Approved for public release, distribution is unlimited / Inter-domain routing connects individual pieces of Internet topology, creating an integral, global data delivery infrastructure. Currently, this critical function is performed by the Border Gateway Protocol (BGP) version 4 [RFC1771]. Like all routing protocols, BGP is vulnerable to instabilities that reduce its effectiveness. Among the causes of these instabilities are those which are maliciously induced. Although there are other causes, e.g., natural events and network anomalies, this thesis will focus exclusively on maliciously induced instabilities. Most current models that attempt to predict a BGP routing instability confine their focus to either macro- or micro-level metrics, but not to both. The inherent limitations of each of these forms of metric gives rise to an excessive rate of spurious alerts, both false positives and false negatives. It is the original intent of this thesis to develop an improved BGP instability prediction model by statistically combining BGP instability metrics with user level performance metrics. The motivation for such a model is twofold. 1) To provide sufficient prior warning of impending failure to facilitate proactive protection measures. 2) To improve warning reliability beyond existing models, by demonstrably reducing both false positives and false negatives. However, our analysis of actual network trace data shows that a widely used BGP instability metric, the total number of update messages received in a time period, is not a good indicator of future user level performance. / Civilian, Department of Defense

Résumé : Ce projet de recherche s'inscrit dans une démarche d'éco-conception de matériaux architecturés à fort potentiel acoustique. On s'intéresse en particulier dans ce travail de thèse à la description des phénomènes de dissipation et de propagation des ondes acoustiques dans un milieu fibreux par une approche multi-échelle et multi-physique. Dans cette étude, il s'agit de décrire les propriétés acoustiques du milieu fibreux à partir d'une description de la géométrie à l'échelle locale de milieux fibreux. Le milieu fibreux est constitué de fibres d'asclépiades, qui sont typiquement des fibres végétales adoptant la forme de longs cylindres. Pour traiter ce problème, la méthodologie employée consiste en quatre étapes principales : (i) caractérisation et modélisation de la géométrie du milieu fibreux; (ii) calcul des paramètres de transport et acoustiques du milieu fibreux reconstruit; (iii) validation expérimentale de propriétés de transport et acoustiques; (iv) évaluation des évolutions de propriétés de transport en fonction des paramètres de la géométrie à l'échelle locale de matériau. En particulier, un modèle isotrope spatialement stationnaire de lignes droites (processus de Poisson) et le tenseur d'orientation angulaire correspondant, constituent des outils de modélisation de la géométrie aléatoire du milieu fibreux permettant de représenter les principales caractéristiques susceptibles d'influencer ses propriétés de transport. Il s'agit ensuite de résoudre les principaux problèmes aux limites gouvernant le comportement acoustique à l'échelle supérieure en appliquant une technique classique d'homogénéisation numérique. On montre dans un premier temps que la méthode développée permet de prédire le comportement en absorption d'un milieu fibreux aléatoire en se basant uniquement sur la description des caractéristiques géométriques du matériau fibreux réel (porosité, rayon de fibres, distributions des orientations angulaires) sans coefficient d'ajustement, validations expérimentales à l'appui. Sur la base de ce travail de reconstruction tridimensionnel, on étudie ensuite systématiquement l'ensemble des phénomènes de transport d'intérêt sur une large gamme de porosité et d'orientations angulaires, de manière à produire des lois qui peuvent être appliquées par la suite par d'autres utilisateurs sur une large gamme de milieux fibreux réels. Finalement, on examine plus particulièrement l'hypothèse classique selon laquelle un milieu fibreux peut être décrit à partir de la moyenne arithmétique du diamètre de ses fibres, afin de cerner les limites de cette approche et ses conditions d'applicabilité lorsque le milieu fibreux présente une distribution étendue de diamètres de fibres ou bimodale. / Abstract : This thesis is concerned with in an eco-design approach for architectured porous materials (fibers made) with high acoustic potential. The project particularly focuses on asclepias (vegetal) fibers and on the description of dissipation and propagation phenomena of sound waves using homogellization techniques. One begins with the characterization of the fibrous medium by scanning electron microscope images from horizontal and vertical cross-sections of a slab of porous sample. Three-dimensional unit cells of the fibrous samples under study are reconstructed with an isotropic model of straight lines (Poisson processes) and by making use of the concept of angular orientation tensor from the previously identified microstructure characteristics (porosity, fiber radii, angular orientation distributions). The transports and acoustic properties are obtamed from numerical computations of unit cell problems (Stokes flow, potential flow and heat conduction) with the Finite Element Method. Validations with experimental data based on permeability and impedance tube measurements are proposed and show a good agreement with the predictive models. The dependence of the effective properties with the geometrical characteristics of the porous microstructure is then analyzed on a large range of porosity (0.75 ÷ 0.09) and explicit relations are provided between the effective acoustic coefficients and the microstructural parameters. The last part of this work is dedicated to the determination of acoustic properties of random fibrous media with bi- and poly dispersed fiber distribution radii. The results are compared with a corresponding mono- dispersed fibrous material (with a single effective fiber radius) which show no significant difference with the initial distributions when the effective radius is small enough. The results of this thesis also point out that an equivalent mono-dispersed fibrous material fails to represent accurately the transport properties of a random fibrous structure when the latter one is described by a fiber distribution radii tending towards a log normal distribution.

