UVM Computer Science Research Day

August 25, 2006

Pictures of the event are now available.
All events held in Votey 367 unless specified otherwise

Schedule of Events

11:00AM: UVM CEMS Dean's Opening Remarks

11:15AM Keynote Presentation

12:15PM: UVM Faculty Presentation

12:35PM: UVM Faculty presentation

1:00PM: Lunch, Votey 351

2:30PM: Distinguished Alumni Presentation

3:30PM: UVM Graduate Student Presentation

3:50PM: Coffee Break, Votey 367

4:15PM: UVM Graduate Student Presentation

4:35PM: UVM Graduate Student Presentation

5:00PM: UVM CS Chair's Closing Remarks


How to Build Secure Network Information Systems
Trevor Jim, AT&T Labs

The Domain Name System (DNS) is one of the fundamental services of the Internet. It was developed in the early days of the Internet, and was designed primarily for functionality, with security as a lesser concern. Recent attacks on DNS have demonstrated that its vulnerabilities are actively being exploited. I will describe some of these attacks on DNS, the efforts underway to secure it, and the technical and economic hurdles that are preventing deployment of a secure solution.

This background motivates a new approach to building distributed information systems like DNS. We adapt off-the-shelf database query languages to implement a peer-to-peer network of database servers. The peers can easily be programmed to provide a variety of secure network services, including directory services, public key distribution systems, distributed access control systems, and network monitoring systems. I will illustrate the power and flexibility of our system by writing the core of DNS resolution in a dozen lines of code.

Status of the VACC
X. Sean Wang, UVM

The Vermont Advanced Computer Center (VACC) is alive and running (literally). I'll talk about its status, and how to take advantage of its resources.

Putting Curry-Howard Theory to Work
Tim Sheard, Portland State University

There has been a lot of recent interest in exploiting the Curry-Howard isomorphism in type systems for more or less traditional programming languages. Types based upon the Curry-Howard isomorphism can express precise properties of programs. Such properties can be either functional or non-functional properties. I.e. they can constrain either the output of the program, or the resources needed to produce the output, or both. In the talk I will introduce the basic ideas from scratch, and give many, many examples that illustrate its use. The talk is not addressed at experts in type systems, but to general computer scientists who are interested in describing properties of their programs precisely.

Our goal is to build a system in which the specification of designs, the definition of properties, the implementation of programs, and the checking that programs adhere to their properties, are all bundled in a coherent manner into a single unified system that appears to the user to be a programming language. We hope to use our system as a broad spectrum language, capable of handling abstract properties as well as implementation minutiae, where the connection between properties and programs is formal and precise.

A Key Predistribution Framework for Wide-Area Wireless Sensor Networks
Katerina Simonova, UVM

Limited resources and unavailability of pre-deployment information make authentication and maintaining confidentiality in Wireless Sensor Network (WSN) a challenging problem. WSN could be deployed in adverse environment, where message protection from eavesdropping is essential. However, well-developed asymmetric cryptography is unsuitable due to substantial resources requirements. Presently, Key Predistribution schemes that provide private keys to sensors prior to deployment is the prevalent technique used to secure communication in WSN. Early key predistribution schemes try to guarantee high connectivity between every pair of sensors in the network. These schemes perform well in small networks, but do not scale well when the size of the network increases. In order to solve the scalability problem, we note that, in wide-area WSN providing high connectivity between every pair of sensors is unnecessary as each sensor could reach only small fraction of sensors. We propose the framework that achieves substantial performance enhancement in wide-area deployment.

Energy-Efficient Resolution Control for Processing of Precision-based Queries in Wireless Sensor Networks.
Biyu Liang, UVM

Tiny wireless sensors have made large scale scientific observation systems possible. However, these tiny sensors operate on limited battery power, and energy efficiency remains a big challenge in very large scale dense sensor network applications, where the network is usually over deployed to extend network life and improve robustness. In this talk, we discuss the issue of energy efficiency in processing of precision-based queries in sensor networks from two aspects: (1) minimizing the amount of data needed to obtain query results of desired accuracy; (2) minimizing the control overhead of gathering this amount of data from sensors.