Computer Sciences Commons^™

Study Of Parallel Programming Models On Computer Clusters With Accelerators, Chenggang Lai

Graduate Theses and Dissertations

In order to reach exascale computing capability, accelerators have become a crucial part in developing supercomputers. This work examines the potential of two latest acceleration technologies, Intel Many Integrated Core (MIC) Architecture and Graphics Processing Units (GPUs). This thesis applies three benchmarks under 3 different configurations, MPI+CPU, MPI+GPU, and MPI+MIC. The benchmarks include intensely communicating application, loosely communicating application, and embarrassingly parallel application. This thesis also carries out a detailed study on the scalability and performance of MIC processors under two programming models, i.e., offload model and native model, on the Beacon computer cluster.

According to different benchmarks, the results …

Exploring User-Provided Connectivity, Mohammad H. Afrasiabi, Roch Guerin

All Computer Science and Engineering Research

Network services often exhibit positive and negative externalities that affect users' adoption decisions. One such service is "user-provided connectivity" or UPC. The service offers an alternative to traditional infrastructure-based communication services by allowing users to share their "home base" connectivity with other users, thereby increasing their access to connectivity. More users mean more connectivity alternatives, i.e., a positive externality, but also greater odds of having to share one's own connectivity, i.e., a negative externality. The tug of war between positive and negative externalities together with the fact that they often depend not just on how many but also which users …

Neural Network Dynamic Progamming Constrained Control Of Distributed Parameter Systems Governed By Parabolic Partial Differential Equations With Application To Diffusion-Reaction Processes, Behzad Talaei, Hao Xu, S. Jagannathan

Electrical and Computer Engineering Faculty Research & Creative Works

In this paper, a novel neural network (NN) adaptive dynamic programming (ADP) control scheme for distributed parameter systems (DPS) governed by parabolic partial differential equations (PDE) is introduced in the presence of control constraints and unknown system dynamics. First, Galerkin method is utilized to develop a relevant reduced order system which captures the dominant dynamics of the DPS. Subsequently, a novel control scheme is proposed over finite horizon by using NN ADP. To relax the requirement of system dynamics, a novel NN identifier is developed. More-over, a second NN is proposed to estimate online the time-varying non-quadratic value function from …

Mechatronics In Electrical Efficiency And Environmental Impact, Sherif Hyseni

UBT International Conference

This paper focuses on the mechatronic technology development, and its influent on electrical efficiency with a direct result in cost and environment. Considering the well-known European Union standard “20-20-20” and the possibility and responsibility of Kosovo in this field, the role of energy efficiency has great impact for future directions. This paper refers to a new technology solution for heating that is available on the local market, called “DAIKIN” by analyzing the functionality of this device and its promise to save up to 40% of the heating costs. Implementation of “DAIKIN” technology in state institutions in Prishtina, and also the …

A New Transfer Impedance Based System Equivalent Model For Voltage Stability Analysis, Yang Wang, Caisheng Wang, Feng Lin, Wenyuan Li, Le Yi Wang, Junhui Zhao

Electrical & Computer Engineering and Computer Science Faculty Publications

This paper presents a new transfer impedance based system equivalent model (TISEM) for voltage stability analysis. The TISEM can be used not only to identify the weakest nodes (buses) and system voltage stability, but also to calculate the amount of real and reactive power transferred from the generator nodes to the vulnerable node causing voltage instability. As a result, a full-scale view of voltage stability of the whole system can be presented in front of system operators. This useful information can help operators take proper actions to avoid voltage collapse. The feasibility and effectiveness of the TISEM are further validated …

Cubic Spline Interpolation By Solving A Recurrence Equation Instead Of A Tridiagonal Matrix, Peter Revesz

School of Computing: Conference and Workshop Papers

The cubic spline interpolation method is proba- bly the most widely-used polynomial interpolation method for functions of one variable. However, the cubic spline method requires solving a tridiagonal matrix-vector equation with an O(n) computational time complexity where n is the number of data measurements. Even an O(n) time complexity may be too much in some time-ciritical applications, such as continuously estimating and updating the flight paths of moving objects. This paper shows that under certain boundary conditions the tridiagonal matrix solving step of the cubic spline method could be entirely eliminated and instead the coefficients of the unknown cubic polynomials …

Designing A Bayer Filter With Smooth Hue Transition Interpolation Using The Xilinx System Generator, Zhiqiang Li, Peter Revesz

School of Computing: Conference and Workshop Papers

This paper describes the design of a Bayer filter with smooth hue transition using the System Generator for DSP. We describe and compare experimentally two different designs, one based on a MATLAB implementation and the other based on a modification of the Bayer filter using bilinear interpolation.

