2007 Projects
A Wearable Device for Increasing the QoL for the Visually Impaired
Advisor: Dr. Hanna
According to the American Foundation for the Blind, there are approximately 10 million blind and visually impaired people in the United States. Walking from place to place is challenging for these people, some of whom use long canes and leader dogs to navigate on foot. The new generation of visually impaired citizens are becoming more sophisticated with more than 1.5 million visually impaired computer users, increasing every year. Little research has been done into wearable medical devices for the visually impaired to assist them with mobility and other prevalant challenges that they face. This research will be towards such a wearable device including issues in ergonomics, environment, use, and wearability, cost, and power consumption. The goal of this work is to significantly increase the quality of life for a visually impaired subject. To assist in the work, an individual who has recently become blind will be a part of the team. Two iterations of the device will be developed based on this research, feedback from the user, and results of a quality of life survey.
Lecture Notes |
Drug Discovery using Machine Learning
Advisor: Dr. Zohdy
In this research project, the Reu students that are selected will have the opportunity to work on exciting and cutting edge research ,on drug actions and discovery by using sophisticated machine learning ,from abundant genomic data. There are three main components of interest; one is the knowledge available about drug targeting,the second is how to apply neural network based machine learing ,and the third has to do with outcome prediction of drugs in the face of variations and uncertainties.
Lecture Notes |
Virtual Nursing Environment for Safe Decision Making
Advisor: Dr. Mili
Practical training is an essential component of nursing students preparation to entering the workforce and making decisions that impact patients health and sometimes lives. The quality of the learning that results from practical
training depends, among other things, on 1) the variety of cases the student is exposed to, and 2) the quality of feedback they receive directly from the patients response to their actions,and indirectly from the verbal and written comments of the supervising faculty. Both of these factors can vary dramatically resulting in important gaps in a graduating nurses preparedness. In this project, we propose to complement the practical training with a virtual trainer. The VI-MED system generates a set of virtual patients whose distribution of symptoms and other
parameters is realistic yet ensures a full coverage of cases including low frequency cases. The VI-MED system keeps a complete log of the patients evolution and reaction to the students actions. Thorough feedback can then be generated from the log. This feedback can be used alone or complemented with comments from supervising nursing faculty. The VI-MED system is developed using Game Studio 6. In this paper, we discuss two technical aspects of the system: The dynamic patient model and the feedback component. The challenge in both of these components arises from the level of uncertainty that permeates health care decision making.
Lecture Notes |
Intrusion Classification, Modeling and Detection for Wireless Sensor Networks.
Advisor: Dr. Fu
As a relatively recent and emerging technology, wireless sensor networks (WSNs) are beginning to be deployed frequently in a wide variety of environments ranging from military and emergency environments to natural and embedded environments. However, WSNs are also of particular interest to adversaries due to their frequent deployments in open and unprotected environments. Denial of Service (DoS) attack is a particularly great threat to WSNs. An effective DoS attack merely has to deplete the resources of the nodes to render them unavailable. WSNs are particularly vulnerable to failure under such attack because of their limited energy, processing capacity, and storage. This project addresses the need for developing intrusion detection techniques for WSNs. Specific tasks that will be carried out under this project are to develop techniques in intrusion classification, modeling, and detection for WSNs. In particular, we will design and implement a preliminary intrusion detection system (IDS) for WSNs.
Lecture Notes |
Development of Computer-Aided Design Software for Reconfigurable Hardware.
Advisor: Dr. Debnath
Programmable logic devices can be reconfigured again and again to
implement different digital systems. Such reconfigurable hardware
can often accelerate a wide variety of compute-intensive applications
because the hardware can be customized through reconfiguration based
on applications. Moreover, reconfigurable hardware can make digital
systems more flexible and fault-tolerant because functionality of
the systems can be changed if necessary even after their deployment.
The objective of the proposed research is to develop a computer-aided
design (CAD) software that can efficiently synthesize digital systems for
implementation in reconfigurable hardware such as complex programmable
logic devices (CPLDs) or field-programmable gate arrays (FPGAs).
Lecture Notes |
Road Testing of UWB ABS Sensor Network.
Advisor: Dr. Li
Our study has shown that UWB technology can provide the capacity and the signal quality desired for intra-vehicle sensor networks. We are currently developing a prototype UWB sensor network for a 4-channel anti-lock breaking system. An indoor 4-channel ABS system has been built. In this REU project, we are going to perform road testing of the developed system and measure the intra-vehicle UWB channel in the in-stationary environment. This UWB sensor network testbed can support future research and development in intra-vehicle sensor networks for various sub-systems and for the integration of many subsystems.
Lecture Notes |
