The evolution of Mobile Internet has led to the growth of bandwidth demanding applications like video streaming and social networking. The required data rates projected for such applications cannot be sustained by current cellular networks. New network architectures like Long Term Evolution (LTE) and LTE Advanced have been carefully engineered and introduced to fulfill such large data rates. The recent introduction of femtocells enabled high data rates and better coverage indoors, without the need for site establishment or upgrading the network infrastructure. Femtocells, however, will potentially suffer from major interference problems due to their expected dense and ad hoc deployment. The main contribution in this thesis is the introduction of a new and a very promising direction in deriving capable and efficient interference mitigation schemes, and comparing this direction to current techniques in the literature. Several works have studied the effect of interference on networks employing femtocells. In this thesis, we also survey such works and provide an overview of the elements considered in mitigating interference. We introduce a new scheme known for its optimality, and use it for frequency assignment in downlink femtocell networks. The algorithm is based on optimization search rather than greedy or heuristic methods. Experimental simulations will be shown to evaluate the proposed scheme against other schemes from the literature. / Thesis (Master, Computing) -- Queen's University, 2012-03-31 02:14:28.549

This study investigates the principles of interpretation in Tanwir al-Miqbas min Tafsir Ibn `Abbas, focusing on its approach and method, sources and points of view. To accomplish this purpose, both historical and literary analysis are employed. Nevertheless, this study does not attempt to resolve the well-known dispute over its authenticity or ascription. / Like other Qur'anic commentaries produced in the classical period, Tanwir al-Miqbas relies heavily on philological analysis in the sense that it provides an interpretation of Qur'anic verses based on their literal meaning and in very simple terms. Similarly, it employs sources that were commonly used in classical tafsir, namely, Qur'anic cross-reference, h&dotbelow;adith, Arabic poetry and isra'iliyat. However, due to its simplistic approach, it is difficult to extract any guiding principles behind its interpretation. In fact, there is little to indicate that this Qur'anic exegesis represents the views of any particular Islamic sect, let alone of any individual person. / Nevertheless, with its simple presentation, Tanwir al-Miqbas brings up many interesting and challenging issues which deserve greater scholarly attention. Further and deeper research into these issues may help us to situate the work within the broader exegetical tradition in Islam.

On appelle suite récurrente stochastique (SRS) dirigée par une suite de matrices aléatoires une suite de variables aléatoires telles que le terme de rang n+1 est obtenu en multipliant celui de rang n par la enième matrice. Cette thèse porte sur le comportement asymptotique de telles suites. Dans la première partie, les matrices sont inversibles et on donne un critère de séparation des exposants de Lyapunov quand la suite de matrices suit une mesure de Gibbs sur un sous-shift de type fini. Dans la seconde partie, les produits se font au sens max-plus. On montre que le comportement des SRS au premier ordre est essentiellement déterminé par celui de certains blocs diagonaux et que la propriété de perte de mémoire, qui assure la stabilité des SRS, est générique. Si une suite de matrices (ou d'applications topicales) aléatoires est i.i.d. et a la propriété de perte de mémoire, alors les SRS qu'elle dirige vérifient des théorèmes limites. Ce résultat est obtenu par la méthode du trou spectral.

Med utgångspunkt i genusteori, intersektionalitet och litteraturteori analyserar denna studie två elevvalda böcker från ett bokprojekt i en gymnasieskola; Hungerspelen av Suzanne Collins och Jag är Zlatan Ibrahimović: min historia av Zlatan Ibrahimović och David Lagercrantz, varav en bok är flickvald och en är pojkvald. Böckerna analyseras utifrån genusperspektiv i förhållande till traditionella könsstereotyper, såsom hur de kvinnliga och manliga karaktärerna beskrivs, hur de talar, hur de påverkas av andra aspekter såsom klass och etnicitet samt hur fördelningen av män och kvinnor ser ut. Utöver nykritikens närläsning av litteraturen analyseras de sammanställda listorna med boktitlar som eleverna har valt, för att urskilja potentiella mönster i förhållande till genus, vilket ställs mot de böcker som i en intervju med läraren bakom bokprojektet framkommer att hon skulle välja att läsa i sin klass. Genrer och författare för de elevvalda böckerna analyseras utifrån befintlig forskning för att kunna skönja om det överensstämmer med forskningen, samt huruvida den litteratur som presenteras av läraren stämmer överens med de preferensramar som finns för flickor och pojkars läsning. Resultatet av litteraturanalysen är att huvudpersonerna i böckerna ständigt gör genus, feminitet och maskulinitet, för att uppnå olika mål och till och med överleva. Zlatan vill framstå som tuff och imponera på andra män medan Katniss måste följa de kvinnliga skönhetsidealen i huvudstaden för att få sponsorer. I båda böckerna är karaktärerna i majoritet män, vilket i Zlatans fall beror på hans karriär som är mansdominerad och följer mönstret för medelklassmannen som skapar sin manlighet i en homosocial sfär. I Hungerspelen är de viktigaste personerna som omger Katniss män, men den är i övrigt mer uppblandad och traditionella könskodade yrken är inte lika könskodade. Båda böckerna problematiserar klass och det är stor skillnad på män och kvinnor från arbetarklass och överklass. Både Katniss och Zlatan gör en klassresa igenom boken. När det kommer till elevernas läsning är det endast en av fyra lärarvalda böcker som passar pojkars läsning, övriga tre stämmer bättre överens med flickors val av litteratur. De böcker som valts av eleverna har en jämn spridning av kvinnliga och manliga författare. Pojkar föredrar att läsa deckare, biografier och fantasy, medan flickor hellre läser böcker som handlar om kärlek och romantik.

