Secure Critical Care Resource Optimization based on Heterogeneous Vital Signs
Year of Dissertation:
2010
Preventable, in-hospital errors account for a substantial number of deaths and injuries in the United States. Various studies estimate that such deaths number between 100,000 and 200,000 each year. One of the key challenges in critical care is a legacy of existing largely wired medical networks, which due to the complexity of their constituent heterogeneous medical devices, limit the ability to optimize the allocation of medical resources such as caregivers. The absence of reliable solutions which address the interoperability of different systems inside critical care units, is principally due to market concerns, since competing vendors do not embrace data sharing standards. In this work, we present a solution that integrates heterogeneous wired legacy systems within a backward compatible wireless interconnect system, providing mobility to caregivers, and the ability to coordinate and optimize their assignment to patients. The design and architecture is able to scale as needed in terms of system load and size. We demonstrate, through simulation, that the system is able to, through the optimization of caregiver assignment, significantly reduce total patient risk within health-care institutions. A prototype implementation of the system, demonstrates that the system has great promise in real-world field deployments, and can be instrumented to be compliant with site security requirements and the HIPAA privacy act.
Geometric Graph Theory and Wireless Sensor Networks
Year of Dissertation:
2012
In this work, we apply geometric and combinatorial methods to explore a variety of problems motivated by wireless sensor networks. Imagine sensors capable of communicating along straight lines except through obstacles like buildings or barriers, such that the communication network topology of the sensors is their visibility graph. Using a standard distributed algorithm, the sensors can build common knowledge of their network topology.
Optimization Problems in Sensor Network Data Collection
Year of Dissertation:
2011
Data collection is one of the most important tasks of many sensor networks. The data collected by sensors is used to monitor and analyze various systems, such as volcanoes, forests, and bridges. Large scale wireless sensor networks can provide timely access to a wealth of data, but obtaining this data is challenged by various resource constraints. This thesis proposes and analyzes solutions to three optimization problems that arise from the conflict between data collection and resource constraints: (1) maximize coverage by a set of sensors when the coverage they provide varies with location; (2) select a subset of the sensors, within some budget constraint, that best predict the data streams produced by all the sensors in the network; and (3) minimize the cost needed to find the top ranking sensor readings according to some criteria. The analyses of these problems use three different views of a sensor network: a coverage-centric view, in which each sensor is valued for its coverage ability; a data-centric view, in which each sensor is valued for the data it provides; and an agent-centric view, in which each sensor is viewed as an independent agent with information of value to the application. By choosing an appropriate view of the network, it is possible to separate the analysis from implementation details and apply well-established techniques from other domains to the problem solution. In this case, methodologies from stochastic and computational geometry, graph theory, and search theory are applied to the respective problems. This thesis presents optimal solutions to the coverage and search problems, approximation bounds on the best possible solution to the selection problem, and quantitative comparisons to alternative solutions to each problem in synthetic environments.
ON THE PROBLEM OF PACKING STEINER TREES OF A GRAPH
Year of Dissertation:
2010
On The Problem of Packing Steiner Trees of a Graph
A Symbolic Exploration of the Joint State Space and the Underlying Argumentation-based Reasoning Processes for Multiagent Planning
Year of Dissertation:
2012
Coming up with coherent behaviors for a team of agents in a non-deterministic environment is a complex process. The problem is further complicated by information regarding the environment being defeasible --- new information will disqualify the old information --- while at the same time this information is distributed, uncertain and possibly inconsistent.
3D SCENE MODELING AND UNDERSTANDING FROM IMAGE SEQUENCES
Year of Dissertation:
2013
A new method for 3D modeling is proposed, which generates a content-based 3D mosaic (CB3M) representation for long video sequences of 3D, dynamic urban scenes captured by a camera on a mobile platform. In the first phase, a set of parallel-perspective (pushbroom) mosaics with varying viewing directions is generated to capture both the 3D and dynamic aspects of the scene under the camera coverage. In the second phase, a unified patch-based stereo matching algorithm is applied to extract parametric representations of the color, structure and motion of the dynamic and/or 3D objects in urban scenes, where a lot of planar surfaces exist. Multiple pairs of stereo mosaics are used for facilitating reliable stereo matching, occlusion handling, accurate 3D reconstruction and robust moving target detection. The outcome of this phase is a CB3M representation, which is a highly compressed visual representation for a dynamic 3D scene, and has object contents of both 3D and motion information. In the third phase, a multi-layer based scene understanding algorithm is proposed, resulting in a planar surface model for higher-level object representations. Experimental results are given for both simulated and several different real video sequences of large-scale 3D scenes to show the accuracy and effectiveness of the representation. We also show the patch-based stereo matching algorithm and the CB3M representation can be generalized to 3D modeling with perspective views using either a single camera or a stereovision head on a ground mobile platform or a pedestrian. Applications of the proposed method include airborne or ground video surveillance, 3D urban scene modeling, traffic survey, transportation planning and the visual aid for perception and navigation of blind people.
