| Title | Corresponding Author(s) | Faculty Sponsor(s) |
Department | Abstract | ||||||||||
| Determination
of Fluid Flow Properties of Permeable Media Using Magnetic Resonance
Imaging |
Jinsoo Uh | A. Ted Watson | Chemical Engineering | The knowledge of fluid flow properties is critical to understand and control diverse processes within permeable media. In this poster, we demonstrate our method to determine permeability distributions using the velocity data measured by magnetic resonance imaging. We use a global optimization procedure that incorporates parallel tempering to interpret the measurements. The parallel tempering algorithm is implemented in the COGRID parallel computing system using message passing interface (MPI). | jinsoouh@engr.colostate.edu | |||||||||
| Computational
Chemistry and Structural Bioinformatics Using an Apple Xserve Cluster |
Terry M. Gray | Anthony K. Rappe | Chemistry | An Apple Xserve cluster is being used for computational chemistry and structural bioinformatics research. There are six dual G5 2.0 GHz nodes (2 GB RAM; Mac OS X Server 10.3; 100 Mb ethernet interconnects). Sun Grid Engine 6 is being used for distributed queuing. This cluster has been used to study homogeneous and heterogeneous methane, propane, and propene oxidation catalysis, propylene polymerization, inorganic and organic (cellulose) nanomaterials, and small peptide conformational analysis. | grayt@lamar.colostate.edu | |||||||||
| Modeling Nanoscale Chemistry | Anthony K. Rappe | Chemistry | Nanoscale materials are of current interest due to their potential for providing novel mechanical or electrical properties. In order to study nanoscale chemistry we have developed an Approximate Pair Theory (APT) that uses parametric models of electronic interactions to permit reasonable descriptions of the electronic changes associated with nanoscale chemistry. This methodology as well as applications of it to nanoscale processes are described in this poster. | anthony.rappe@colostate.edu | ||||||||||
| The Virtual Institute for Satellite Integration Training at the Cooperative Institute for Research in the Atmosphere (CIRA) | Dan Bikos, Jeff Braun, John
Weaver, Bernie Connell |
CIRA | The primary mission of the Virtual Institute for Satellite Integration Training (VISIT) is to accelerate the transfer of research results based on atmospheric remote sensing data into National Oceanic and Atmospheric Administration / National Weather Service (NWS) operations. This transfer is accomplished through the education of NWS forecasters on the latest techniques to integrate remote sensing data, especially from satellite. The distance learning approaching is based primarily on using a combination of the internet and a conference phone line. The program utilizes a software application known as VISITview which provides two-way interactive capability. Over 14,000 students have participated in VISIT teletraining since April 1999. | bikos@cira.colostate.edu | ||||||||||
| A Unified Transaction Processing Model for Advanced Transactions | Tai Xin, Yajie Zhu | Indrakshi Ray | Computer Science | Advanced applications, consisting of highly cooperative activities, have rapid growth recently. Mechanisms for reliably executing and deploying these applications are greatly needed. The goal of this research is to produce a reliable transaction processing model for advanced transactions, which can be used to develop complex yet well-formed advanced transactions, correctly ensure the constraints of dependencies in execution/failure recovery, and have the power to survive malicious attacks. | xin@cs.colostate.edu | |||||||||
| An aspect
oriented approach to early software design |
Y. Raghu Reddy | Robert France | Computer Science | Developers of complex software systems are often required to address security, faulttolerance and other crosscutting concerns. The solutions that realized these crosscutting concerns are often spread across and tangled with other concerns in an architectural design. As a result it can be diffcult to modify and evolve these concerns. Localizing parts of an architectural design that describe crosscutting concerns can help developers better manage the evolution of dependability features. In this research, we develop an aspect-oriented modeling (AOM) approach that supports separation of crosscutting concerns from other concerns during architectural design. In the proposed AOM approach, the features are addressed using aspect-oriented architecture models (AAM). An AAM consists of a set of aspect models and a primary model. An aspect model describes the crosscutting concerns and the primary model describes the base architecture of the system. The aspect models and the primary model are composed to produce the logical architecture of the system. Bindings and composition directives determine how the composition is done. Composition of aspect model and primary model behaviors may produce descriptions of undesirable emergent behavior, as a result of the interactions between the aspect and the primary model behaviors. The composed model needs to be analyzed to identify such behavior. The proposed research provides techniques for (1) representing the aspect models and the primary models using UML, (2) composing the aspect models and the primary model in a verifiable manner, and (3) analyzing the composed models. The techniques will be developed using simple applications that involve modeling of security and fault tolerance dependability concerns. Once the techniques are developed they will be applied on moderately large case studies. | raghu@cs.colostate.edu | |||||||||
| Code
Generation in Polyhedral Model |
DaeGon Kim | Sanjay Rajopadhye | Computer Science | Automatic parallelization of a large class of loop programs is based on the polyhedral model, where programs are given a mathematical description based on polyhedra, affine mappings and dependences. Code generation, the last step in parallelization, re-produces a program from a transformed description. We present a correctness proof of Quillere et al. algorithm. We also report experiments on various code generation algorithms and the integration of code generation in the alpha framework. | kim@cs.colostate.edu | |||||||||
| Data Quality
Control, Modeling and Performance Analysis of TRABOL: A Streaming Protocol
for High-Bandwidth Radar Data |
Aditya Maroo, Tarun Banka | V. Chandrasekar, Anura Jayasumana | Computer Science | TCP-friendly Rate Adaptation Based on Losses (TRABOL) protocol is targeted for high-bandwidth real-time digitized radar data transmission. A sample selection mechanism is designed and implemented that complements TRABOL to meet the end-user data quality requirements. The performance evaluation results prove that TRABOL is TCP-friendly and its degree of friendliness can be adjusted using its operating parameters. The empirical verification of the analytical model for TRABOL proves its validity. | maroo@cs.colostate.edu | |||||||||
| Decision
Tree Ensembles for Mental Task Classification |
Tim O'Connor | Charles Anderson | Computer Science | Mental task classification using electroencephalogram (EEG) signals to create a brain computer interface is a difficult problem because patterns across electrodes and time must be deciphered using weak, noisy signals. Understanding the classification is even more difficult because patterns across electrodes and time must be deciphered. Singular value decomposition (SVD), data lagging, clustering, and ensembles of under sampled decision tree classifiers are used to create an effective classifier. Very small sets of consistent logical rules can then be extracted from this ensemble. These small rule sets are accurate and provide a basis for understanding mental task classification. | oconnt@cs.colostate.edu | |||||||||
| Dynamic Power
Minimization for Combinational Circuit Testing as a Travelling Salesman
Problem |