A Game-Theoretic Analysis Of The Nuclear Non-Proliferation Treaty, Peter Revesz

School of Computing: Conference and Workshop Papers

Although nuclear non-proliferation is an almost universal human desire, in practice, the negotiated treaties appear unable to prevent the steady growth of the number of states that have nuclear weapons. We propose a computational model for understanding the complex issues behind nuclear arms negotiations, the motivations of various states to enter a nuclear weapons program and the ways to diffuse crisis situations.

Estimating The Flight Path Of Moving Objects Based On Acceleration Data, Peter Revesz

School of Computing: Conference and Workshop Papers

Inertial navigation is the problem of estimating the flight path of a moving object based on only acceleration measurements. This paper describes and compares two approaches for inertial navigation. Both approaches estimate the flight path of the moving object using cubic spline interpolation, but they find the coefficients of the cubic spline pieces by different methods. The first approach uses a tridiagonal matrix, while the second approach uses recurrence equations. They also require different boundary conditions. While both approaches work in O(n) time where n is the number of given acceleration measurements, the recurrence equation-based method can be easier updated …

Organizing Video Search Results To Adapted Semantic Hierarchies For Topic-Based Browsing, Jiajun Wang, Yu-Gang Jiang, Qiang Wang, Kuiyuan Yang, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

Organizing video search results into semantically structured hierarchies can greatly improve the efficiency of browsing complex query topics. Traditional hierarchical clustering techniques are inadequate since they lack the ability to generate semantically interpretable structures. In this paper, we introduce an approach to organize video search results to an adapted semantic hierarchy. As many hot search topics such as celebrities and famous cities have Wikipedia pages where hierarchical topic structures are available, we start from the Wikipedia hierarchies and adjust the structures according to the characteristics of the returned videos from a search engine. Ordinary clustering based on textual information of …

Traccs: Trajectory-Aware Coordinated Urban Crowd-Sourcing, Cen Chen, Shih-Fen Cheng, Aldy Gunawan, Archan Misra, Koustuv Dasgupta, Deepthi Chander

Research Collection School Of Computing and Information Systems

We investigate the problem of large-scale mobile crowd-tasking, where a large pool of citizen crowd-workers are used to perform a variety of location-specific urban logistics tasks. Current approaches to such mobile crowd-tasking are very decentralized: a crowd-tasking platform usually provides each worker a set of available tasks close to the worker's current location; each worker then independently chooses which tasks she wants to accept and perform. In contrast, we propose TRACCS, a more coordinated task assignment approach, where the crowd-tasking platform assigns a sequence of tasks to each worker, taking into account their expected location trajectory over a wider time …

Click-Through-Based Subspace Learning For Image Search, Yingwei Pan, Ting Yao, Xinmei Tian, Houqiang Li, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

One of the fundamental problems in image search is to rank image documents according to a given textual query. We address two limitations of the existing image search engines in this paper. First, there is no straightforward way of comparing textual keywords with visual image content. Image search engines therefore highly depend on the surrounding texts, which are often noisy or too few to accurately describe the image content. Second, ranking functions are trained on query-image pairs labeled by human labelers, making the annotation intellectually expensive and thus cannot be scaled up. We demonstrate that the above two fundamental challenges …

Cama: Efficient Modeling Of The Capture Effect For Low Power Wireless Networks, Behnam Dezfouli, Marjan Radi, Kamin Whitehouse, Shukor Abd Razak, Hwee-Pink Tan

Research Collection School Of Computing and Information Systems

Network simulation is an essential tool for the design and evaluation of wireless network protocols, and realistic channel modeling is essential for meaningful analysis. Recently, several network protocols have demonstrated substantial network performance improvements by exploiting the capture effect, but existing models of the capture effect are still not adequate for protocol simulation and analysis. Physical-level models that calculate the signal-to-interference-plus-noise ratio (SINR) for every incoming bit are too slow to be used for large-scale or long-term networking experiments, and link-level models such as those currently used by the NS2 simulator do not accurately predict protocol performance. In this article, …

Localization Of Near-Field Radio Controlled Unintended Emitting Sources In The Presence Of Multipath Fading, Nurbanu Guzey, Hao Xu, Sarangapani Jagannathan

Electrical and Computer Engineering Faculty Research & Creative Works

Localization of near-field unintended emitting radio controlled (RC) devices under a multipath environment is considered in this paper using a uniform linear array (ULA). Since received signals are dependent on both angle of arrival (AoA) and distance from the RC devices to the ULA in the near-field scenario, traditional localization schemes based on received signal strength indicator, time difference in arrival, which only estimate either AoA or distance, are unsuitable. Therefore, a novel smooth 2-D multiple signal classification (MUSIC) near-field localization scheme is developed to locate RC devices under a multipath environment with a possible location error of 0.5 m. …

