Thomas F. Hain
University of South Alabama
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Thomas F. Hain.
IEEE ACM Transactions on Networking | 1995
Christopher Ward; Cheong H. Choi; Thomas F. Hain
A data link control protocol for low earth orbit satellite networks has been developed to overcome throughput limitations inherent in event-based positive acknowledgment automatic-repeat-request class protocols. This was accomplished by relaxing a reliability constraint, specifically, the in-sequence delivery constraint. This relaxation results in a new class of link layer service, that of reliable datagram, and permits a fresh approach to link layer protocol design. The authors present one such protocol, the low altitude multiple satellite data link control (LAMS-DLC) protocol. They derive the throughput efficiency of LAMS-DLC and compare the results with HDLC (selective reject). The analysis is verified using an event-based simulation. Measurements suggest that LAMS-DLC provides near optimal throughput efficiencies in the target environment, while using significantly less buffer space than that required for HDLC-SREJ. >
International Journal on Artificial Intelligence Tools | 2011
Yan Zhou; Oleksandr Grygorash; Thomas F. Hain
We propose two Euclidean minimum spanning tree based clustering algorithms — one a k-constrained, and the other an unconstrained algorithm. Our k-constrained clustering algorithm produces a k-partition of a set of points for any given k. The algorithm constructs a minimum spanning tree of a set of representative points and removes edges that satisfy a predefined criterion. The process is repeated until k clusters are produced. Our unconstrained clustering algorithm partitions a point set into a group of clusters by maximally reducing the overall standard deviation of the edges in the Euclidean minimum spanning tree constructed from a given point set, without prescribing the number of clusters. We present our experimental results comparing our proposed algorithms with k-means, X-means, CURE, Chameleon, and the Expectation-Maximization (EM) algorithm on both artificial data and benchmark data from the UCI repository. We also apply our algorithms to image color clustering and compare them with the standard m...
conference on computers and accessibility | 2006
Thomas J. Hubbell; David D. Langan; Thomas F. Hain
This paper discusses a research project targeted at the design and implementation of an interface intended to allow manually disabled people to more easily perform the task of programming. It proposes a Speech User Interface (SUI) targeted for this task. Voice was selected as the means of input as an alternative to the keyboard and mouse. Traditional programming IDEs tend to be character and line oriented. It is argued that this orientation is not conducive to voice input, and so a syntaxdirected programming interface is proposed. To test the viability of this combination of voice with a syntax-directed approach, an editor named VASDE (Voice-Activated Syntax-Directed Editor) was implemented using ECLIPSE as the underlying platform for development. This paper describes the syntax-directed interface, VASDE, and some of the lessons learned from initial usability studies.
Computers & Graphics | 2005
Thomas F. Hain; Athar L. Ahmad; Sri Venkat R. Racherla; David D. Langan
We present two related algorithms for flattening (generating polyline approximations for) curves associated with planar cubic Bezier segments. One flattens the path curve, and the other flattens the left and right offset curves. The algorithm for flattening path curves yields an average of 67% of the vertices generated by recursive subdivision, while maintaining flatness to within 4% of the specified value, and runs 37% faster. The algorithm for flattening offset curves generates 70% of the vertices as the methods based on recursive subdivision, such that 94% of all subsegments fall within 20% of the flatness criterion. This latter code runs as fast as recursive subdivision.
technical symposium on computer science education | 2005
Vishal D. Modak; David D. Langan; Thomas F. Hain
Mobile agents are a technology that is applicable in several courses. However, the development of applications using mobile agents can be difficult and time-consuming for students. To address this problem, a tool called Mobile Agent Development Environment (MADE) is presented. MADE facilitates the rapid creation of mobile agent based applications through a pattern-based code generating wizard. MADE organizes and offers the creation of mobile agents based on their characteristics and behavior. MADE is based on the IBM Aglets API. MADE also offers easy-to-use plug-ins for course specific domains such as database and networking. Initial testing has shown that 55% of the code needed for simple applications can be generated quickly and automatically using the described tool.
brazilian symposium on computer graphics and image processing | 2004
Thomas F. Hain; Sri Venkat R. Racherla; David D. Langan
A fast algorithm for generating a polyline approximation (flattening) for the offset curves of a cubic Bezier curve segment is described. It is shown to be more efficient than the standard recursive subdivision method by generating only 70% as many segments, but, just as importantly, 94% of all segments fall within 20% of the flatness criterion. The code runs as fast as recursive subdivision.
acm southeast regional conference | 1998
Thomas F. Hain; Somnath Gulve
Problems in computational geometry are characterized by a large and varied input donain, with many test cases to be covered in testing. The interactive, visual test method described here has been found to be extremely efective in generating su&ient test case coverage to produce a high degree of confidence in the robustness of a code solving such a problem. It is believed that the described technique has broad applicability in computational geometry algorithm testing, as well as other graphical areas.
Computer Science Education | 1992
Thomas F. Hain; Stephen Matthew Jodis
The two‐quarter Senior Design Project course given in the Computer Science and Engineering Department of Auburn University gives undergraduate students a significant computer system development experience by working on a project from proposal to prototype. Heavy emphasis is placed on team‐ and project‐management principles, team presentations, the application of standard structured development methodologies, and the generation of formal deliverables after each project phase. Project ideas are canvassed from faculty, graduate students, local industry and community, and from the students themselves. We believe that the knowledge and experience acquired by students from working in a team on a substantial project gives them an insight and perspective that is unattainable through the standard lecture paradigm. The course has received enthusiastic endorsement from industry, accrediting boards, and participating students (albeit at the end of the course!).
Disability and Rehabilitation: Assistive Technology | 2008
David D. Langan; Thomas F. Hain; T. J. Hubbell; J. Frøseth
Purpose. The paper presents the design and implementation of voice-based interface to support two types of application development activities (coding, and visual GUI design) by manually disabled programmers. Method. Prototype programs were implemented, and tests were performed for both proposed voice-based interface models. Results. The interface design concepts, and the results and lessons learned of the usability studies are discussed. Conclusions. The usability studies indicated that the interface models were sound, although the voice recognition aspects left something to be desired.
Computer and Information Science | 2008
David D. Langan; Thomas F. Hain; W. Christopher Camery
Code entry and editing by manually disabled computer programmers is difficult in conventional GUI-based development environments, since these rely heavily on mouse and keyboard use. The advent of accurate and responsive speech recognition technologies has made the speech user interface (SUI) a viable option for input. While high level programming constructs can be entered in a fairly natural way us / ing voice, expressions may be very complex in structure, and may use a wide variety of variables, literals, punctuation marks and operator symbols. A SUI-based syntax-directed editor was previously presented, but its interface model did not include expression editing in its model and implementation. This paper fills this gap by presenting a SUI-based model for entering and editing programming language expressions. A Java implementation was evaluated by a group of programmers to verify the functionality and to test the ease of use of the model. The results of this evaluation are presented and discussed.