Alodeep Sanyal, Artem Sokolov | L. Darrell Whitley, Yashwant Malaiya | Computer Science | We perform a quantitative comparison of two TSP techniques and explore implications of combining them on dynamic power dissipation problem during combinational circuit testing. We found our results to be significantly better than those reported in the literature. | alodeep@cs.colostate.edu | |||||||||
| Finding
connectivity in microchip artwork design |
Clem Izurieta | Robert France | Computer Science | Scan based or Line Sweep methods are a traditional mechanism to traverse the physical layout, or artwork of a microchip. These traversals are incremental in nature. They typically traverse the artwork from bottom to top (in the y direction) and from left to right (in the x direction) in a traditional Cartesian plane. As the traversal occurs, client applications need to know what causes a given event. For example, an analysis algorithm may be interested in finding out every time two rectangle shapes overlap or touch. This is important because such an overlap could represent a short in a circuit. The number of rectangles used to represent real world designs of artwork in a microchip is in the order of one million per metal layer, thus various optimization techniques are necessary to breakdown the complexity of such designs. This poster describes a semi-greedy technique coupled with an object oriented callback design pattern to aid in traversing and reporting events of chip data. | cizuriet@colostate.edu | |||||||||
| Finding
Shortcuts: Infeasible Solutions in
Scheduling Problems |
Mark Rogers | Adele Howe, L. Darrell Whitley | Computer Science | Search algorithms that include infeasible solutions have proved to be efficient in solving scheduling problems. Researchers conjecture that these algorithms yield two key benefits: 1) they focus on the boundary between feasible and infeasible solutions, where active constraints may yield optimal values, and 2) moves that include infeasible solutions may uncover short-cuts to optima. We report results on a project to develop methods that identify and measure short-cuts when they occur. | rogersma@cs.colostate.edu | |||||||||
| ICMP
Traceback Packet Authentication for BGP Path Validation |
Eunjong Kim | Indrajit Ray, Dan Massey | Computer Science | Nearly all network applications rely on the global Internet routing infrastructure to compute routes and deliver packets. Unfortunately, Border Gateway Protocol (BGP), the Internet's global routing protocol, lacks basic authentication and monitoring functionalities. False Internet routes can be easily introduced leading to denial of service or misdirected traffic. To provide efficient validation mechanisms for each path information, we use ICMP (Internet Control Message Protocol) traceback message with source AS-PATH information. It uses efficient off-line verification technique with which each router independently and dynamically keeps track of local database. Our solution allows a destination to monitor its routes, detect false paths used by remote sites, and record routing data for later forensic analysis in the event of an attack. In addition, it does not require monifications to the BGP protocol and hence can be easily deployed. | kimeu@cs.colostate.edu | |||||||||
| Middleware Transparent Software Development | Devon M. Simmonds | Robert France, Sudipto
Ghosh |
Computer Science | Changing the middleware of a distributed application is risky and difficult when middleware functionality is spread across and tangled with business functionality. Middleware Transparent Software Development (MTSD) is an aspect-oriented approach to distributed application development. The approach supports the separation of middleware from business functionality. In MTSD, business functionality is developed without considering middleware concerns. Middleware-specific functionality is provided by independently developed aspects. Composing middleware aspects with business functionality produces a middleware-specific distributed application. The separation of business functionality from middleware concerns eases the task of migrating business functionality to new middleware platforms. | simmonds@cs.colostate.edu | |||||||||
| Modeling Local Search: A First Step Toward Understanding Hill-climbing Search in Oversubscribed Scheduling | Mark Roberts | Adele Howe, L. Darrell Whitley | Computer Science | Previous results for a real-world domain, scheduling communications requests for the Air Force Satellite Control Network (AFSCN), suggested that the best search methods perform a greedy random walk. We examine the degree to which this observation is accurate for AFSCN, and take steps toward generalizing this model for scheduling image requests for a set of Earth Observing Satellites (EOS). Our results indicate that this model is reasonably accurate for AFSCN, though it is less accurate for EOS. | mroberts@cs.colostate.edu | |||||||||
| Optimized
Compilation of Visual Programs for Image Processing on Adaptive Computing
Systems |
Monica Chawathe | Wim Bohm | Computer Science | We show a high level approach for embedded, image processing applications on FPGAs. Examples of applications are Probing, Synthetic-Aperture-Radar prescreening and wavelet based image comprension. | monica@cs.colostate.edu | |||||||||
| Systolic
Architectures for Back Propagation Neural Networks |
Alodeep Sanyal | Sanjay Rajopadhye | Computer Science | Back propagation, a well known technique in the artificial neural networks, is computation intensive and lends itself to a high degree of parallelism. These features motivate a systolic design of hardware to implement the back propagation algorithm. We present implementation of a 1-dimensional linear systolic array for this algorithm and claim that it reports better running time and therefore an improvement over the existing hardware solution. We synthesize the architecture by an automatic systolic architecture synthesis tool called MMAlpha. The synthesized IP core for the back-propagation algorithm is interfaced and mapped to a regular cell-based architecture (viz. a Field Programmable Gate Array or FPGA). Presently we are analyzing the performance of this architecture compared to a serial machine for various neural net applications. | alodeep@cs.colostate.edu | |||||||||
| Using FIFOs in Hardware
Software Codesign or FPGA based Embedded Systems |
Charles Ross | Wim Bohm | Computer Science | We compare FSLs (A simple FIFO design) to Generic busses and we show that in hardware accelerator based Embedded FPGA applications, FSLs are preferable. | rossc@cs.colostate.edu | |||||||||
| A Monte Carlo Simulation Package for Simulating Multisite Adsorption | James Knight | Simon Tavener, Dr. Bruce
Parkinson, and Todd Mytkowicz |
Computer Science, Mathematics, Chemistry | A software package for simulation of a large class of adsorption/desorption systems is presented. We give simulation and experimental results of multisite adsortion and desorption on various lattice structures. | nate@cs.colostate.edu | |||||||||
| Solving the Research to Operations "Valley of Death" while Optimizing Science Algorithm Development | Philip Partain, Ken Eis, Don
Reinke, and Dale Reinke |
Cooperative Institute for Research in the Atmosphere | Porting research code to operational data processing in a cost-effective manner while simplifying the job of the algorithm developer is a challenge for today's satellite missions. To address this issue, CIRA has developed the ADAPT system to perform data processing activities for the NASA CloudSat mission. Comprised of several innovative and generic software applications, ADAPT can easily be tailored to other missions to provide a low-cost and robust data processing solution. | partain@cira.colostate.edu | ||||||||||
| Analog Precoder and
Equalizer Design for Multichannel Communication |
