Project Description:

Distributed Computing Systems for Science and Engineering

The Cooperative Computing Lab (CCL) at the University of Notre Dame creates, operates, and studies distributed computing systems consisting of hundreds to thousands of computers.  Such distributed systems are used to attack "grand challenge" problems in nearly every field of scientific study.  Our lab works with users in fields such as astronomy, biology, biometrics, data mining, physics, and mathematics to solve problems that are too large or too long running to attack using a single computer.

Summer RET projects with the CCL will introduce participants to the use of large distributed systems to attack scientific problems. Participants will have direct access to a 500-CPU computing cluster, and the ability to harness thousands more CPUs using national computing facilities. Summer projects will be tailored to the skills and interests of the applicant, but may include deploying a new application, improving performance or reliability, or studying usage patterns.

A background in computer science, physical sciences, life sciences, or mathematics is necessary.  Some experience in programming is required. More information is available online at http://www.nd.edu/~ccl and from Douglas Thain, dthain@nd.edu.

Interference mitigation/management in wireless communications: 
This project will study how various kinds of interferences occur in wireless communication systems and examine techniques that are employed to mitigate those interferences to offer improved communication, e.g., improve data throughput rates or enhance signal
quality or reduce drop-off rates in a celluar call.   In cellular systems, there are at least two common sources of interferences:  1.  intersymbol interference due to the physical phenomenon that multiple copies of signals travel through multiple paths which lead to
destructive interference at the receiving end;  2.  inter-cell and intra-cell interference due to the need for each cell to serve numerous customers with limited bandwidth.  Conventional approaches to mitigating such interferences include adaptive equalization and adaptive beamforming.  Central to those methods are signal estimation and detection based on statistical principles.  For example, statistical estimation method is needed to estimate the channel state information (such as signal-to-noise ratio) to design and adjust the equalizer and the beamformer to match the channel condition.  This project will include studies of statistical estimation methods for those applications.
Yih-Fang Huang

Adaptive filtering for biological systems:
Adaptive filters have been employed to the development of various medical instrumentation including diffuse optical tomography (DOT), electrical impedance tomography (EIT) and heart-rate monitoring systems.  Among various adaptive filtering techniques, Kalman filter has been studied extensively and some studies have shown that Kalman filter can be used to improve the resolution of DOT and EIT.  One of the difficulties associated with the implementation of Kalman filter, however, is numerical instability.  This project aims to explore further the suitability of Kalman filter and develop innovative variations of Kalman filter for improved suitability to medical applications.
Yih-Fang Huang