traditional physiological techniques such as patch clamp, design and construction of PC boards and interfaces, and development of methods for cell manipulation. This project involves review of electronics and biological literature in addition to design, simulation, testing, cell culture and physical construction.

Prerequisites: Integrated circuits and/or MEMS and/or physiology experience


3. Detection of Humans Carrying Concealed Objects
Prof. R. Chellappa

Computer vision methods for detecting humans carrying concealed objects will be developed. Using kinematic models for human motion, we will estimate the joint angles and torques to then determine the loading conditions on different parts of the body. Methods for synthesizing human motion under various loading condition will also be developed.


4. High Speed Optical Switching for Fiber Optic Communications
Prof. J. Goldhar

The student would participate in optical testing and characterization of novel optoelectronic devices, data acquisition, data analysis and comparison of results with computer simulations. The research would be conducted at the Laboratory for Physical Science which is a 15 minute walk from campus. It is also reachable by the UM shuttle.

Prerequisites: U.S. citizenship and ENEE 322-Signal and System Theory and ENEE 381-Electromagnetic Wave Propagation are the courses directly relevant to the research. Without these courses, students with a solid math and physics background should be able to learn the essential theory as needed during the project.


5. Maryland Optics Group Summer Research Project
Prof. C. Davis

Participants in the UMCP Electrical and Computer Engineering MERIT program will work on the fundamental physics, operation, and applications of high bandwidth (>1Gb/s) directional wireless links using free space optical (FSO) (laser links through the atmosphere) and radio frequency transceivers. Fundamental issues that affect FSO links include scintillation, spatial, and temporal correlation functions. The primary application for which these directional links are used in out test network is for real-time, high definition video surveillance, with automatic object recognition and event detection. Students interested in software development for real-time motion detection, size discrimination, vehicle speed, crowd formation, and other measures of ?events? in a high resolution HDTV image will find this project timely and fascinating.


6. Polarization Dependence in Nonlinear Fiber Optics: Modeling and Experiment
Prof. T. Murphy

Most currently-used methods for modeling nonlinear propagation in fiber optic do not account for the evolution of the polarization state. A software model has been developed that accounts for the variations in polarization via a numerical approximation and predicts the behavior of a system with different input polarization states. The model uses two methods, an approximation and an exact equation, which have been derived from the nonlinear Schrodinger equation. The approximation closely agrees with experimental data in which the polarization state frequently changes compared to the length of the fiber. The exact method is free of this constraint but takes longer to simulate. The software tool developed for this project will help to better understand experimental observations of polarization dependence, and will help in designing nonlinear optical switches that are insensitive to polarization.



7. Pursuit-evasion and Swarming Games with Robots
Prof. P. S. Krishnaprasad

In this project, students will explore robot-robot interaction via the study of pursuit-evasion and swarming problems. These will be cast as games with dynamic models, constraints and payoffs. The exploitation of localization technology such as Cricket for indoors and GPS outdoors will be part of the experience. Students will learn to write principled feedback control laws in high level motion control languages to investigate robot-robot interactions. They will learn how to use modern estimation and tracking methods to get the most out of sensors. Interested students may wish to examine the website created by students in a similar project in summer 2006.


8. Software Toolbox for Media Security and Forensics
Prof. Min Wu

In the recent decade, new devices and powerful software have made it possible for consumers worldwide to access, create, and manipulate multimedia data. Internet and wireless networks offer ubiquitous channels to deliver and exchange such multimedia information. However, the potential offered by the information technology era cannot be fully realized without a guarantee on the security and protection of multimedia data. The focus of this project is on security and the forensic technologies to ensure that multimedia content is being used by authorized users for authorized purposes and to gather solid evidence to hold violators accountable. When security is compromised, intellectual rights are violated, or authenticity is forged, forensic methodologies and tools are employed to reconstruct what happened to the content.

Analogous to human fingerprints in criminal forensics, the ?fingerprints? in the digital domain take two forms, intrinsic fingerprints that are naturally left by certain processing operations and extrinsic fingerprints that are proactively inserted by system designers. These digital fingerprints can be exploited to provide evidence on the violations to media content, by whom, when, where, and how. Students participating in this project will be exposed to many exciting results from the latest research and contribute to the development of a software toolbox for media security and forensics. The toolbox will provide a collection of reusable modules, many of which are common in handling different types of multimedia signals (such as image, video, and audio) and in different applications (such as tampering detection and traitor tracing). The successful deployment of this toolbox will enable fast prototyping of media security and forensic systems for new applications with a variety of multimedia signals.