Zhifei Fan | Louis Scharf | Electrical and Computer Engineering | This paper addresses joint analog precoder and equalizer design for multichannel data transmission over the frequency-selective additive Gaussian noise (AGN) channel. The design goal is to maximize mutual information rate subject to a transmit power constraint. We assume a continuous channel model with precoder transmissions for M subchannels that lie in an $n$ dimensional linear subspace of L^2(R). | fanz@engr.colostate.edu | |||||||||
| Attenuation
Statistics for X Band Radar Design |
Wanyu Li, Delbert Willie | V. Chandrasekar | Electrical and Computer Engineering | Signal attenuation due to rain is very important in design of weather radars operated at higher frequencies. | wanyu_li@engr.colostate.edu | |||||||||
| Canonical
Coordinate Geometry of Precoder and Equalizer Designs for Multichannel
Communications |
Zhifei Fan | Louis Scharf | Electrical and Computer Engineering | In this paper we show that all known designs for precoder/equalizer design are, in fact, decompositions of a virtual two-channel problem into a system of canonical coordinates, wherein variables in the canonical message channel are correlated only pairwise with corresponding variables in the canonical measurement channel. This finding clarifies the geometry of precoder/equalizer design and illustrates that these designs decompose the two-channel communication problem into the Shannon channel. | fanz@engr.colostate.edu | |||||||||
| Coherent-Based Feature Extraction for Cloud Analysis using Satellite Imagery Data | Amanda Falcone, Adam Kankiewicz,
Donald Reinke |
Mahmood Azimi-Sadjadi | Electrical and Computer Engineering | Atmospheric products derived from the MODerate resolution Imaging Spectroradiometer (MODIS) instrument are widely accepted as the state-of-the-art by the meteorological community. These products are more useful to the meteorological community than the traditional cloud labeling results. However, these products are not available at regular temporal frequency over one specific region. Algorithms that create MODIS products cannot simply be applied to other satellite data. Thus, an innovative method is needed to estimate MODIS-like products using geostationary satellite imagery, which has a higher temporal frequency. This paper presents a canonical coordinate decomposition (CCD)-based method to estimate MODIS channels using imagery from the geostationary satellite Meteosat 8. The estimated data can subsequently be used to arrive at cloud phase products. | afalcone@engr.colostate.edu | |||||||||
| Compact soft x-ray lasers extend the limits of dense plasma interferometry | J. Filevich, E.C.Hammarsten,
J.Grava, M. Purvis, R.F. Smith, J. Dunn, R. Keenan, J.R. Hunter, S.J. Moon,
J. Nilsen, V.N. Shlyaptsev |
J.J. Rocca, M.C. Marconi | Electrical and Computer Engineering | We have designed a novel interferometer to work together with compact soft x-ray lasers at wavelengths of 46.9 and 14.7 nm. The design uses diffraction gratings to form the two arms in a skewed Mach-Zehnder configuration. This grating interferometer shows high throughput, excellent stability and a resistance to damage caused by plasma debris. The design also permits customization of the gratings to function at any selected EUV wavelength, making the diffraction grating interferometer very versat | rage@engr.colostate.edu | |||||||||
| Data
Transport Challenges in Emerging High Bandwidth Real-TIme Collaborative
Adaptive Sensing Systems |
Tarun Banka, Brian Donovan, | V. Chandrasekar, Anura Jayasumana | Electrical and Computer Engineering | A new paradigm of networked radars is on the verge of becoming a reality. The many advantages offered by these networks of small and low cost radars are so overwhelming that they offer the opportunity to potentially revolutionize the science of radar meteorology. Potential benefits include significant cost savings, novel paradigms for radar design and deployment, and the ability to measure and predict weather phenomena in ways that were hitherto not possible. NSF CASA (Collaborative Adaptive Sensing of the Atmosphere) Engineering Research Center is a multidisciplinary, multi-university effort that is redefining the landscape of meteorological radar design and operation through the paradigm of networked remote sensing. CASA is a prime example of emerging distributed collaborative and adaptive systems. Many existing transport protocols are able to meet requirements of end applications like voice or video streaming by adapting to dynamic network conditions due to link degradation or congestion. Need to support high bandwidth associated with radar sensors, heterogeneous QoS requirements of different end applications, and underlying heterogeneous (wired and wireless) networking infrastructure are factors pushing the existing transport protocols to their limits. Application specific minimum acceptable and maximum desired rates, and distinct data quality requirements for different clients need to be satisfied under dynamic network conditions. In this poster we concentrate on such unique challenges imposed on the transport protocols for collaborative and adaptive systems. | tarunb@engr.colostate.edu | |||||||||
| Design of CMOS
Trans-Impedance Amplifier for On-chipWaveguide Based Optical Clock
Distribution System |
Charles Thangaraj | Tom Chen | Electrical and Computer Engineering | This paper presents a design of a CMOS trans-impedance amplifier | charles@engr.colostate.edu | |||||||||
| Efficient Nonlinear Frequency Conversion With A Dynamically Structured Nonlinearity | Klaus Hartinger | Randy Bartels | Electrical and Computer Engineering | klaus@engr.colostate.edu | ||||||||||
| Environmentally
adaptive acoustic transmission loss prediction with neural networks |
Gordon Wichern, Michael Mungiole | Mahmood Azimi-Sadjadi | Electrical and Computer Engineering | Prediction of acoustic transmission loss (TL), or the attenuation of sound pressure level (SPL) is a complex problem dependent on a variety of physical parameters. Prediction of the TL using a numeric parabolic equation (PE) method is often accepted as a method of providing accurate TL prediction, but the large computational time is a hinderance in applications requiring real-time situation awareness. In order to overcome these extreme computational requirements a neural network-based environmentally adaptive TL prediction method is proposed and developed in this paper. This method uses multiple back-propagation neural network (BPNN) predictors, each trained on specific environmental conditions, and then probabilistically combines the outputs of these predictors in a fusion center to obtain a final TL estimate. This method is implemented on a data set generated using a PE model for a wide range of geometric and environmental parameters. The results are then benchmarked against a single neural network-based prediction scheme. | gwichern@engr.colostate.edu | |||||||||
| Failure-Tolerant
Path Planning for Kinematically Redundant Manipulators |
Rodrigo S. Jamisola | Anthony Maciejewski, Rodney Roberts | Electrical and Computer Engineering | This work considers kinematic failure tolerance when obstacles are present in the environment. It addresses the issue of finding a collision-free path such that a redundant robot can successfully move from a start to a goal position and/or orientation in the workspace despite any single locked-joint failure at any time. An algorithm is presented that searches for a simply-connected obstacle-free surface with no internal local minimum or maximum in the configuration space that guarantees the existence of a solution. The method discussed is based on the following assumptions: a robot is redundant relative to its task, only a single locked-joint failure occurs at any given time, the robot is capable of detecting a joint failure and immediately locks the failed joint, and the environment is static and known. Examples are given using a three degree-of-freedom planar robot and a seven degree-of-freedom commercially available redundant robot. | jamisola@engr.colostate.edu | |||||||||
