Completed MS Projects/2005-2006
From CSWiki
Contents |
[edit] Jules Nya Baweu
Tamper Detection and Localization for XML Data Using Fragile Watermarks
Advisor: H. Guo
The success of the Internet has made the communication very easy between parties and XML is one of the most used standard for information representation and exchange. A huge amount of data is crossing the Internet on the daily basis. Since the Internet is an unreliable network, XML data can be easily captured and tampered by unauthorized users. This study is proposing a scheme that uses fragile watermarking to detect and localize any modification made to an XML data. This scheme is distortion free since it doesn’t introduce any distortion to the XML data. This scheme uses an algorithm that distributes elements of the XML data into groups in a secure manner according to certain parameters. The watermark is verified without the need to have the initial XML data, and also, the embedding and the verification process in each group are independent. Experimental results and security analysis show that the chance to detect and locate change in the XML data is very high.
[edit] Laila Sherif Mahmood
Energy Aware Security Protocols in Heterogeneous Sensor Networks
Advisor: V. Crespi
The thesis focuses on the security aspect of sensor networks. Sensor networks and ad hoc networks exhibit common characteristics like lack of physical infrastructure and dynamic topology. However, since sensor networks are subjected to energy constraints, the security protocols devised in ad hoc wireless networks cannot be directly applied to sensor networks. Most of the existing security protocols in sensor networks are not energy efficient when it comes to networks that change dynamically and those requiring a quick re-establishment of group keys. We propose a security protocol that takes advantage of the heterogeneity of sensors and is suitable for dynamically changing sensor networks. The thesis focuses on the security aspect of sensor networks. Sensor networks and ad hoc networks exhibit common characteristics like lack of physical infrastructure and dynamic topology. However, since sensor networks are subjected to energy constraints, the security protocols devised in ad hoc wireless networks cannot be directly applied to sensor networks. Most of the existing security protocols in sensor networks are not energy efficient when it comes to networks that change dynamically and those requiring a quick re-establishment of group keys. We propose a security protocol that takes advantage of the heterogeneity of sensors and is suitable for dynamically changing sensor networks.
[edit] Kamyar Miremadi
Building a Peer to Peer Message Passing Environment by Utilizing Reflection in .NET
Advisor: B. Parviz
Nowadays many projects based on peer to peer computing (one form of grid computing) are being conducted world widely. Grid computing (or the use of a computational grid) is a system which enables the users apply the resources of many computers in a single network endeavoring to find solutions to a single problem at the same time — usually to a scientific or technical problem that requires a great number of computer processing cycles or an access to large amounts of data.
MPI is a standard protocol that is widely utilized in many of the parallel and distributed applications. In other words, MPI is a communicating programming model for general purposes, particularly in parallel and distributed applications .Our solution benefits from .Net Web Service Technology for Communication Services and Reflection for Execution Model. Reflection is the ability of a program to modify its own behavior in the course of its evaluation.
This study implements class libraries in .NET to make them available to the programmers and enable them to write a quick and easy to use parallel and distributed program as well as internal classes that are used in our communication and execution model.
A poster based on this work was presented at The 2006 International Conference on Parallel and Distributed Processing Techniques and Applications, Las Vegas, June 26-29, 2006.
[edit] Chris Orona
CSNS: Computer Science Network Services
Advisor: C. Sun
The Computer Science Department at Cal State LA has been planning services that would handle many of the needs of the students and faculty to enhance the learning process. Currently some of these needs are met by several different sites maintained by the department, including a turn-in server for assignments, a CSNetwork site for student surveys and department announcements, a server that manages information for graduate students, and a server run by the ACM Student Chapter which features discussion forums for many courses.The goal of the CS Network Services (CSNS) Project is to consolidate these services into a single site, making things much easier for students, faculty, and staff as they would only have to check a single site for all of their CS-related academic needs. …
The system is a web application which will use the Java 2 Enterprise Environment (J2EE) along with the popular open-source Spring Framework.
[edit] Robert Ritchey
Measuring With Jugs: A Solution in Finite Domain
Advisor: R. Abbott
The “measuring with jugs problem” involves pouring water between jugs to measure out a specific quantity of water. The object is to find the proper sequence of fill operations, pour operations (from one jug to another) and emptying operations to arrive at the desired quantity of water which may be in either jug. In the past, solutions to this puzzle have always involved creating a series of operations by computing various possible paths and choosing the one that leads to a solution. This paper takes a different, novel route to the solution to this. An answer is found numerically using constraint programming and the results are then translated into a path. At first, a simplified version of this puzzle using only two jugs will be discussed. A mathematical solution to this problem will be presented and the solution will then be translated into code in the Mozart-Oz programming language.
A paper based on this work was presented at The 2006 International Conference on Artificial Intelligence, Las Vegas, June 26-29, 2006.
