Thesis (M.S.)--Brigham Young University. Dept. of Computer Science, 2006. / Includes bibliographical references (p. 45-48).

Zugl.: Paderborn, Universiẗat, Diss., 2004.

Thesis (Ph. D.) University of Missouri-Columbia, 2006. / The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file (viewed on August 3, 2007) Vita. Includes bibliographical references.

Presentation version of MIT M.S. EECS/TP P Thesis, May 1997

Internet 0 is proposed as a local area network that supports extremely small network devices with very little capacity for computation, storage, or communication. Internet 0 addresses the issue of connecting very small, inexpensive devices such as lightbulbs and heating vents with their controllers. To achieve this effectively, Internet 0 assumes both that operating between communicating end-nodes should not require third-party support, and that IP will be available all the way to those end-nodes. Several simplifying assumptions are made in Internet 0 to achieve this. The objective of this paper is to explore issues of design in a context where federation of an Internet 0 net either with other Internet 0 nets or the global Internet becomes important. The question we ask is whether the end-node in such an Internet 0 needs to know more or behave differently in such a federated environment, and how one might achieve such federation. We explore three aspects of network design in this study: addressing and routing, traffic collision and congestion control, and security. In each case, based on analysis, we conclude that to reach our goals in a generalizable and extensible fashion, a third party service will be needed to act as an intermediary, and propose that a single service should provide all the required federation services. / Singapore-MIT Alliance (SMA)

Due to the high measuring cost, the monitoring of power quality is non-trivial. This work is aimed at reducing the cost of power quality monitoring in power networks. Using a real-world power quality dataset, this work adopts a learn-from-data approach to obtain a device latent feature model, which captures the device behavior as a power quality transition function. With the latent feature model, the power network could be modeled, in analogy, as a data-driven network, which presents the opportunity to use the well-investigated network monitoring and data estimation algorithms to solve the network quality monitoring problem in power grid. Based on this network model, algorithms are proposed to: 1) intelligently place measurement devices on suitable power links to reduce the uncertainty of power quality estimation on unmonitored power links, 2) estimate the power quality in unmonitored segments of a power network, using only a small number of measurement points, and 3) identify a potential malfunction device in the network. The meter placement algorithms use entropy-based measurements and Bayesian network models to identify the most suitable power links for power quality meter placement. Evaluation results on various simulated networks including IEEE distribution test feeder system show that the meter placement solution is efficient, and has the potential to significantly reduce the uncertainty of power quality values on unmonitored power links. After deploying power quality meters on selected links, a MaxEnt-based approach is presented to estimate the power quality on the unmonitored lines. Compared to other existing methods such as MCEM, the MaxEnt-based approach is much faster while maintaining similar estimation accuracy. Convergence time of the MaxEnt algorithm is particularly important when the network size increases and we need to do the estimation in real time. Finally, using readings from our metered locations, we propose a prediction model that derives an acceptable device behavior to identify a potential malfunction device in the power grid. Simulation results show that our predictive model accurately detects the malfunction devices in the power network and can be used to make proper recommendations of device maintenance and replacement. / Graduate

The global aim of this dissertation is to develop advanced statistical modeling to understand the genetic basis of complex polygenic traits. In order to achieve this goal, this dissertation focuses on the development of (i) a novel methodology to detect genetic variants with different inheritance patterns formulated as a Bayesian model selection problem, (ii) integration of genetic data and non-genetic data to dissect the genotype-phenotype associations using Bayesian networks with family-based data, and (iii) an efficient technique to model the family-based data in the Bayesian framework. In the first part of my dissertation, I present a coherent Bayesian framework for selection of the most likely model from the five genetic models (genotypic, additive, dominant, co-dominant, and recessive) used in genetic association studies. The approach uses a polynomial parameterization of genetic data to simultaneously fit the five models and save computations. I provide a closed-form expression of the marginal likelihood for normally distributed data, and evaluate the performance of the proposed method and existing methods through simulated and real genome-wide data sets. The second part of this dissertation presents an integrative analytic approach that utilizes Bayesian networks to represent the complex probabilistic dependency structure among many variables from family-based data. I propose a parameterization that extends mixed effects regression models to Bayesian networks by using random effects as additional nodes of the networks to model the between-subjects correlations. I also present results of simulation studies to compare different model selection metrics for mixed models that can be used for learning BNs from correlated data and application of this methodology to real data from a large family-based study. In the third part of this dissertation, I describe an efficient way to account for family structure in Bayesian inference Using Gibbs Sampling (BUGS). In linear mixed models, a random effects vector has a variance-covariance matrix whose dimension is as large as the sample size. However, a direct handling of this multivariate normal distribution is not computationally feasible in BUGS. Therefore, I propose a decomposition of this multivariate normal distribution into univariate normal distributions using singular value decomposition, and implementation in BUGS is presented.