Vireo @ Trecvid 2014: Instance Search And Semantic Indexing, Wei Zhang, Hao Zhang, Ting Yao, Yijie Lu, Jingjing Chen, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

This paper summarizes the following two tasks participated by VIREO group: instance search and semantic indexing. We will present our approaches and analyze the results obtained in TRECVID 2014 benchmark evaluation

Distributed Owl El Reasoning: The Story So Far, Raghava Mutharaju, Pascal Hitzler, Prabhaker Mateti

Computer Science and Engineering Faculty Publications

Automated generation of axioms from streaming data, such as traffic and text, can result in very large ontologies that single machine reasoners cannot handle. Reasoning with large ontologies requires distributed solutions. Scalable reasoning techniques for RDFS, OWL Horst and OWL 2 RL now exist. For OWL 2 EL, several distributed reasoning approaches have been tried, but are all perceived to be inefficient. We analyze this perception. We analyze completion rule based distributed approaches, using different characteristics, such as dependency among the rules, implementation optimizations, how axioms and rules are distributed. We also present a distributed queue approach for the classification …

A Cross Layer Routing Scheme For Passive Rfid Tag-To-Tag Communication, Haifeng Niu, S. Jagannathan

Electrical and Computer Engineering Faculty Research & Creative Works

A routing protocol for Passive RFID tag-to-tag networks is not reported in contrast with routing protocols for battery powered mobile wireless ad hoc and sensor networks. In this paper, we provide a cross-layer approach for passive RFID tag-to-tag communication. In the data link layer, we designed the medium access control (MAC) protocol which is suitable for passive tag-to-tag communication. In the network layer, we developed the optimal link cost multipath routing (OLCMR) protocol by using modulation depth as the link cost. Simulation results verify that proposed routing protocol consumes less energy when it is compared to optimum link state routing …

Strategies For Sensor Data Aggregation In Support Of Emergency Response, X. Wang, A. Walden, M. Weigle, S. Olariu

Computer Science Presentations

PDF of a powerpoint presentation from the Military Communications Conference (MILCOM), Baltimore, Maryland, October 7, 2014. Also available on Slideshare.

Indoor Acoustic Mapping And Geolocation, Thomas Shyy

Open Access Theses

The Global Positioning System (GPS) is widely used in navigations of flights, vehicles and travelers nowadays. However, the signals are not always available and there are some cases where the receptions of the GPS signal can be lost because of the presence of urban canyons, buildings, tunnels and forests. Another down side for the GPS is that indoor positioning can be really difficult because of the greater signal interference from the indoor.

On the other hand, the concept of echolocation has also been used for positioning purposes. Echolocation offers a promising approach to improve the quality of life of people …

Pushing The Boundaries Of Tractable Ontology Reasoning, David Carral, Cristina Feier, Bernardo Cuenca Grau, Pascal Hitzler, Ian Horrocks

Computer Science and Engineering Faculty Publications

We identify a class of Horn ontologies for which standard reasoning tasks such as instance checking and classification are tractable. The class is general enough to include the OWL 2 EL, QL, and RL profiles. Verifying whether a Horn ontology belongs to the class can be done in polynomial time. We show empirically that the class includes many real-world ontologies that are not included in any OWL 2 profile, and thus that polynomial time reasoning is possible for these ontologies.

Wi-Fi Real Time Location Systems, Benjamin A. Doll

Open Access Theses

This thesis objective was to determine the viability of utilizing an untrained Wi-Fi real time location system as a GPS alternative for indoor environments. Background research showed that GPS is rarely able to penetrate buildings to provide reliable location data. The benefit of having location information in a facility and how they might be used for disaster or emergency relief personnel and their resources motivated this research. A building was selected with a well-deployed Wi-Fi infrastructure and its untrained location feature was used to determine the distance between the specified test points and the system identified location. It was found …

Rapid Indirect Trajectory Optimization On Highly Parallel Computing Architectures, Thomas Antony

Open Access Theses

Trajectory optimization is a field which can benefit greatly from the advantages offered by parallel computing. The current state-of-the-art in trajectory optimization focuses on the use of direct optimization methods, such as the pseudo-spectral method. These methods are favored due to their ease of implementation and large convergence regions while indirect methods have largely been ignored in the literature in the past decade except for specific applications in astrodynamics. It has been shown that the shortcomings conventionally associated with indirect methods can be overcome by the use of a continuation method in which complex trajectory solutions are obtained by solving …

Impact Of License Selection On Open Source Software Quality, Benjamin J. Cotton

Open Access Theses

Open source software plays an important part in the modern world, powering businesses large and small. However, little work has been done to evaluate the quality of open source software. Two different license paradigms exist within the open source world, and this study examines the difference in software quality between them. In this thesis, the author uses technical debt as a measure of software quality. Eighty open source projects (40 from each paradigm) were downloaded from the popular open source hosting website SourceForge. Using complexity, code duplication, comments, and unit test coverage as inputs to the SonarQube technical debt model, …