This thesis considers how fiscal and monetary policy should be conducted in resourcerich economies. It consists of three papers addressing: whether governments should spend, save or invest volatile oil income; the assets they should save in; and how monetary policy should respond. The first, “Eight principles for managing resource wealth”, shows that capital-scarce countries should save relatively less against oil price volatility, and invest more in domestic capital. They also should prepare for volatility in advance, and treat savings as a source of income rather than a temporary buffer. To show this the paper develops a framework that nests a variety of existing results, which are presented in eight principles. The second, “The Elephant in the Ground: Oil extraction and asset allocation in sovereign wealth funds”, shows that governments should use sovereign wealth funds to offset oil price risk, extract oil faster if its price is pro-cyclical, and use precautionary savings to manage any residual volatility. To do this it combines three strands of literature for the first time: on continuous-time portfolio theory, oil extraction and precautionary savings. The third, “Optimal monetary responses to oil discoveries”, addresses the anticipation effects around an oil discovery. It shows that the terms of trade will need to appreciate twice: once when oil is discovered and consumers anticipate future revenues; and again when the government begins spending the revenues. Oil wealth will give the monetary authority an incentive to appreciate the terms of trade, in addition to stabilising domestic inflation and the output gap. Optimal policy is well-approximated by a standard monetary rule that also responds to expected changes in the natural level of output.

The objective of this thesis is to examine real exchange rate volatility, with a particular focus on investigating the causes of exchange rate jumps. While the predominant approach in the literature is to examine the interaction between nominal rigidities and nominal shocks, this thesis examines the volatility that arises from real rigidities and shocks. Trying to better understand the transmission of real shocks to the exchange rate is a worthwhile task, given the substantial evidence that these shocks and rigidities are important for explaining other economic fluctuations. This thesis develops theoretical models that examine the contributions of specific real rigidities to exchange rate volatility. Chapter 1 introduces our baseline specification - a frictionless model, with the exception of capital adjustment costs. This baseline generates very mild exchange rate fluctuations. Additional rigidities are required to generate volatility of the magnitude that is typically observed. Chapter 2 finds that introducing imperfect asset substitutability - specifically, home asset bias - goes a little towards achieving this. When investors are biased, the exchange rate must adjust by more to equilibrate asset markets. This greater burden of adjustment on the exchange rate along the short run path typically translates to larger jumps after shocks. Similarly, Chapter 3 shows that augmenting the baseline with banks and financial frictions raises exchange rate volatility. The key point is that, in the presence of financial frictions, there is a risk premium that widens after negative shocks, increasing the required adjustment of the exchange rate. A fourth chapter extends Chapter 3 and shows that unconventional credit policy, while beneficial in some respects, nonetheless entails nontrivial costs because it invites moral hazard by encouraging banks to be more highly leveraged, which increases exchange rate and consumption volatility. So, the overall message is that, in the presence of plausible real frictions - including (i) capital adjustment costs, (ii) imperfect asset substitutability, and (iii) financial frictions - real shocks can generate a plausibly significant degree of real exchange rate volatility. This thus posits an additional explanation of exchange rate jumps that complements the predominantly monetary literature.

