Mar 19, 2008
2:30pm - 3:30pm

Goodwin 524

Software architectural evaluation becomes a familiar practice in software engineering community for developing quality software. Architectural evaluation reduces software development effort and costs, and enhances the quality of the software by verifying the addressability of quality requirements and identifying potential risks. There have been several methods and techniques to evaluate software architectures with respect to the desired quality attributes such as maintainability, usability and performance. This paper presents a discussion on different software architectural evaluation methods and techniques using a taxonomy. The taxonomy is used to distinguish architectural evaluation methods based on the artifacts on which the methods are applied and two phases (early and late) of software life cycle. The artifacts include specification of a whole software architecture and its building blocks: software architectural styles or design patterns. The role of this paper discussion is to review different existing well known architectural evaluation methods in order to view the state of the art in software architectural evaluation. This paper also concentrates on summarizing the importance of the different evaluation methods, similarities and difference between them, their applicability, strengths and weaknesses.

Mar 11, 2008
2:30pm - 3:30pm

Goodwin 524

This summary paper describes the general properties and features of two systems of semantic representation: Richard Montague's Intensional Logic and Ray Jackendoff's Conceptual Structures. Each system of representation is based upon a different type of semantic theory, and thus each one is meant to express different semantic and linguistic phenomena. The basic concepts of how each formalism works and their expressiveness will be explored. Several linguistic phenomena will be explained and used as test cases to demonstrate the expressiveness and limitations of each semantic formalism. Additionally, the first order predicate calculus is explained and used to express linguistic phenomena as an aid for the reader.

Nove 23, 2007
12:30pm - 1:30pm

Goodwin 521

Wireless networks have made great gains in usability and popularity. However, inherent limitations on cell capacity and coverage still exist. There are also dead-spots and hotspots problems in these networks. Ad hoc multi-hop relaying enhances cell capacity and coverage, alleviates the dead-spots problem, and helps to ease congestion in hotspots. However, multi-hopping also increases packet delay. Effective channel assignment is key to reducing delay. Existing channel assignment schemes are heuristics and may not guarantee optimal solutions in terms of minimum delay. In this paper, we provide an optimal channel assignment (OCA) scheme for ad hoc TDD W-CDMA multi-hop cellular networks (MCN) to minimize packet delay. OCA can also be used as an un-biased tool for the comparison among different network topologies, network densities, and protocols. To the best of our knowledge, this is the first time that a minimum delay optimal channel assignment is proposed in a multi-hop cellular environment.

Sept 25, 2007
2:30pm - 3:30pm

Goodwin 524

We outline a research proposal whose overall goal is to contribute to the definition of a formal semantics for UML, and indeed visual behavioural modelling languages in general. Specifically, we aim to validate the three-layer semantic architecture, used as a way of explaining the behavioural semantics of UML. The validation includes a definition of the semantics of UML actions and activities, as well as a prototype interpreter.

Sept 5, 2007
1:30pm - 2:30pm

Goodwin 524

Abstract: Selecting an informative subset of SNPs, generally referred to as tag SNPs, to genotype and analyze is considered to be an essential step toward effective disease association studies. However, while the selected informative tag SNPs may characterize the allele information of a target genomic region, they are not necessarily the ones directly associated with disease or with functional impairment. To address this limitation, we present a first integrative SNP selection system that simultaneously identifies SNPs that are both informative and carry a deleterious functional effect – which in turn means that they are likely to be directly associated with disease. We formulate the problem of selecting functionally informative tag SNPs as a multi-objective optimization problem and present a heuristic algorithm for addressing it. We also present the system we developed for assessing the functional significance of SNPs. To evaluate our system, we compare it to other state-of-the-art SNP selection systems, which conduct both information-based tag SNP selection and function-based SNP selection, but do so in two separate consecutive steps. Using 14 datasets, based on disease-related genes curated by the OMIM database, we show that our system consistently improves upon current systems.

Aug 21, 2007
2:30pm - 3:30pm

Goodwin 524

Abstract: In computer-assisted surgery, a computer display provides 3D guidance of tools during surgery. However, current guidance displays have a number of drawbacks. This thesis considers the advantages of a prototype guidance system for drilling tasks. We attached a small LCD screen and video camera to a surgical drill. The screen's point of view was controlled by moving the drill. A user study showed that the tool-mounted LCD screen can provide significantly better positional and angular accuracy than the conventional display. The study also showed that a video camera might not be useful for guidance.

Aug 9, 2007
2:30pm - 3:30pm

Goodwin 524

Abstract: We look at a variant of the Hamilton circuit problem, where the input is restricted to hexagonal grid graphs. A hexagonal grid graph has a vertex set that is a subset of the grid points of a regular hexagonal tiling of the plane and edges corresponding to hexagon sides. We show that Hamilton circuit in hexagonal grid graphs is NP-complete.

Jun 19, 2007
2:30pm - 3:30pm

Goodwin 524

Security is a critical issue for software systems, especially for those systems which are connected to networks and the Internet, since most of them suffer from various malicious attacks. Intrusion detection is an approach to protect software against such attacks. However, security vulnerabilities that are exploited by intruders cut across multiple modules in software systems and are difficult to address and monitor. Aspect-oriented software development (AOSD) is a programming paradigm that enables modularization of these kinds of concerns, called cross-cutting concerns. A number of work have utilized AOSD to address security issues of software systems, but none of them has employed AOSD for intrusion detection.

