Master of Science / Department of Computing and Information Sciences / Eugene Vasserman / As medical systems expand to allow for the increase the number of devices, new ways to protect patient safety have be developed. The Integrated Clinical Environment, ICE, standard sets up a set of standards that define what an integrated hospital system is. Within the specification is a direct call for a forensic logger that can be used to review patient and system data. The MDCF is one implementation of the ICE standard, but it lacked a key component the ICE standard requires, a logger. Many loggers exist in industry, with varying rates of success and usefulness. A medically sound logger has to be able to completely retell exactly what happened during an event, including patient, device, and system information, so that the right medical professional can provide the best care. Several loggers have been built for MDCF, but few were practical due to the invasiveness of the service. A universal subscriber, a service that is able to connect to all publishing data streams, logging service was built for the MDCF which has the ability to record all information that passes over the MDCF messaging service. This implementation was then stress tested with varying numbers of devices and amounts of data. A reviewing tool was also built that allows for replay of device data that is similar to the original device UI. Future work will include looking into storing system information such as state changes within MDCF and system health. There is also a push to further integrate the forensic reviewer into the core MDCF UI.

Master of Science / Department of Computer Science / Eugene Vasserman / There is continuing growth in the need to secure critical infrastructures from malicious adversaries. These adversaries can attack systems from different forms. They can physically break in and steal something important, or they can attack from the cyber realm in order to steal critical information. This project combines the modeling process for physical spaces along with a logic reasoning tool that can identify the state of a networked device in order to analyze large enterprise systems for combined cyber-physical vulnerabilities. Using a pure model checker would not be able to handle the near infinite states that a computer or networked device may be in. Therefore this new approach combines the use of a logic analyzer tool that with a well-defined set of rules that reasons about the security and trustworthiness of devices in the model. While there has been long study of how to secure a building from intrusion, and much research about defense against cyber attacks, there is always a large gap between the two in practice. This approach may no longer be sufficient against today’s adversaries and offers little to no defense against insider threats. Combining the two in this new form allows for a more complete security view and protection against more advanced adversaries. Then this thesis shows how this approach meets a series of requirements for an effective vulnerability analysis. This is achieved by executing a model based on a real world facility with a series of induced faults that would on their own not be enough to be a vulnerability but tied together would have series consequences. This thesis shows how this approach can then be used to detail potentially unseen vulnerabilities and develop fixes for them to help create a more secure facility.

Master of Science / Department of Computing and Information Sciences / Daniel A. Andresen / Tutorials and Quiz Android Mobile application is an application which gives you brief tutorials on various programming languages and technologies to the user who has installed this application. It gives video tutorials and training on different technologies. The mobile application also contains sample code, interview questions, and answers. This application also conducts a quiz to the user who is interested in a particular technology for checking their knowledge on the subject. It has different levels like Basic Level and Advanced Level quiz. This application conducts a quiz in the form of multiple choice questions. After the completion of the quiz the application generates reports based on the quiz conducted. The user can also test their knowledge on those technologies based on a timed quiz, we can include the above technology questions in this application. We can give a time frame for each question and all the questions are to be answered in that specified time period. Say for example we can give 20 seconds to answer a question. If the user gives an incorrect answer, will be exited out of the quiz.

Master of Science / Computing and Information Sciences / Daniel A. Andresen / One among the many inventions of Benjamin Franklin is the Glass Armonica, a musical instrument whose sound source was a series of resonating glass vessels. However, the Irish musician Richard Pockrich is typically credited as the first to play an instrument composed of glass vessels, called the Glass Harp in 1741, by rubbing his fingers around the rims. In this project “Singing Wine Glasses”, the principle of Franklin’s glass armonica is demonstrated with a wine glass. One hand is used to hold the glass steady at the base. The rim of glass is gently pressed with a moistened finger of the other hand and drawn in a circle around. When the pressure and amount of moisture are just right, the slight friction between the finger and the rim of glass causes vibrations in the sides of the glass. At a particular frequency, called the resonant frequency, the sides of the glass will vibrate most easily. The resonant frequency of wine glasses is typically within the range of human hearing (20-20,000 Hz), so the resulting resonant vibration is heard as a tone. The glass starts to sing when the vibration gets the molecules moving at their natural frequency. The resonant frequency changes with the amount of water filled in the glass. This android application deals with virtual glasses that serve the purpose of wine glasses filled with different amounts of water. Swiping on the glass edges would produce music, as per Franklin’s principle. The users would be free to select the number of glasses they want to play, and the amount of water-level in each glass. This application would also come with an enhanced feature of sustaining a particular note until the finger is released from the glass.

Doctor of Philosophy / Department of Computing and Information Sciences / Gurdip Singh / Many frameworks have been proposed which provide distributed algorithms encapsulated as middleware services to simplify application design. The developers of such algorithms are faced with two opposing forces. One is to design generic algorithms that are reusable in a large number of applications. Efficiency considerations, on the other hand, force the algorithms to be customized to specific operational contexts. This problem is often attacked by simply re-implementing all or large portions of an algorithm. We propose InDiGO, an infrastructure which allows design of generic but customizable algorithms and provides tools to customize such algorithms for specific applications. InDiGO provides the following capabilities: (a) Tools to generate intermediate representations of an application which can be leveraged for analysis, (b) Mechanisms to allow developers to design customizable algorithms by exposing design knowledge in terms of configurable options, and (c) An optimization engine to analyze an application to derive the information necessary to optimize the algorithms. Specifically, we optimize algorithms by removing communication which is redundant in the context of a specific application. We perform three types of optimizations: static optimization, dynamic optimization and physical topology-based optimization. We present experimental results to demonstrate the advantages of our infrastructure.

