RITE Project Descriptions: Summer 2005
1. Artificial
Bat Echolocation for Biological Modeling and Robotics
Prof. T. Horiuchi
The Computational Sensorimotor Systems Lab (CSSL) will have a project this summer for 1-2 MERIT undergraduates to design, simulate, and construct circuits and PC boards for a small acoustic sonar system for investigating biological algorithms of bat echolocation. This project will involve some reading of electronics as well as biological literature in addition to the circuit simulation and physical construction. Integrated circuit design may be possible. This work will support on-going research in the lab and will involve students directly in the research topics of the lab.
Course Prerequisites: Junior-level analog electronics and signal processing (Equivalent to UMCP courses ENEE 312 and ENEE 322).
2. Automatic Speaker-Independent Speech Recognition
Prof. C. Espy-Wilson
There is an increasing demand for automatic speaker-independent speech recognition solutions for desktops and wireless handheld devices. The development of accurate, efficient, robust and speaker-independent digital signal processing algorithms for speech recognition is the subject of intensive on-going research in the Department. Many of the research activities involved, such as the evaluation of algorithms across different databases and the development of fast algorithms, are suitable projects for undergraduate students. Typical subprojects include the conversion of MATLAB code into C, the evaluation and comparison of different algorithms in the various components of the recognition system, testing of the recognition system on different languages, and the application of these algorithms on environmental sounds to see how well they are characterized and distinguished from speech.
3. 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 for speaker normalization, integration of algorithms into the speaker verification system, evaluation and comparison of different algorithms in the various components of the system.
4. Biometrics and Robot Control
Prof. P. Abshire and
Prof. T. Horiuchi
It is possible to use a variety of measurements from the human body in order to control the motion of external robotic devices. In this project students will refine methods for multi-channel electromyography (EMG) to control a robotic device. Limitations of existing systems are few control channels, slow response time, and limited sensory feedback to the user.
Subprojects include circuit design and simulation, characterization of EMG signals, signal classification techniques, design and construction of PC boards and interfaces, and development of robotic interface. This project involves review of electronics and biological literature in addition to design, simulation, testing, and physical construction.
Prerequisites: analog circuits and/or controls and/or MATLAB programming experience and willingness to wear skin surface electrodes.
5. 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, 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, and physical construction.
Prerequisites: integrated circuits and/or MEMS and/or physiological experience.
6. Design of Wireless Ultra-Wideband Communication
Prof. R. Liu
The universal serial bus (USB) is now making inroads into computer systems. Almost every computer that one buys today comes with one or more USB connectors on the back. However, all of current USB connectivities are wired. All of USB devices, such as printer, scanner, external hard drive, and digital camera, should be connected to computers with cables. One challenging problem is that the wireless USB should provide the same high-speed and effective connections as the wired USB with a high data rate of 480 Mbits/s. In order to accommodate the high data rate, one possible solution is to use the emerging ultra-wideband (UWB) transmission technique, a wireless technology designed for short-range, personal area networks, or PANs. Recently, the Federal Communications Commission (FCC) has mandated that UWB radio transmission can legally operate in the range from 3.1 GHz to 10.6 GHz, at a transmit power of -41dBm/MHz. The use of UWB technology under the FCC guidelines.
In this project, we propose to design wireless UWB transmission systems to support high data rates from 55 Mbits/s to 480 Mbits/s. The proposed research consists of the following issues. First, the trade-offs between dierent existing UWB transmission techniques will be investigated. There are two general ways to use the 7.5 GHz license-free UWB spectrum, resulting in dierent UWB transmission methods that can fundamentally affect the design of wireless USB. Second, different system design issues such as low-complexity constraints, strict power limitations, and scalability and flexibility will be addressed. We will modify the existing UWB scheme to accommodate all of the system design issues. Third, a wireless USB system will be implemented. We will develop softwares that include all of the algorithms deployed at both transmitter and receiver sides. With FPGA boards provided by our industrial partner Maryland Semiconduct Inc. (MSI), we will transit the softwares into the FPGA boards. Together with MSI's high-frequency RF technique, we plan to build a test-bed to test the performance of the high-speed wireless UWB system.