Algorithms and Hypothesis Selection in Dynamic Homology Phylogenetic Analysis
Year of Dissertation:
2010
Phylogeny and alignment estimation are two important, and closely related biological problems. In the typical alignment problem, insertions, deletions, and substitutions need to be inferred, to understand the evolutionary patterns of life. With the technological advances of the last 20 years, phylogenetic analyses will grow to include complete chromosomes and genomes. With these data sets, not only insertions, deletions, and substitutions, but also rearrangements such as duplications, translocations, transpositions, and inversions must be taken into consideration.
Timed Modal Epistemic Logic
Year of Dissertation:
2012
There will be three parts in this thesis. The first part is a survey of epistemic logic. Epistemic logic was first introduced by philosophers, and later found its applications in fields such as Computer Science and Economics. The survey will cover both the philosophical debates and applications of epistemic logic, and then discussions of the logical omniscience problem will follow.
AN ADAPTIVE AND INTEGRATED MULTIMODAL SENSING AND PROCESSING FRAMEWORK FOR LONG RANGE MOVING OBJECT DETECTION AND CLASSIFICATION
Year of Dissertation:
2013
In applications such as surveillance, inspection and traffic monitoring, long-range detection and classification of targets (vehicles, humans, etc) is a highly desired feature for a sensing system. A single modality will no longer provide the required performance due to the challenges in detection and classification with low resolutions, noisy sensor signals, and various environmental factors due to large sensing distances. Multimodal sensing and processing, on the other hand, can provide complementary information from heterogeneous sensor modalities, such as audio, visual and range sensors. However, there is a lack of effective sensing mechanisms and systematic approaches for sensing and processing using multimodalities. In this thesis, we described a systematical framework for Adaptive and Integrated Multimodal Sensing and Processing (thereafter, the AIM-SP framework) that integrates novel multimodal long-range sensors, adaptive feature selection and learning-based object detection and classification for achieving the goal of adaptive and integrated multimodal sensing and processing. Based on the AIM-SP framework, we have made three unique contributions. First, we have designed a novel multimodal sensor system called Vision-Aided Automated Vibrometry (VAAV), which is capable of automatically obtaining visual, range and acoustic signatures for moving object detection at a large distance. Second, multimodal data, acquired from multiple sensing sources, are integrated and represented in a Multimodal Temporal Panorama (MTP) for easy alignment and fast labeling. Accuracy of target detection can be improved using multimodalities. Further, a visual reconstruction method is developed to remove occlusions, motion blurs and perspective distortions of moving vehicles. With various types of features extracted on aligned multimodal samples, we made our third contribution on feature modality selection using two approaches. The first approach uses multi-branch sequential-based feature searching (MBSF) and the second one uses boosting-based feature learning (BBFL).
OPTIMIZATION ALGORITHMS FOR PROXY PLACEMENT IN CONTENT DISTRIBUTION NETWORKS
Year of Dissertation:
2011
Advisor:
Kaliappa Ravindran
Popular web sites receive an enormous share of Internet traffic. These sites have a competitive motivation to offer better service to their clients at lower cost. One of the solutions is to using content distribution network (CDN). When we optimize proxy placement in content distribution network we want to maximize the user experience in the mean time to minimize the resource consumption. We are dealing with a multi-objective optimization problem for CDNs in the thesis: the content latency experienced by users and the system resources expended to deliver content. After the topology model for content distribution network, we listed metrics to evaluate content distribution network. Then we defined the objective function for optimizing proxy placement. After reviewing the existing proxy placement algorithms we used genetic algorithm to solve the proxy placement problem. We first apply the genetic algorithms to a simple network, and find that the results are better than greedy algorithm and identical to the global optimal. Then we apply the genetic algorithm to a realistically larger network topology. The genetic algorithm can find good solutions to the proxy placement problem. But compared to greedy algorithm, it is less efficient. So when response time is important, genetic algorithm is not the best choice. To optimize the genetic algorithm we found if we choose higher mutation rate and step size at the beginning of the genetic operation and decrease the mutation rate and step size as we go through the genetic operation, the results are much better. For the initial populations, if we want quick response to the sudden change of client access we can choose smaller initial population, if we have sufficient response time especially in initial setting up the network we should choose larger initial populations. Under optimized genetic algorithm, we experiment three different network examples with their sizes between the simple network and large network. Since the network sizes are smaller we are able to run the exhaustive search to find the global optimal. In all cases the results from genetic algorithm outperform greed algorithm and close to global optimal. Although it is not guaranteed to find the global optimal solution as the simple network did, a robust good solution is enough to solve real world problems.