| Fluidic Intracavity Laser Diode (FILD) Sensor | D. Kumar, L. Shao | Kevin Lear | Electrical and Computer Engineering | A microfluidic channel is integrated to the laser cavity of an electrically injected vertical cavity surface emitting laser (VCSEL) diode to construct a fluidic intracavity laser diode (FILD) sensor. Device fabrication and output power versus injection current measurements are presented. Larger fluid index in the cavity is observed to decrease the laser’s threshold current and increase slope efficiency. Possible mechanisms are examined. | dhiraj@engr.colostate.edu | |||||||||
| K-band
Radiometer based on Monolithic Microwave and Millimeter wave Integrated
Circuit (MMIC) technology for Water Vapor Retrieval |
Flavio Iturbide-Sanchez, Luis Batres | Steven C. Reising | Electrical and Computer Engineering | Water vapor constitutes the most dynamic component of the earth’s troposphere and it influences almost all atmospheric processes from radiative heating on a climatic scale to local hydrological budgets and severe weather events. Radiosonde balloon packages and active remote-sensing instruments operating at optical wavelengths (Lidars) are both possible methods for profiling water vapor. The performance of these large and expensive systems, however, degrades during daylight hours, and their use is limited to clear-sky conditions. The advantage of microwave radiometers is their ability to inexpensively and continuously monitor atmospheric water vapor within a controlled spatial volume under nearly all weather conditions. At present, the vast majority of ground-based and airborne microwave remote sensing instrumentation is produced using waveguide-based or connectorized discrete microwave components, which are high in cost and large in volume. Recent maturation of Monolithic Microwave and Millimeter-wave Integrated Circuit (MMIC) technology, mainly developed for the wireless communications and defense industries, has provided potential reductions in the volume, weight and cost of microwave and millimeter-wave radiometers for Earth remote sensing. Furthermore, small and low-cost radiometers can be designed and produced in relatively large quantities, making feasible the implementation of two and three-dimensional (2-D and 3-D) arrays of sensors to retrieve the 3-D distribution of tropospheric water vapor. An example of the reductions afforded by MMIC technology is the prototype Miniaturized Water Vapor profiling Radiometer (MWVR), developed at the University of Massachusetts and recently moved to Colorado State University. The MWVR has been designed to obtain all available information on the humidity profile by using four frequencies in the upper sideband of the 22.2 GHz water vapor resonance and one Ka-band frequency to retrieve integrated liquid water. The K-band and Ka-band radiometers for the MWVP will be field tested, calibrated and validated using intercomparisons with reliable humidity profile data. Finally, a coordinated network of four Miniaturized Water Vapor Profilers will be designed and produced to measure the 3-D distribution of water vapor in the atmosphere using tomographic inversion. | fiturbid@ecs.umass.edu | |||||||||
| Novel Laser Diodes
for Telecom |
Ovidio Anton | Carmen Menoni | Electrical & Computer Engineering | Semiconductor lasers are ubiquitous in many of today’s electronic systems including communication, computing, data storage, and sensing systems. In this work we describe recent experiments of a novel class of laser diodes for telecom that promise higher modulation bandwidths and temperature stability. To investigate the laser behavior we have developed innovative electro-optical tools whose results are analyzed using a comprehensive model of the laser diode, whose details will be described. | anton@engr.colostate.edu | |||||||||
| Markov Decision Process with Action Fairness Constraints | Sudhir Moola, Zhi Zhang | Edwin K. P. Chong | Electrical and Computer Engineering | Many problems in wireless networks, sensor networks, financial planning, etc., can be modelled as Markov decision Problems. Present solution techniques for Markov decision problems do not provide for action fairness. In this work we present a solution for Markov decision process with action fairness constraints. We solve for Markov decision process with action fairness for both average reward criterion and expected discounted cost criterion. | msudhir@engr.colostate.edu | |||||||||
| Markov
Decision Processes and Particle Filters for Sensor Management |
Lucas W. Krakow, Yun Li | Edwin Chong | Electrical and Computer Engineering | We study the problem of sensor scheduling for target tracking, to determine which sensors to activate over time to trade off tracking performance with sensor usage costs. We approach this problem by formulating it as a partially observable Markov decision process (POMDP), and develop a Monte Carlo solution method using a combination of particle filtering for belief state estimation and sampling-based Q-value approximation for lookahead. | liyun@engr.colostate.edu | |||||||||
| Modeling
Multi-Fiber Shared-Wavelength-Conversion Switch Architectures and Conversion
Resources Allocation Algorithms for Share-per-link WCS |
Abdulgader Habiballa | Anura Jayasumana | Electrical and Computer Engineering | Dynamic Conversion Resources Allocation Algorithm, CRAA, will be a key technique for tackling the Routing and Wavelength Assignment problem. The adaptation of a certain dynamic CRAA influences the complexity of the optical switch control plane and accordingly the performance of the share-per-link switch. We propose three different algorithms for dynamically allocating conversion resources. Analytical models are developed for these algorithms to verify their efficiency. | habiball@engr.colostate.edu | |||||||||
| Opportunistic
Transmission Scheduling in Wireless Networks |
Zhi Zhang | Edwin Chong | Electrical & Computer Engineering | In wireless networks, the link capacity is usually a scarce resource that needs to be shared efficiently and fairly among users. A transmission scheduling scheme is called opportunistic if it exploits the time-varying nature of the wireless channel. Our goal is to find an optimal scheduling policy for selecting users to transmit at each time opportunistically, while maintaining the specific fairness constraints. | zhang@engr.colostate.edu | |||||||||
| Optimization of
Individual Well Adaptive Body Biasing (IWABB) Using a Multiple Objective
Evolutionary Algorithm |
Justin Gregg | Electrical and Computer Engineering | Mitigating the effects process variations on power-frequency distributions can be done by using a system of locally-generated body biases. This system allows for highly localized circuit optimizations with very little overhead in silicon area and routing resources. We present a multiple objective algorithm to find near-optimal configurations of these biases which can be applied during post-fabrication testing. The system can improve an initial yield of 12% to 77%. | jlgregg@engr.colostate.edu | ||||||||||
| Optimization techniques in portfolio management | Ramin Zahedi | Edwin Chong | Electrical and Computer Engineering | We’re trying to find a portfolio selection that has the best performance in the stock market. This problem has been around for quite a while and has been challenged by many people. Cover(1991) proposed a new algorithm called “Universal Portfolios” which is competitive with best constant rebalanced portfolio chosen after the stock outcomes are revealed. Our goal is to find a strategy, using optimization techniques such as Markov Decision Processes, to fine the best portfolio selection. | zahedi@engr.colostate.edu | |||||||||
| Overcoming the
effects of correlation in delay measurements using inter-packet gaps |