7. Human Movement Modeling Using a Multi-Camera System
Prof. R. Chellappa
Modeling and analysis of human movement using multiple cameras is a challenging problem with applications in sports video analysis, early diagnosis of movement-related disorders and monitoring the recovery of patients who have undergone knee surgery. In this summer's project, the student will be expected to help with developing software for self-calibration of multiple cameras used for data acquisition as well as work on algorithms for extracting the kinematics of joints and body parts using non-rigid models.
8. Machine Recognition of Disquised Faces
Prof. R. Chellappa
Although significant research has been done over the past fifteen years in the area of machine recognition of faces, we still do not have a working algorithm for recognizing disguised faces. In this effort, the summer student will develop novel algorithms using a parts-based face recognition approach for handling disguised faces.
9. Multimedia Watermarking and Steganography
Prof. M. 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 to exchange such multimedia information. However, the potential offered by the information technology era cannot be fully realized without the guarantee on the security and protection of multimedia data. Digital watermarking are schemes to embed imperceptible data in digital multimedia for a variety of applications, including ownership protection, tampering detection, access control, and annotation. With a few variations, the same technologies can be used as steganographic tools for covert communications.
This project is to develop a software toolbox for watermarking and steganography. The toolbox will provide a collection of reusable modules, many of which are common in watermarking different types of multimedia signal (such as image, video, and audio) and in different applications (such as ownership protection, tampering detection, and digital fingerprinting). The successful deployment of this toolbox will enable fast prototyping of watermarking and steganography systems for new applications with a variety of multimedia signal. Besides being exposed to exciting results from latest research and the state-of-the-art development tools, students in this project will be able to create their own pictures and music with invisible images and silent messages hidden in.
10. Resource Allocation for Wireless
Communication Networks
Prof. S. Ulukus
The number of wireless data customers, as well as their demand for higher and higher data rates, has made efficient allocation of resources of paramount importance. One important characteristic of wireless communication networks that distinguishes them from wired networks is that in wireless communication networks the transmissions of multiple users take place in the same fixed frequency bandwidth, that is, users' communications interact with each other through unintentional interference they create for each other. A main thrust of this project is in the allocation of physical layer resources to a network of mobile users (terminals) communicating with fixed information sources (base stations in cellular wireless networks) and investigations of energy-aware, energy-efficient, and energy-constrained wireless networking. The controllable resources in the physical layer are transmit powers, transmit waveforms, number of parallel transmissions, modulation constellation sizes, error correction coding mechanisms, and receiver filters of the users. The individual or joint optimum selection of these resources for individuals or networks of users has been an important research issue. The research typically involves mathematical modeling of the particular problem, solving it, and ultimately coming up with distributed algorithms to implement the solution. This area of research is very versatile in terms of finding research problems for undergraduate students with various backgrounds. Depending on the problem formulation (e.g., the set of resources considered, and system and implementation constraints), one can come up with problems with varying degrees of difficulty, and involving varying amounts of mathematical analysis and practical implementation aspects. We expect that selected undergraduates will work together with graduate research assistants under faculty supervision to contribute to these tasks.
11. Sensing and Control for Mobile Robots
Prof. P. S. Krishnaprasad
Knowledge of a robot's own state as well as the state of the environment is necessary to devise safe guidance and control laws for mobile robots in complex environments (that may include other robots, stationary obstacles, people, etc.). Robots operating outdoor can be given access to the global positioning system (GPS) to determine location to varying degrees of precision. Since GPS signals do not penetrate indoor, alternative methods of localization are being explored, including the specialized application of ultrasound and radio frequency beacons, vision systems, and wireless carrier signals used for computer networks. In this project, students will explore the nature and use of such location sensing techniques for mobile robots operating indoor and outdoor, together with feedback control laws that are appropriate for these differing modes of location sensing. They will also explore how to couple location sensing to maps, ultrasonic obstacle sensing etc. Students will develop control software using MDLe, a motion control language for robotics developed in the Intelligent Servosystems Laboratory (ISL). Strong programming skills and interest in algorithm development are a must. Students with interest and skills in hardware will be given an opportunity to contribute to an ongoing project on a hovercraft robot under development in ISL.
Students participating in this project will receive instruction in the basic principles of location sensing, GPS, robot motion control, and allied topics. They will have access to several commercial robots equipped with ISL software. They will work with graduate students in the lab and outdoor to collect data, test control and sensing algorithms, and investigate subtleties that arise when complex physical systems interact with software.