This thesis argues that state-building in Sudan in the modern era cannot be understood without a multilevel analysis of the links between water, civilisation and power. More particularly, it focuses on the hydropolitical economy of the Al-Ingaz Revolution since its launch in 1989. I analyse the efforts by Sudan's military-Islamist leaders at material and immaterial transformation of society through visions of hydro-engineering civilisation. “Economic Salvation” -the rescue of Sudan’s economy through a “hydro-agricultural mission” that will create an ‘Islamic’ middle class- is central to this ideology. The hydro-agricultural mission is a revolutionary attempt at Islamist state-building through a hyper-ambitious Dam Programme and an Agricultural Revival in Sudan’s riverain core. It intends to entrench Al-Ingaz in power by delivering for those riverain constituencies and external partners on the Arabian Peninsula and in East Asia deemed critical to continued hegemony. This thesis is fundamentally about Islamist Sudan's hydropolitical economy, but makes broader contributions. First, it highlights how, far from being exceptional, the hydro-agricultural mission is deeply embedded in historical ways of thinking about water, civilisation and power in Sudan and the Nile Basin more broadly, echoing assumptions, policy prescriptions and logics of political control and high-modernist development that have been salient for almost 200 years. In the past, grand state-building projects, predicated on the dream of controlling the water to control the people, have been characterised by high levels of violence and developmental mirages in the desert. I show why, under military-Islamist rule, this experience is being repeated in Sudan. Second, this thesis is situated in wider debates in the early 21st century, with fears about resources crunches proliferating amidst rising global commodity prices and the impact of climate change. The idea that environmental scarcity, as an exogenous variable, is the main shaper of societies and their politics is enduring, but both theoretically and empirically misguided. Moreover, it has often been manipulated by elites in processes of power and wealth accumulation that reproduce the very societal and ecological problems they claim to be resolving. I argue that the links between water, civilisation and power in Sudan highlight not just the endogeneity of environmental scarcity to political-economic processes, but also the violent consequences of a modernist paradigm that is seen by ruling elites as both enlightened science and the route to hegemony while reproducing conflict at the local, national and regional level.

This thesis deals with stress testing as a process that helps to assess the impact of potential adverse shocks on the soundness of a financial system. First section is dedicated to non-technical discussion about stress testing and to some methodological issues. The main focus lies on the system-wide macroeconomic stress testing. The empirical part of the thesis is a contribution to macroprudential analysis of the quality of the aggregate loan portfolio in the Czech Republic. This study adopts a vector autoregression model applied to the Czech banking sector in order to judge its stability and present some evidence on macroeconomic variables affecting the Czech banking system. As a measure of the strength of the loan portfolio is used the stock of non-performing loans vis-à-vis total loans in the sector. The thesis follows the widely used methodology and seeks to identify significant macroeconomic risk factors affecting the loan portfolio quality. The latter part aims also to forecast the most likely development of the loan portfolio.

Commuting became an inseparable part of everyday life of contemporary society and an analysis of commuting flows has a fundamental meaning for capturing social- geographical organization of society. The main object of this thesis is to identify changes in geographical aspects of commuting to work and school in the Benešov district during the period 2001-2011. The second object is to evaluate the impact of macroregional centre of Prague on the commuting in Benešov district. The district is, taking into consideration the nearness and a good traffic connesction to Prague, appropriate example for assessing the impact of macroregional centre on neighbouring micro-regions. This impact is followed by an analysis of commuting flows and also by a questionnare survey. The purpose of the questionnare survey is to capture how people feel about commuting to work and labour market in Prague. Keywords: commuting, socio-geographical regionalization, macro-region, micro- region, functional region, Benesov, Prague, Vlasim, Votice.