Recent advances in biology have revolutionized our understanding of living systems. For the first time, it is possible to study the behavior of individual cells. This has led to the discovery of many amazing phenomena. For example, cells have developed intelligent mechanisms for foraging, communicating, and responding to environmental changes. These diverse functions in cells are controlled through biochemical networks consisting of many different proteins and signaling molecules. These molecules interact and affect gene expression, which in turn affects protein production. This results in a complex mesh of feedback and feedforward interactions. These complex networks are generally highly nonlinear and stochastic, making them difficult to study quantitatively. Recent studies have shown that biochemical networks are also highly modular, meaning that different parts of the network do not interact strongly with each other. These modules tend to be conserved across species and serve specific biological functions. However, detect- ing modules and identifying their function tends to be a very difficult task. To overcome some of these complexities, I present an alternative modeling approach that builds quantitative models using coarse-grained biological processes. These coarse-grained models are often stochastic (probabilistic) and highly non-linear. In this thesis, I focus on modeling biochemical networks in two distinct biological systems: Dictyostelium discoideum and microRNAs. Chapters 2 and 3 focus on cellular communication in the social amoebae Dictyostelium discoideum. Using universality, I propose a stochastic nonlinear model that describes the behavior of individual cells and cellular populations. In chapter 4 I study the interaction between messenger RNAs and noncoding RNAs, using Langevin equations.

Dysfunction of cholinergic and dopaminergic systems has been implicated in memory function de cits that are core pathology and associated features of several neurological disorders. However, in order to develop more effective treatments, it is crucial to better understand how different aspects of learning and memory are modulated by these neuromodulatory systems. Using optogenetic stimulation or silencing, this thesis aims to investigate the effects of cholinergic and dopaminergic modulation on various hippocamal learning and memory processes. To understand how these neuromodulatory systems modulate hippocampal network activity, I first examined their effects on hippocampal local field potentials in urethane-anaesthetised mice. I demonstrated that optogenetic cholinergic activation suppressed slow oscillations, shifting brain activity to a state dominated by theta and gamma oscillations. In contrast, dopaminergic activation suppressed gamma oscillations. Second, to directly probe the effects of neuromodulation on different stages of spatial learning, I acutely activated or inactivated cholinergic or dopaminergic neurons during various behavioural tasks. My findings suggested that cholinergic activation, solely during the reward phase of a long-term spatial memory task, slowed place learning, highlighting the importance of temporally-precise neuromodulation. Moreover, dopaminergic stimulation may enhance place learning of a food rewarded task, supporting a role for dopamine in spatial learning. In addition, I tested the effects of cholinergic and dopaminergic modulation on reversal learning and found that cholinergic inactivation and dopaminergic activation appear to impair this process. Together, these findings emphasise the importance of cholinergic and dopaminergic modulation in learning and memory. They suggest that precise timing of neuromodulator action is critical for optimal learning and memory performance, and that acetylcholine and dopamine support complementary processes that allow for effective learning and adaptation to changing environments.

In this thesis I contribute to the growing literature on the structure of covert networks by exploring the organisation and functioning of two new groups. (1) The Right Club, a Right-wing, Pro-German group active in the UK at the outbreak of World War Two, and (2) The leadership group of the Provisional Irish Republican Army (PIRA) between 1969 and 1986. Specifically, I focus upon the formation of these groups, and how, and indeed if, they maintained covertness in practice. Whilst there has been a wealth of research in this area, many studies simply assume covertness and its impact upon structure due to the illegal nature of their case studies. In this thesis I develop a more nuanced concept of covertness, and a more detailed analysis of the myriad factors which affect the structure of a clandestine group. I employ a mixed methods approach combining Social Network Analysis with qualitative inquiry of the environment and processes which influence the functioning of each group. The qualitative analysis, which was guided by factors identified in the existing covert networks literature, in the Social Movements literature, and by dynamics noted in work on the Sociology of Secrecy, is used to explore and explain the sociometric findings. This provides a more in-depth, more sociological understanding of clandestine organisation than that which currently exists in this field of research. However, more and varied case studies analysed in this way are also necessary if we are to improve our understanding of the structure and functioning of covert groups. With this knowledge more sensitive and successful deradicalisation and/or destabilisation techniques can be crafted.