Nischal M. Piratla | Anura Jayasumana | Electrical and Computer Engineering | The delay of packets are characterized, emphasizing on effects of cross traffic, sending rate, and packet size. Measurements show that modeling delay with high sending rates, as a fraction of bandwidth, is difficult due to correlations among the values. Correlations among inter-packet gaps (IPG) at these rates, are negligible. At lower rates, delay correlations are negligible, and distribution can be used as a delay model. Using relation between delay and IPG, delay is approximated as Markov Pro | Nischal.Piratla@colostate.edu | |||||||||
| Performance Tradeoffs of Shared Limited Range Wavelength Conversion Schemes in Optical WDM Networks | Fahad Al-Zahrani | Anura P. Jayasumana | Electrical and Computer Engineering | Performance of optical switches that employ limited-range wavelength converters is investigated. Different wavelength converter ranging architectures, complemented with conversion resources sharing, are considered in this research work. Simulation results demonstrate that the performance improvement obtained by full range wavelength conversion can almost be achieved by using ranged LRWC. | fahad@engr.colostate.edu | |||||||||
| Processor Allocation for Tasks that is Robust Against Errors in Computation Time Estimates | Prasanna V. Sugavanam, Syed Amjad Ali, Mohammad Al-Otaibi, Mahir Aydin, Kumara Guru, Aaron Horiuchi, Yogish G. Krishnamurthy, Panho Lee, Ashish Mehta, Mohana Oltikar, Ron Pichel, Alan J. Pippin, Michael Raskey, Vladimir Shestak, Junxing Zhang | H. J. Siegel and A. A. Maciejewski | Electrical and Computer Engineering | Heterogeneous computing systems composed of interconnected machines with varied computational capabilities often operate in environments where there may be sudden machine failures, higher than expected load, or inaccuracies in estimation of system parameters. Makespan (defined as the completion time for an entire set of tasks) is often the performance feature that is optimized in such systems. It is important that the makespan of a resource allocation (mapping) be robust against errors in task computation time estimates. The problem of optimally mapping tasks onto machines of a heterogeneous computing environment has been shown, in general, to be NP-complete. Therefore, heuristic techniques to find near optimal solutions to this mapping problem are required. The goal of this research is to find a static mapping of tasks so that the robustness of the desired system feature, makespan, is maximized against the errors in task execution time estimates. Seven heuristics to derive near-optimal solutions and an upper bound to this problem are designed and evaluated | prasanna@engr.colostate.edu | |||||||||
| Range-velocity ambiguity mitigation for dual velocity ambiguity mitigation for dual polarized weather radars | Nitin Bharadwaj | V. Chandrasekar | Electrical and Computer Engineering | The center for Collaborative Adaptive Sensing of Atmosphere (CASA) is established by NSF to develop a network radars for targeted applications such as tornado, flash flood monitoring. Radar operating for such applications will suffer from range overlay and velocity folding problems. In this research waveforms to mitigate range overlay and velocity folding are presented. The design is based on many practical constraints and data requirements for radar meteorology. | nitin@engr.colostate.edu | Range Range-velocity ambiguity mitigation for dual velocity ambiguity mitigation for dual polarized weather radars | ||||||||
| Requirement Analysis of Antenna Performance for Tornado Detection | Yanting Wang | V. Chandrasekar, V.N.
Bringi |
Electrical and Computer Engineering | Tornado as an observation features sharp gradients, which means very good sidelobe performance is required for the antenna. A conceptual model is built for tornado; and based on that the impact of antenna sidelobe performance is analyzed. A set of design curves are presented for the antenna sidelobe figures and radar simulation is conducted. The presented design curves can act as governing guidelines in antenna design for tornado detection and observation. | ytwang@engr.colostate.edu | |||||||||
| Scheduling File
Transfers over Highspeed Networks |
Hua Li | Edwin Chong | Electrical & Computer Engineering | In the context of connection-oriented highspeed networks, we have studied a greedy scheduling algorithm, Variable Bandwidth List Scheduling (VBLS), to manage file transfers. We analyzed the file delay performance and the bandwidth allocation fairness among files of different sizes under VBLS. Our simulation results show that the system achieves high utilization, and that VBLS favors large files over small files. | huali@engr.colostate.edu | |||||||||
| Target Tracking with
Limited Communication in Sensor Networks |
Patricia R. Barbosa | Edwin Chong | Electrical and Computer Engineering | We consider the problem of tracking a target that moves according to a Markov chain on a sensor network. Sensing information is transmitted through a feedback channel whose capacity is scarce. We are interested in finding the minimum requirements on channel capacity such that tracking is possible. The natural interplay between querying, coding and information theory is explored, and preliminary results show that optimal tracking strategies are related to Huffman coding. | pbarbosa@engr.colostate.edu | |||||||||
| Thermal impact and management for high-speed VCSELs | A. N. Al-Omari, Z. Liu, J. Flint, D. Kumar | Kevin Lear | Electrical and Computer Engineering | Chip-to-chip optical interconnects would benefit from directly modulated vertical cavity surface emitting lasers (VCSELs) operating at higher data rates. Thermal limits are a major limitation to increased VCSEL bandwidth. Metal wrapped and flip-chip bonded structures have been realized with significantly reduced thermal resistance. Comparisons with thermal simulations are made. | ahmad@engr.colostate.edu | |||||||||
| Truly CMOS Compatible on-chip optical interconnects | Abdul Matheen Raza, Guang Wei
Yuan, Charles Thangaraj, Bob Pownall |
Kevin Lear, Tom Chen | Electrical and Computer Engineering | CMOS based waveguides coupled to high-speed silicon photodetectors have been designed to meet requirements for on-chip clock distribution. Light is routed through typical bends and splitters in the CMOS waveguide and is coupled at the end of the waveguide segment to a traveling wave silicon photodetector that offers both high speed and responsivity. Optical and electronic modeling of the detector performance and experimental results from amorphous silicon devices are presented. | amraza@engr.colostate.edu | |||||||||
| 17 GHz polyimide
planarized VCSELs for chip-to-chip optical interconnects |
A. N. Al-Omari | Kevin Lear | Electrical and Computer Engineering | Vertical cavity surface emitting lasers (VCSELs) are considered to be the bottleneck for chip-to-chip optical interconnects. High speeds, oxide-confined, VCSELs are being developed for this application. The devices exhibit –3dB modulation bandwidths up to 17GHz and are fabricated using a reproducible, simple process. VCSELs circuit model based on physical features of the VCSEL die is obtained. Measured values of the circuit elements were extracted from the S11 parameters. | ahmad@engr.colostate.edu | |||||||||
| Computer-Interactive
Keys for the Larvae and Early Juveniles of Selected Southwestern U.S. Fishes
(Hands-on Experiementation) |