This master thesis analyzes the question, how norm authority constitutes itself in the context of school education. It is based on firsthand data from several adio- and audiovisual recordings of classes in two German schools (trade school and high school). The project provides a description of the mechanism by which the teacher establishes himself as the norm-authority and how he is able to assert this authority towards the student, the norm subject. At the same time, the reaction of the norm subjects towards the intervention of the norm authority into their language production is analyzed. Theoretically, the essay is embedded in the social force field of a standard-variety ("soziales Kräftefeld einer Standardvarietät") by Ulrich Ammon and the language management theory. By following the steps of the language management process, the handling of the language norm is described and an analysis is made of how language norms are actively adopted into the learning process. The analysis and evaluation of the class recordings underline the dynamic character of language norms and show the procedural nature of norm formation and their adoption. Key Words: norm, variety, social force field, language management, macro and micro level, norm authority, norm subject

This diploma thesis is tasked to evaluate the company Kordárna Inc. with the view of its sale within the frame of reorganization as a nonliquidating solution to its insolvency. The evaluation of the company Kordárna Inc. is performed to the date of January 1st, 2009. By then the company, due to many unfavorable conditions, falls into existential problems that have to be immediately solved. The reorganization plan is allowed and processed within the framework of higher creditors' satisfaction. An integral part of this plan is also the desirable evaluation of the company Kordárna Inc. with the view of its sale to a new investor. Within these issues, both the detailed external and internal analysis of the company and the financial analysis of the 5 preceding fiscal periods are processed (years 2004-2008). Consequently, based on many assumptions, the financial plan for the 4 following fiscal periods (2009-2012) is conducted, and the evaluation of the company Kordárna using 4 evaluation methods -- FCFF, FCFE, EVA and Liquidation value methods -- is processed.

La photo-catalyse est une voie très prisée pour la dépollution de l’eau ou de l’air. Le photo-catalyseur le plus employé est le dioxyde de titane (TiO2) mais l’activité photo-catalytique peut fortement varier d’une poudre à l’autre. De plus, il est très avantageux, voire nécessaire, de le déposer sur un support, pour une manipulation aisée, notamment s’il s’agit d’un monolithe poreux. Le but de cette thèse, menée en collaboration entre le LMCPA et l’IEM, est de produire un photo-catalyseur supporté : une mousse céramique, issue de précurseurs céramiques polymériques (voie PDC), revêtue de TiO2, tout en visant une cristallisation et une intégration du TiO2 dans la mousse, en une seule étape, par voie hydrothermale. Nous avons étudié la synthèse de TiO2 par voie hydrothermale formant ainsi des poudres de diverses natures (anatase, brookite, oxyde hydraté de titane et mélanges de ces phases) ; certaines présentent une activité photo-catalytique supérieure, dans certaines conditions, à celle de la référence commerciale le P25 de Degussa. Cette étude a également permis de produire de la brookite pure à une température relativement basse, bien plus performante que le P25 et l’anatase dans certaines conditions. Parallèlement, divers polymères précéramiques ont été investigués pour la confection de supports macro-poreux via la méthode des charges sacrificielles. Nous avons ainsi réalisé des mousses en céramique amorphe en SiBCN et SiBC, hautement poreuses, ouvertes et robustes. Enfin, les poudres de TiO2 les plus efficaces ont été déposées sur les mousses, par la voie hydrothermale, pour former le photocatalyseur supporté dont l’activité photo-catalytique a été évaluée. / Photocatalysis is a method of choice for water and air depollution. Titanium dioxide (TiO2) is the most used photocatalyst but photocatalytic activity can widely differ from one powder to another. Moreover, it is useful, or even necessary, to immobilize it on a support; in particular, a porous monolithic support for an easier handling. The purpose of this PhD thesis, consisting in a collaboration between the LMCPA and the IEM, is to produce a supported photocatalyst: an amorphous ceramic foam produced from polymeric ceramic precursors (PDC route) coated with TiO2. The objective is also to target a one pot process the crystallization and the incorporation of the photocatalyst, in a single stage, through a hydrothermal treatment. We studied the hydrothermal synthesis of TiO2 powders composed of various crystalline phases (anatase, brookite, titanium oxide hydrate and mixtures of thereof); some of them showed higher photocatalytic activity than the commercial reference, Degussa’s P25. This study also lead to the synthesis of pure brookite, at a relatively low temperature, much more efficient than P25 and anatase under certain conditions. At the same time, various preceramic polymers were studied for the production of macroporous supports through the sacrificial filler technique. We were able to produce highly porous, opened and robust amorphous ceramic foams in the Si-B-C-N and Si-B-C systems. Lastly, the foams were coated with the most efficient TiO2 powders in order to produce supported photocatalysts by the hydrothermal route; their photocatalytic activity was then evaluated.