Digital Image Correlation Of Heterogeneous Deformations In Polycrystalline Material With Electron Backscatter Diffraction, Javier Esquivel

Open Access Theses

This work establishes the ability to conduct digital image correlation (DIC) investigations at varying length scales. DIC allows for a computational method of strain field measurements using multiple images to track random speckle patterns on material surfaces. The use of a powder silicon oxide speckle allows for high optical magnification correlation using conventional load frames. Self-assembling gold nanoparticles provide sub-micron resolution speckle patterns to study microstructure influences on deformation using scanning electron microscopy. The complex microstructure in aerospace grade aluminum and nickel-based superalloys, give rise to varied deformation fields, which can be studied using electron backscatter diffraction. Specimen preparation techniques, …

Dynamic Wireless Power Transfer System For The Unmanned Aerial Vehicles, Tae Sup Lee

Open Access Theses

UAVs have limitless applications to help our daily lives for the autonomous operations. UAVs have a limited power capacity due to weight constraints and are therefore not able to travel long distances. Ground stations for recharging UAVs throughout different points can increase the flight time of the UAVs with the current UAV battery capacity. This study investigates how the wireless charging system for the ground station can be made more robust when there are misalignments. The wireless charging system is monitored by an Arduino, micro controller, to assess the current condition of charging. The Arduino is able to change the …

Functional Programming Abstractions For Weakly Consistent Systems, Sivaramakrishnan Krishnamoorthy Chandrasekaran

Open Access Dissertations

In recent years, there has been a wide-spread adoption of both multicore and cloud computing. Traditionally, concurrent programmers have relied on the underlying system providing strong memory consistency, where there is a semblance of concurrent tasks operating over a shared global address space. However, providing scalable strong consistency guarantees as the scale of the system grows is an increasingly difficult endeavor. In a multicore setting, the increasing complexity and the lack of scalability of hardware mechanisms such as cache coherence deters scalable strong consistency. In geo-distributed compute clouds, the availability concerns in the presence of partial failures prohibit strong consistency. …

Theory Of Topological Insulators And Its Applications, Parijat Sengupta

Open Access Dissertations

An important pursuit in semiconductor physics is to discover new materials to sustain the continuous progress and improvements in the current electronic devices. Traditionally, three material types are in use: 1) Metals 2) Semiconductors 3) Insulators. All the three material types are classified according to the energy gap between conduction and valence bands derived from band theory of solids. Recent theoretical predictions and confirmed by experimental observations have provided evidence that there exists materials which behave as insulators in the bulk but possess gapless conducting states on the surface. These new class of materials are called topological insulators (TI). In …

Digital Provenance - Models, Systems, And Applications, Salmin Sultana

Open Access Dissertations

Data provenance refers to the history of creation and manipulation of a data object and is being widely used in various application domains including scientific experiments, grid computing, file and storage system, streaming data etc. However, existing provenance systems operate at a single layer of abstraction (workflow/process/OS) at which they record and store provenance whereas the provenance captured from different layers provide the highest benefit when integrated through a unified provenance framework. To build such a framework, a comprehensive provenance model able to represent the provenance of data objects with various semantics and granularity is the first step. In this …

Algebraic Level Sets For Cad/Cae Integration And Moving Boundary Problems, Kritika Upreti

Open Access Dissertations

Boundary representation (B-rep) of CAD models obtained from solid modeling kernels are commonly used in design, and analysis applications outside the CAD systems. Boolean operations between interacting B-rep CAD models as well as analysis of such multi-body systems are fundamental operations on B-rep geometries in CAD/CAE applications. However, the boundary representation of B-rep solids is, in general, not a suitable representation for analysis operations which lead to CAD/CAE integration challenges due to the need for conversion from B-rep to volumetric approximations. The major challenges include intermediate mesh generation step, capturing CAD features and associated behavior exactly and recurring point containment …

Image Analysis Using Visual Saliency With Applications In Hazmat Sign Detection And Recognition, Bin Zhao

Open Access Dissertations

Visual saliency is the perceptual process that makes attractive objects "stand out" from their surroundings in the low-level human visual system. Visual saliency has been modeled as a preprocessing step of the human visual system for selecting the important visual information from a scene. We investigate bottom-up visual saliency using spectral analysis approaches. We present separate and composite model families that generalize existing frequency domain visual saliency models. We propose several frequency domain visual saliency models to generate saliency maps using new spectrum processing methods and an entropy-based saliency map selection approach. A group of saliency map candidates are then …