Darrel E. Snyder | Sean C. Seal | Fishery and Wildlife Biology/Larval Fish Laboratory | Computer-interactive keys for the larvae and early juveniles of selected southwestern fishes are provided for hands-on experimentation. One covers the Catostomidae (suckers) of the Upper Colorado River Basin and is intended to be used with the recently expanded and updated guide to those larvae by Snyder and Muth (2004, Colorado Division of Wildlife Technical Publication 42). The others are intended for use with a guide to (selected) Native Cypriniform Fish Larvae of the Gila River Basin by Snyder et al. (2005, Final Report to the U.S. Bureau of Reclamation, Phoenix, Arizona) and include keys to the families of fishes, the native Catostomidae, and selected native Cyprinidae (minnows). Computer-interactive keys are more flexible and user-friendly tools for specimen identification than printed dichotomous or polychotomus keys. Among other features, users of such keys can limit consideration to only likely candidate species, have available characters listed in best (most diagnostic) order for remaining taxa, and select from that list in any desired sequence, bypassing characters that are unfamiliar, difficult to assess, or based on structures that are damaged or missing. Depending on the similarity of taxa included, such keys may be easier to prepare and are almost always much easier to correct, modify, or expand than more traditional printed keys.. Data sets for the keys presented here were prepared in DELTA format (DEscriptive Language for TAxonomy) and transformed to files for use by Intkey, a widely used program available free over the internet (http://delta-intkey.com). | desnyder@cnr.colostate.edu | |||||||||
| Web-Based
Distance Education And Native Professional Communities: Social And Cultural
Constraints |
Sonya J. Le Febre | Forest Rangeland Watershed Stewardship | This study used an anonymous survey to assess the cultural and social conflict experienced by reservation-based professionals in an on-line graduate course. Students experienced no cultural conflict. The biggest social constraint was a busy schedule. Persistence was not high: 36 people were interested, 10 registered, and 4 finished. Tribal governments and colleges want to increase employee qualifications, but funds are limited and staff loads make it difficult to fulfill course obligations. | slefebre@lamar.colostate.edu | ||||||||||
| Digital Colorado: A New Information Paradigm for the 21st Century | Patrick J. Pellicane, Denver
Burns, Celedonio Aguirre-Bravo, Raymond Czaplewski, Rafael Moreno |
Forest, Rangeland, & Watershed Stewardship | A proposed project that is designed to integrate data from diverse sources in a process that will allow users to address issues relevant to the welfare of Colorado's people and environment in a web-based system. | patp@cnr.colostate.edu | ||||||||||
| The National Park Service Geology-GIS Geodatabase Data Model: A Story of Migration II | Stephanie A. O'Meara, Heather
Stanton, James Chappell, Anne R. Poole, Greg Mack |
Geosciences | The NPS Geologic Resources Division is currently migrating from a coverage/shapefile-based geology-GIS data model to a geodatabase-based data model using XML and custom code. The data model includes feature classes, subtypes, domains, topology rules, tables and relationship classes that are being implemented using the XML format. After implementation and debugging, the migration of existing geology-GIS data to the geodatabase data model will commence. | jim_chappell@partner.nps.gov | ||||||||||
| Adoption of
information technology by faculty for teaching in the liberal arts and
natural sciences |
Don Zimmerman & Teresa Yohon | Journalism & Technical Communication | Higher education institutions are investing millions of dollars in IT for teaching. And yet, national studies suggest a low IT adoption rate by faculty for teaching. In an extensive literature review we have not found any studies reporting IT adoption for teaching by faculty at Carnegie Doctoral/Research Extensive institutions (CSU’s classification). We investigated four research questions. | Don.zimmerman@colostate.edu | ||||||||||
| The Broadcast
Flag: Fair Use Infringement or Valid
Copy Protection? |
Peter B. Seel | Journalism and Technical Communication | The FCC’s adoption of “Broadcast Flag” copy protection standards in 2005 for digital television (DTV) may prevent the copying of high-value digital programs in the future. This research analyzes the key technical and policy issues involved in this ruling and their effects on broadcasters, consumers, and the consumer electronics industry in the U.S. | peteseel@colostate.edu | ||||||||||
| Concentration: Mathematics of Information Security | Anton Betten, Robert
Liebler |
Mathematics | The poster presents the Mathematics Concentration "Mathematics of Information Security" | betten@math.colostate.edu | ||||||||||
| GAP -- A system for
discrete mathematics |
Alexander Hulpke | Mathematics | GAP is an open system for computations in discrete mathematics and algebraic structures, in particular computational group theory which has been developed over the last 15 years. One of the four development centers is Fort Collins (the other 3 are in Europe). I want to show capabilities of the system as well as some of the challenges faced by development. | hulpke@math.colostate.edu | ||||||||||
| Atmospheric Research in the Arctic Circle: The ALOMAR-Webber Sodium Lidar System | Phillip Acott, Jia Yue, Sean
Harrell, Biff Williams |
C.Y.(Joe)She, David Krueger | Physics | The ALOMAR-Weber sodium LiDAR (Light Detection And Ranging), located within the Arctic Circle in Andenes, Norway, uses an innovative sum frequency generation (SFG) method to generate the continuous light at 589nm to seed the transmitter of the LiDAR for simultaneous temperature, horizontal wind, and sodium density measurements in the mesopause (80-110Km). This method was developed at Colorado State University. After four years of operation, improvements to this unique and remote SFG system have yielded increased: longevity, stability, and the accuracy of the data. The poster will briefly describe the transmitter layout. The current SFG system possesses an improved ion-beam-sputtered coating on the resonator, deployed recently in March 2004. Additions to the electronics allow for a more stable laser lock on the sodium D2a peak, resulting in improved system stability. Changes in the way we sample the sodium spectrum yields greater resolution in frequency space making data more accurate. Since travel expenses from the U.S. for trained scientists to operate and troubleshoot the system are high, it makes remote control and monitoring a necessity. Remote control and monitoring via the internet will be highlighted. These improvements make the SFG a reliable and accurate solid-state solution for generating seed light for sodium lidar and other applications. ALOMAR (Arctic Lidar Observatory for Middle Atmospheric Research) is an international facility devoted to multi-instrument research of the earth’s middle atmosphere. In addition to several LiDAR systems, there are Radar systems, a rocket launch facility, and many passive infrared instruments and imagers. A recent joint lidar-rocket-radar campaign devoted to the study of the polar summer mesosphere produced 7 high quality papers featured as the cover in a special December 28, 2004, issue of Geophysics Research Letters (GRL). Our LiDAR data showing 40 K change in less than 1 km and a photograph of our dual laser beams are inserts on the cover. | acott@lamar.ColoState.EDU | |||||||||
| The CSU sodium
lidar facility: Current observation capability and science enabled |
Tao Yuan, Tao Li | David A. Krueger, Chiao-Yao (Joe) She | Physics | Lidar (Light Detection And Ranging) is one of the most powerful atmospheric remote sensing techniques for atmospheric research. Starting in the late 80’s, the Na lidar group has been observing the mesopause (80-110km in altitude, also called the “ignorosphere”) by probing the motions of naturally occurring atmospheric Na atoms with a laser system and associated detection equipment developed at Colorado State University. | titus@lamar.colostate.edu | |||||||||
| Secondary
Teachers' Attitudes About Changes in the Public Vocational Technological
School |