Master of Science / Department of Computing and Information Sciences / William H. Hsu / Alternative splicing is a mechanism for generating different gene transcripts (called iso- forms) from the same genomic sequence. Finding alternative splicing events experimentally is both expensive and time consuming. Computational methods in general, and EST analy- sis and machine learning algorithms in particular, can be used to complement experimental methods in the process of identifying alternative splicing events. In this thesis, I first iden- tify alternative splicing exons by analyzing EST-genome alignment. Next, I explore the predictive power of a rich set of features that have been experimentally shown to affect al- ternative splicing. I use these features to build support vector machine (SVM) classifiers for distinguishing between alternatively spliced exons and constitutive exons. My results show that simple, linear SVM classifiers built from a rich set of features give results comparable to those of more sophisticated SVM classifiers that use more basic sequence features. Finally, I use feature selection methods to identify computationally the most informative features for the prediction problem considered.

Master of Science / Department of Computing and Information Sciences / David A. Gustafson William Hsu / In this study, we identify a new method of tracking motion over a sequence of images using feature point clusters. We identify and implement a system that takes as input a sequence of images and generates clusters of SIFT features using the K-Means clustering algorithm. Every time the system processes an image it compares each new cluster to the clusters of previous images, which it stores in a local cache. When at least 25% of the SIFT features that compose a cluster match a cluster in the local cache, the system uses the centroid of both clusters in order to determine the direction of travel. To establish a direction of travel, we calculate the slope of the line connecting the centroid of two clusters relative to their Cartesian coordinates in the secondary image. In an experiment using a P3-AT mobile robotic agent equipped with a digital camera, the system receives and processes a sequence of eight images. Experimental results show that the system is able to identify and track the motion of objects using SIFT feature clusters more efficiently when applying spatial outlier detection prior to generating clusters.

Doctor of Philosophy / Department of Computing and Information Sciences / Matthew Dwyer / John M. Hatcliff / To analyze an open system, one needs to close it with a definition of its environment, i.e., its execution context. Environment modeling is a significant challenge: environment models should be general enough to permit analysis of large portions of a system's possible behaviors, yet sufficiently precise to enable cost-effective reasoning. This thesis presents the Bandera Environment Generator (BEG), a toolset that automates generation of environment models to provide a restricted form of modular model checking of Java programs, where the module's source code is the subject of analysis along with an abstract model of the environment's behavior. Since the most general environments do not allow for tractable model checking, BEG has support for restricting the environment behavior based on domain-specific knowledge and assumptions about the environment behavior, which can be acquired from a variety of sources. When the environment code is not available, developers can encode their assumptions as an explicit formal specification. When the environment code is available, BEG employs static analyses to extract environment assumptions. Both specifications and static analyses can be tuned to reflect domain-specific knowledge, i.e., to describe domain-specific aspects of the environment behavior. Initially, BEG was implemented to handle general Java applications; later, it was extended to handle two specific domains: Graphical User Interfaces (GUI) implemented using the Swing/AWT libraries and web applications implemented using the J2EE framework. BEG was evaluated on several non-trivial case studies, including industrial applications from NASA, SUN, and Fujitsu. This thesis presents the domain-specific environment generation for GUI and web applications and describes BEG, its extensible architecture, usage, and how it can be extended to handle new domains.

Master of Science / Department of Computing and Information Sciences / Mitchell L. Neilsen / Safety-critical systems embedded in avionics and automotive systems are becoming increasing complex. Components with different requirements typically share a common distributed platform for communication. To accommodate varied requirements, many of these distributed real-time systems use FlexRay communication network. FlexRay supports both time triggered and event-triggered communications. In such systems, it is vital to establish a consistent view of all the associated processes to handle fault-tolerance. This task can be accomplished through the use of a Process Group Membership Protocol. This protocol must provide a high level of assurance that it operates correctly. In this thesis, we provide for the verification of one such protocol using Model Checking. Through this verification, we found that the protocol may remove nodes from the group of operational nodes in the communicating network at a fast rate. This may lead to exhaustion of the system resources by the protocol, hampering system performance. We determine allowable rates of failure that do not hamper system performance.

Master of Science / Department of Computing and Information Sciences / Daniel A. Andresen / Increasing web content and Internet traffic is making web prediction models popular. A web prediction model helps to predict user requests ahead of time, making web servers more responsive. It caches these pages at the server side or pre-sends the response to the client to reduce web latency. Several prediction techniques have been tried in the past; Markov based prediction models being the most popular ones. Among these, the All-K[superscript]th -order Markov model has been found to be most effective. In this project, a Markov tree is designed, which is a fourth order model but behaves like an All-K[superscript]th-order Markov model because of its ability to recognize different order models according to the height of the tree. It has dual characteristics of good applicability and predictive accuracy. A Markov tree gives a complete description on the frequency with which a particular state occurs, and the number of times a path to a particular state is used, to access its child nodes. Further, the model can be pruned to eliminate states that have very little contribution towards the accuracy of the model. In this work, an evolutionary model is designed that makes use of a fitness function. The fitness function is a weighted sum of precision and the extent of coverage that the model offers. This helps to generate a model with reduced complexity. Results indicate that this model performs consistently with good predictive accuracy among different log files. The evolutionary approach helps to train the model to make predictions commensurate to current web browsing patterns.