In this thesis, we propose an AOSD framework for developing intrusion-aware software systems. We identify and specify vulnerabilities and potential attacks against the target software systems. These attack scenarios and intrusion detection aspects are modeled using an aspect-oriented Unified Modeling Language (UML) profile. Based on the aspect-oriented attack scenario model, the intrusion detection aspects are implemented and woven into the target system via an aspect-oriented programming Language. The resulting target system is an intrusion-aware system, which has the ability to detect the intrusions automatically. We present an experimental evaluation by applying this framework for some of the most common attacks included in the Web Application Security Consortium (WASC) web security threat classification. The experimental results demonstrate that the framework is effective in specifying and implementing intrusion detection and can be applied for a wide range of attacks.

Jun 12, 2007
2:30pm - 3:30pm

Goodwin 524

The weird quantum mechanical effects governing the behavior of sub-atomic particles are about to revolutionize the way we perform computation and manipulate information. This thesis is a testimony to the indelible mark that quantum mechanics has already left on computer science. Specifically, we have investigated some of the consequences of manipulating information at the quantum level on data security, parallel processing, universality and computability.

The importance of parallelism for quantum information processing and its consequence on universality is demonstrated through a series of evolving computing paradigms for which only a parallel approach can guarantee a reliable solution. We also bring a necessary clarification to the much disputed problem regarding the comparison between the computational power of a quantum machine and that of a conventional computer.

Jun 4, 2007
12:00pm - 1:00pm

Goodwin 524

In this paper, we survey kinetic visibility problems. Unlike the research in traditional visibility problems, the researchers have also taken moving objects into consideration. These problems have been addressed from different practical aspects, such as computational graphics and robotic design. Over the years, some general combinatorial results of the visibility graph have been shown. Leveraging temporal coherence has been used in kinetic visibility studies for different scenes. Data structures that have been developed to support the kinetic operations, and kinetic algorithms that have been designed to achieve efficient visibility computation and maintanence are reviewed in this paper. Open problems related with kinetic visibility will be addressed from different perspectives. Both theoretical and practical work needs to be done in future studies.

May 3, 2007
2:30pm - 3:30pm

Goodwin 521

Many orthopaedic procedures performed on the knee require the segmentation of bones from CT images. Four features of the CT data cause these difficulties: (i) gaps or weakly defined edges in the cortical bone; (ii) diffused bone boundaries; (iii) non-uniform texture of the bone across its surface and (iv) narrow channels between bones. In addition to these, the variability of bone tissue (from spongy trabecular bone to dense cortical bone) in the knee further contribute to the difficulty of segmentation of bone. We present a proof of concept study for a hybrid semi-automated segmentation technique for the knee. The hybrid approach combines the use of active contours and a knee atlas constructed from MRI data. The active contour method that we used overcomes the problems of small capture range and inability to converge into concavitites, which are common with active contour methods. The atlas provides a good initial approximation of the bone geometry, and can provide anatomical labelling for a specific patient's anatomy. Our segmentation is performed in three dimensions (rather than on a per-slice basis, as is typical with CT segmentation), giving a smoother segmentation, minimizing the user interaction, and reducing segmentation time.

March 19, 2007
1:30pm - 2:30pm

Goodwin 524

This dissertation presents a framework for the study of passive knee kinematics. Computational models for solving both the forward and the inverse kinematics of a prosthetic knee joint were formulated. For forward kinematics, the model computed the predicted knee motion as a function of joint parameters; these parameters included patient-specific ligament data as well as the design and the surgical placement of the prosthesis. The model also provided three-dimensional graphics visualization of the predicted kinematics. It was applied to two commercially available prosthetic designs and was validated experimentally. Simulation results suggested several heuristic rules for surgical strategies for maximizing the post-operative knee function.

March 8, 2007
1:30pm - 2:30pm

Goodwin 524

As a result of advances in hardware technology such as multi-core processors there is an increased need for concurrent software development. Unfortunately, developing correct concurrent code is more difficult than developing correct sequential code. This difficulty is due in part to the many different, possibly unexpected, executions of the program, and leads to the need for special quality assurance techniques for concurrent programs such as randomized testing and formal state space exploration. This talk focuses on the complementary relationship between such different state-of-the art quality assurance approaches in an effort to better understand the best bug detection methods for concurrent software. An approach is presented that assesses testing and formal analysis tools using metrics to measure the effectiveness and efficiency of each technique at finding concurrency bugs. Using program mutation, the assessment method creates a range of faulty versions of a program and then evaluates the ability of various testing and formal analysis tools to detect these faults. The approach is implemented and automated in an experimental mutation analysis framework (ExMAn) which allows results to be easily reproducible. To demonstrate the approach, I will present the results of a comparison of testing using the IBM concurrency testing tool ConTest and the NASA formal state analysis system Java PathFinder. I will conclude with a discussion of other potential applications of this research as well as areas for future work.

January 17, 2007
2:30pm - 3:30pm

Goodwin 521

DNA computing is an unconventional branch of computing that uses DNA molecules and molecular biology experiments to perform computations. This thesis evaluates the feasibility of implementing molecular codebreaking, a DNA computing technique that uses a known-plaintext attack to recover an encryption key, and double encoding, a proposed error resistance technique that is designed to reduce the number of false negatives that occur during DNA computation. This thesis evaluates molecular biology experiments that make up a molecular codebreaker, a theoretical DNA computer that has never been attempted in the laboratory until now. Molecular techniques such as ligation, gel electrophoresis, polymerase chain reaction (PCR), graduated PCR, and bead separation were carried out, reported upon and found to be feasible with certain reservations. An implementation of the error resistance technique of double encoding, where bits are encoded twice in a DNA strand, was designed and attempted. Although the implementation was unsuccessful, several issues associated with double encoding were identified, such as encoding adaptation problems, strand generation penalties, strand length increases, and the possibility that double encoding may not reduce the number of false negatives.

January 15, 2007
11:30am - 12:30pm

Goodwin 736