Ching-Chiech Ho | George A. Morgan | School of Education | This study investigated: (a) teachers’ attitudes about their vocational school teaching environment, especially changes involving new technologies, (b) teachers’ current and preferred roles in decision making, and (c) teachers’ attitudes about their teaching assignments. The study also explored possible effects of the predictor variables (age, whether or not the teacher had an administrative position, educational level, and the respondent’s specialty area) on the three dependent variables. | jcho@lamar.colostate.edu | |||||||||
| The
Relationship of Institutional Support/Constraints and Instructors'
Characteristics to EFL Taiwanese Teachers' Use of Technology |
Shiaoli Pai | Don Quick, George Morgan | School of Education | The purpose of the study is to investigate the aspects of the availability, frequency of use, and complexity of technology use in teaching English speaking/listening skills in Taiwan. The researchers conclude that instead of increasing the availability of the technology or providing technical support in the institute, encouraging more colleague interaction in using technology, facilitating activities for preparing teaching as well as promoting computer usage in the class/lab will actually motivate. | shirley_pai@yahoo.com | |||||||||
| An
Integrated Display and Analysis Methodology for Multi-Variable Radar
Data |
Brenda A. Dolan | Steven A. Rutledge | Atmospheric Science | Software was developed for near real-time analysis and display of data from a network of four Doppler radars. The integrated radar algorithms include dual-Doppler derived winds, hydrometeor identification, and rainfall estimation. | bdolan@atmos.colostate.edu | |||||||||
| Using
Information Technology in the Diagnosis of Plant Disorders |
Tamla D. Blunt | Ned Tisserat | Bioagricultural Science and Pest Management | In the global environment, rapid dissemination of information about new and emerging plant pathogens is critical to growers and industry personnel. The National Plant Diagnostic Network was formed in response to the threat of bio-terrorism and to orchestrate the flow of information to diagnosticians and regulatory agencies. The NPDN is divided into five regions, Northeast, North Central, Southern, Great Plains and Western regions. | tamla.blunt@colostate.edu | |||||||||
| A
Formal Approach to Incorporating Access Control Features into
Applications |
Eunjee Song | Indrakshi Ray, Robert France | Computer Science | Access control features are often spread across and tangled with other functionality in an application design. This makes understanding, analyzing, and modifying them in a design difficult. Aspect-oriented modeling (AOM) techniques can be used to support separation of access control concerns from other application concerns. Our AOM-based approach supports rigorous specification, verification, and analysis of access control functionality, and facilitates the evolution of its incorporated models. | song@cs.colostate.edu | |||||||||
| An
Inexpensive System for Pattern Analysis of EEG Signals as a Brain-Computer
Interface |
Chuck Anderson | Computer Science | Patterns in electroencephalogram (EEG) signals recorded from the scalp have been found that can be used to discriminate between several mental activities. Such patterns can be used to construct a new mode of communication for paralyzed persons. A hardware-software design of an inexpensive system for EEG recording and real-time analysis is shown as a first step towards such a brain-computer interface. Joint work with Michal Kirby (Math). | anderson@cs.colostate.edu | ||||||||||
| Brain
Wave Tracking and Classification:
Sigma Point Filter, Data Reduction, Classification |
R. Arif Albayrak | Charles W. Anderson | Computer Science | The goal of my studies on EEG signals is to develop a novel tracking method for the brain waves that results in providing weight vectors that minimizes the error for the purpose of classification of mental tasks in real time. | albayrak@cs.colostate.edu | |||||||||
| eIdentity
WebAuth: A Centralized, Web Based,
Authentication Protocol |
Eric Galyon | Daniel Massey | Computer Science | Web applications require identification and authentication for providing services to a specefic end user. Existing methods of authentication have limitations such as requiring numerous credential to be created and managed, allowing insecure access to credential stores, or requiring specialized software to be installed on clients and servers. eIdentity WebAuth provides a standard protocol that helps centralize and share authentication credentials, increases the security of accessing and using the credentials, and minimizes end user requirements. | etg@colostate.edu | |||||||||
| Evaluation
of Motion in Selective Attention |
Trent Williams | Bruce Draper | Computer Science | trent@cs.colostate.edu | ||||||||||
| Evolving
Cooperative Strategies for UAV Teams |
Marc D. Richards | L. Darrell Whitley | Computer Science | We present a Genetic Programming approach to evolve cooperative controllers for teams of UAVs. Our focus is a collaborative search mission in an uncertain and/or hostile environment. | mdr@cs.colostate.edu | |||||||||
| Factors Affecting Web Genre Classification | Adele Howe, Elizabeth Sugar
Boese |
Computer Science | Retrieving relevant documents over the Web is an overwhelming task when search engines return thousands of Web documents. Sifting through these documents is time consuming and sometimes leads to an unsuccessful search. One problem is that most search engines rely on matching a query to documents based solely on topical keywords. However, many users of search engines have a particular genre in mind for the desired documents. | boese@cs.colostate.edu | ||||||||||
| Identifying
Algorithm Performance Factors for an Oversubscribed Scheduling Problem |
Laura Barbulescu | Adele Howe, L. Darrell Whitley | Computer Science | We identify key algorithm performance factors when scheduling for the U.S.A. Air Force Satellite Control Network. The search space is dominated by plateaus, favoring algorithms like a genetic algorithm and Squeaky Wheel Optimization, which take large steps through the space. We considered two local search versions: one excels, the other performs poorly. We present empirical results to characterize the search space and explain the performance discrepancy for the two local search versions. | laura@cs.colostate.edu | |||||||||
| Measuring
Mobility: Evolutionary algorithms and
global search |
Monte Lunacek | L. Darrell Whitley | Computer Science | The global search properties of heuristic search algorithms are not well understood. In this poster, we introduce a new metric, mobility, that quantifies the dispersion of local optima visited during a search. This allows us to explore two questions: How dispersed are the local optima visited during a search? How does mobility relate to algorithm performance? We compare local search with two evolutionary algorithms, CHC and CMA-ES, on a set of non-separable, non-symmetric, multi-modal test functions. Given our mobility metric, we show that algorithms visiting more disperse local optima tend to be better optimizers. | lunacek@cs.colostate.edu | |||||||||
| Modeling
Reward Functions for Incomplete State Representations via Echo State
Networks |
Keith Bush | Charles Anderson | Computer Science | The Echo State Network (ESN) architecture is used as a recurrent network strategy for approximating the Q-function of the mass-spring-damper linear dynamical system when only partial state is observable. The ESN architecture's approximation performance is compared against feed forward neural networks given perfect state information (FNN) and a finite window of time-delayed partial state (TDNN), respectively. Both feed forward representations are known to perform well in approximating the Q-function in this problem domain. We demonstrate that the ESN, given partial state, well-represents temporally dependent rewards and exhibits similar performance to FNN and TDNN architectures in approximating the Q-function during on-line learning. | kbush@cs.colostate.edu | |||||||||
