FY96 Technology Transfer Awards
The goal of this project is to develop a suitable scheme for tracking multiple sinusoids in noisy signals. The system consists of severalfilters, decimators and adaptive algorithms. The proposed method isexpected to accurately measure the frequency of each sinusoid, andcontinuously track and enhance the sinusoids. One of the adaptivealgorithms used is called the Fast Conjugate Gradient (FCG) algorithm,developed by the PI as part of an earlier CASI project. This algorithm isused along with other adaptive algorithms and filters for developing theoverall system of this project. The developed scheme will also be studiedfor implementation in a fixed point processor. The algorithms and filtersof this scheme will be theoretically analyzed for quantization noise.
RACEWIN: Resource Allocation and Admission Control Evaluator for Wireless Information Networks. Assoc. Prof. C. Edward Chow, Department of Computer Science, University of Colorado - Colorado Springs, $35,260. Collaborating Company:
U S WEST Advanced Technologies.
Mail: C. Edward Chow
The proposed research deals with the development of efficient algorithms and simulation tools for PCS Wireless Information Network Planning and Management. A software design system called Resource Allocation and Admission Control Evaluator for Wireless Information Networks (RACEWIN) will be built to facilitate the design and analysis of efficient algorithms for cell size determination, hand-off procedure, and traffic management in the future wireless information networks, such as the emerging PCS networks. The RACEWIN system will assist the network administrators in their network planning and management tasks. It will facilitate the network designers to improve the network efficiency and reliability.
PDA-Based Graphical Interchange for Field Service and Repair. Asst. Prof. Wayne V. Citrin, Department of Electrical & Computer Engineering, and Asst. Prof. Mark D. Gross, Department of Environmental Design,
University of Colorado at Boulder, $46,842. Collaborating Company:
U S WEST Advanced Technologies.
Mail: Wayne V. Citrin
Mail: Mark D. Gross
In order to address the problems of field service workers who service and maintain complex infrastructure systems, the principal investigators propose to develop a PDA-based graphical interchange system. The system would allow field workers to download graphical information (plans, diagrams, photos) from a host server, over a wired or wireless connection, to a personal digital assistant (PDA), a small handheld computer supporting pen-based text and diagram entry. The field worker will use the downloaded graphics in his or her task, make annotations on the graphics to reflect the tasks performed, and upload the annotated graphics to the host computer, where the annotations would be interpreted, incorporated into the diagrams, and integrated into the database for later use by other workers.
Efficient Access Methods for Multidimensional Data. Asst. Prof. Scott Leutenegger and Asst. Prof. Mario A. Lopez, Department of Mathematics and Computer Science,
University of Denver, $36,538. Collaborating Company:
MetaComp Inc.
Mail: Scott Leutenegger
Mail: Mario A. Lopez
This project will concentrate on developing techniques for efficient loading, retrieval, and update of multi-dimensional data. The PIs expect to adapt techniques from computational geometry, a field that has matured independently of databases, as well as to develop new ones. The PIs intend to transfer their research into the extensible hybrid database system being developed by MetaComp. A database product that incorporates the PIs' proposed techniques for efficient support of multi-dimensional data will be competitive in the fast growing markets of scientific and geographic databases.
An Efficient Geometric Engine for CAD Tools. Asst. Prof. Mario A. Lopez, Department of Mathematics and Computer Science,
University of Denver,
$26,602. Collaborating Company:
Advanced Software Technologies, Inc.
Mail: Mario A. Lopez
The researchers will develop a geometric engine applicable to a wide range of design methodologies. The engine will include a data model and efficient algorithms for a number of geometric problems, including: object intersection searching, nearest object searching, and computation of object contours. An efficient geometric engine is an important component of the Meta-CASE tool being developed by Advanced Software Technologies, Inc. The PIs expect it to be useful in other CAD applications as well.
Emerging Technologies in Construction: Integrated Information Processes for the 21st Century. Asst. Prof.
Anthony Songer
; Civil, Environmental, and Architectural Engineering;
University of Colorado - Boulder; $30,768. Collaborating Company: Flatiron Structures Company.
Mail: Anthony Songer
This is the second year of a two-year project. During the coming year, extensive testing will be conducted to address the issue of visualization. Using control/experimental comparison of different users for the same project, data will be collected on the planning processes, planners and planning product. A project by the industry partner will be used to conduct the testing.
From Program Comprehension to Software Maintenance Support Tools. Prof.
Anneliese von Mayrhauser, Department of Computer Science, Colorado State University, $38,500. Collaborating Company:
IBM Corporation.
Mail: Anneliese von Mayrhauser
The researchers propose to extend their own work on understanding industrial software and translate findings into requirements for software maintenance support tools. Specifically, they will evaluate which of the features of IBM's Lemma environment support code comprehension, which do not, and which additional features Lemma should have to provide better support.
The First Break Seismic Detection Problem: Representation Issues for Neural Net, Fuzzy Logic & Heuristic Solutions. Assoc. Prof.
Darrell Whitley,
Department of Computer Science,
Colorado State University, $44,110. Collaborating Company: Landmark Advanced Products Group.
Mail: Darrell Whitley
"First break picking" is the problem of identifying the first peak or reflected signal in a seismic trace. First break picking is done as a preprocessing step that sets the stage for subsequent signal processing of seismic signals. Most traditional statistical and analytical approaches to this problem are inadequate and heuristic solutions with extensive human interaction are common. Neural networks have been applied to this problem, but only with modest success. The researchers will attempt to find improved solutions to this problem in two ways. First, they will work to improve the problem representation in an effort to achieve better neural network results. Second, they will recast an existing heuristic solution in the form of a fuzzy logic system.
Predicting Program Behavior to Support Instruction Level Parallelism. Asst. Profs. Dirk Grunwald and Benjamin Zorn and Assoc. Prof. James Martin, Department of Computer Science, and Prof. William Waite, Department of Electrical and Computer Engineering,
University of Colorado - Boulder, $60,000. Collaborating Company:
Hewlett-Packard Company.
Mail: Dirk Grunwald
Mail: Benjamin Zorn
Mail: James Martin
Mail: William Waite
The principal investigators will investigate the trade-offs between information gathered and the optimizations that are enabled by information collected. In particular, they will investigate the range of possible gathering times from compile-time (via source), to link time (via executable), to load time (incorporating shared library information) to execution time (using profiling or hardware). The research will investigate specific optimizations, such as branch prediction and load latency prediction, and look at the impact of having different information available on overall program performance.
Progressive Refinement Approach to Factory Planning. Assoc. Prof. William J. Wolfe, Department of Computer Science and Engineering,
University of Colorado - Denver, $13,000. Collaborating Company:
CTA Incorporated.
Mail: William Wolfe
Planning and scheduling problems are notoriously difficult because of vagueness, uncertainty and intractability. A schedule usually depends on external schedules. Thus, a schedule is part of a larger system of interacting schedules, any one which can initiate a sequence of changes that ripples through the enterprise. Consequently, the current research in algorithms needs a more realistic context. We focused on an approach called "Progressive Refinement", which takes iterative repair, constraint-envelopes, and distributed decision making and integrates them into a rolling horizon of evolving schedules. We begin with an out-day schedule and progressively refine the constraints, allocations and commitments as the day of operation approaches. The critical issue: how to integrate new information (outages, canceled orders, etc.), while maintaining plan continuity and stability?
FY96 Technology Transfer Awards
FY95 Technology Transfer Awards
To view CASI Members