| Optimal Software Pipelining and Register Tiling | Lakshminarayanan Renganarayana,
Ramakrishna Upadrasta |
Sanjay Rajopadhye | Computer Science | Best performance for a given class of programs is achieved by using an optimal combination of program transformations. Program optimizations interact in subtle ways. For the class of uniform dependence loops, we formulate an optimization problem that captures the interaction between three transformations viz., skewing, tiling and permutation. We show how the optimization problem can be decomposed and solved efficiently to obtain an optimal combination of transformations. | ln@cs.colostate.edu | |||||||||
| Predicting
Malicious Attack From Authorized Insider |
Nayot Poolsappasit | Indrajit Ray | Computer Science | nayot@cs.colostate.edu | ||||||||||
| Reinforcement
Learning Control with Robust Stability |
Chuck Anderson | Computer Science | An artificial neural network is placed in parallel with a conventional controller and trained via reinforcement learning to fine tune the performance of the controller. Stability while learning is guaranteed by representing the nonlinearities and time-varying parameters of the network with integral quadratic constraints, allowing the application of robust control theory to analyze the stability of the feedback loop. An application to heating and cooling is discussed. Joint work with Peter Young (ECE) and Douglas Hittle (ME). | anderson@cs.colostate.edu | ||||||||||
| Simplifying
Reductions |
Seema Gautam | Sanjay Rajopadhye | Computer Science | The polyhedral model enables very high level programming and powerful static analysis. Expressing computations in the model offers an opportunity to decrease their complexity. We develop the foundations for complexity reduction by equational transformations. We give necessary and sufficient conditions for their validity and present an algorithm for choosing optimal transformations. We extend our algorithm to retain optimality while exploiting further simplification through algebraic properties. | ggupta@cs.colostate.edu | |||||||||
| Stochastic
Quadsearch |
Bryan Schlief | L. Darrell Whitley | Computer Science | Recently a Gray code probability distribution function was developed that allows for mutations of evolution stategy values that mimic properties of mutations of binary Gray codes. This algorithm performs poorly on test functions in which there are ridges in the search space that are not aligned with dimensions specified by the problem. We extend this work and develop an evolution strategy that directs where samples are taken from, such that large mutations are generated first, followed by successively smaller mutations. This new algorithm performs better on rotated test functions. | schlief@cs.colostate.edu | |||||||||
| Reorder
Density (RD): A Formal, Comprehensive
Metric for Packet Reordering |
Nischal M. Piratla, Abhijit A. Bare | Anura Jayasumana | Electrical & Computer Engineering | Increase in link speeds, increased parallelism within routers/switches, QoS, point to future networks with increased packet reordering. A formal analysis of reordering is carried out and Reorder Density (RD) metric is defined for measurement and characterization of packet reordering. RD captures the amount and degree of reordering, and used to define the reorder response of networks under stationary conditions. It is shown that the reorder response of the network formed by cascading two subnets. | Nischal.Piratla@colostate.edu | |||||||||
| Resource
Allocation for Periodic Applications in a Shipboard Environment |
Vladimir Shestak | H.J. Siegel | Electrical & Computer Engineering | Providing efficient workload management is an important issue for a large-scale heterogeneous distributed computing environment where a set of periodic applications is executed. The considered distributed system is expected to operate in an environment where the input workload is likely to change unpredictably, possibly invalidating a resource allocation that was based on the initial workload estimate. The tasks consist of multiple application strings, each made up of an ordered sequence of applications. There are quality of service (QoS) constraints that must be satisfied for each string. This work addresses the problem of finding a robust initial allocation of resources to application strings that is able to absorb some level of unknown input workload increase without rescheduling. An allocation feasibility analysis and four heuristics are designed for finding a near-optimal allocation of resources. The performance of the proposed heuristics is evaluated and compared using simulation. The proposed heuristics also are compared to a mathematically derived upper bound. | shestak@engr.colostate.edu | |||||||||
| Temperature
dependence of Temporally and Spectrally Resolved Photoluminescence study in
InGaAsN Quantum Well |
Lifang Xu, O. Antón | D. Patel, G. Vaschenko, C.S. Menoni | Electrical & Computer Engineering | We investigated directly the dynamical interplay of optically-generated excitons and unbound electron-hole pairs in InGaAsN/GaAs single quantum well by time resolved Photoluminescence intensity and spectroscopy. We observed an anomalous high energy peak existing in time resolved PL spectra during the short time delay for temperature varying from 20K to 100K. We suggest this anomalous peak as a proof the existence of self-formed quantum dot | lfxu@engr.colostate.edu | |||||||||
| Nano-Imaging with an Extreme Ultraviolet Laser | Courtney Brewer |
Carmen Menoni | Electrical and Computer Engineering | We have demonstrated nanometer scale imaging in both transmission and reflective modes using 47 nm wavelength light from a tabletop capillary-discharge laser. This compact imaging system employs a Sc/Si multilayer-coated Schwarzschild condenser and a freestanding zone plate objective. Images with a spatial resolution better than 140 nm were obtained. These results foretell the development of compact imaging tools for nanoscience and nanotechnology. | brewerca@holly.colostate.edu | |||||||||
| Network
Emulator Tool For Large Ip Networks; Analysis And Methodologies |
Daniel A. Vivanco | Anura Jayasumana | Electrical and Computer Engineering | Today’s Internet researchers rely on two main methodologies for designing and evaluating novel and existing network protocols and architectures. On one hand, packet-by-packet simulation based approaches offer repeatable and controlled experimentation. Unfortunately these methods are not suitable for reproducing the experiment’s random nature and excessive computational overhead is demanded for large network scenarios. On the other hand, networking testbeds mimic more closely realistic characteristics, but constructing such testbeds is labor intensive and costly. As an alternative, network emulators have emerged as a potential solution. This approach represents a hybrid combination of the two mentioned methodologies, capable of providing interaction between real world and simulated network elements. In this research we describe in detail the design and testing of the Overall Trend Replicating Network Emulator Tool (OTRENET), which aims to fulfill the above necessity. Two main topics are covered in this research. First, the design of an adaptive average traffic sampler by time frame segmentation algorithm, which aims to mimic the overall behavior of real network scenarios without including unnecessary computation/time overhead. Second, the design and testing of a new scaling methodology for simulating-emulating large IP Networks in real time. | dvivanco@engr.colostate.edu | |||||||||
| Bound electron contribution to soft x-ray laser interferograms of dense plasmas | J. Filevich, F. Smith, J. Dunn ,
R. Keenan, J.R. Hunter, S.J. Moon, J. Nilsen, A. Ng, V.N. Shlyaptsev |
|||||||||||||