This paper studies if the Minimax theorem holds for the behaviour of trained and untrained actors in the field. This is explored with data from 1043 football penalty kicks from professionals of the German Bundesliga and for 268 penalty kicks from untrained players. Minimax makes good predictions about the collective patterns emerging from the behaviour of experienced actors, as well as about their individual strategic actions. However, this is not true for untrained actors. In the next step it is explored if, the professional players learned their behaviour, or if they were selected into their roles because they had the required abilities. The data suggests that the professionals were selected by the competitive conditions of professional sports.

This thesis presents a comparative study of two modern Muslim biographers of the Prophet of Islam, Muhammad Husayn Haykal and Sayyid Ja$ rm sp{c}$far Murtada, the one a Sunni the other a Shispc$ = i. In dealing with their works on the { it s = i ra}, the intention is to focus on some aspects of the life of the Prophet of Islam from the time when he resided in Makka. After outlining the authors' scholarly backgrounds, sources, methods, and purposes, etc., the thesis deals with certain specific issues concerning the Prophet's birth, his nursing and the reason behind the custom of having foster-mothers, the controversial stories of the splitting of his chest and of his participation in the sacrilegious war, his state both before and after the announcement to him of his Mission, the authors' interpretations of his night journey and ascent to heaven and lastly their discussion about the Prophet and miracles. This research should not be considered as historical in nature; rather, it aims at clarifying the similarities and differences which may be discovered between our two biographers, and at comparing their respective Sunn = i and Sh = i$spc$ = i approaches.

We consider the problem of learning the parameters of a structured output prediction model, that is, learning to predict elements of a complex interdependent output space that correspond to a given input. Unlike many of the existing approaches, we focus on the weakly supervised setting, where most (or all) of the training samples have only been partially annotated. Given such a weakly supervised dataset, our goal is to estimate accurate parameters of the model by minimizing the regularized empirical risk, where the risk is measured by a user-specified loss function. This task has previously been addressed by the well-known latent support vector machine (latent SVM) framework. We argue that, while latent SVM offers a computational efficient solution to loss-based weakly supervised learning, it suffers from the following three drawbacks: (i) the optimization problem corresponding to latent SVM is a difference-of-convex program, which is non-convex, and hence susceptible to bad local minimum solutions; (ii) the prediction rule of latent SVM only relies on the most likely value of the latent variables, and not the uncertainty in the latent variable values; and (iii) the loss function used to measure the risk is restricted to be independent of true (unknown) value of the latent variables. We address the the aforementioned drawbacks using three novel contributions. First, inspired by human learning, we design an automatic self-paced learning algorithm for latent SVM, which builds on the intuition that the learner should be presented in the training samples in a meaningful order that facilitates learning: starting frome easy samples and gradually moving to harder samples. Our algorithm simultaneously selects the easy samples and updates the parameters at each iteration by solving a biconvex optimization problem. Second, we propose a new family of LVMs called max-margin min-entropy (M3E) models, which includes latent SVM as a special case. Given an input, an M3E model predicts the output with the smallest corresponding Renyi entropy of generalized distribution, which relies not only on the probability of the output but also the uncertainty of the latent variable values. Third, we propose a novel learning framework for learning with general loss functions that may depend on the latent variables. Specifically, our framework simultaneously estimates two distributions: (i) a conditional distribution to model the uncertainty of the latent variables for a given input-output pair; and (ii) a delta distribution to predict the output and the latent variables for a given input. During learning, we encourage agreement between the two distributions by minimizing a loss-based dissimilarity coefficient. We demonstrate the efficacy of our contributions on standard machine learning applications using publicly available datasets.

Thesis (Ph. D.)--University of Washington, 1998. / Vita. Includes bibliographical references (leaves [584]-599).

Made available in DSpace on 2014-12-17T14:07:03Z (GMT). No. of bitstreams: 1 LeonardoFM.pdf: 4456392 bytes, checksum: 36e6d8718ecd158955c24006e7afca9a (MD5) Previous issue date: 2009-10-30 / The mining industry is responsible for the generation of waste from their natural process of extraction. The mining impacts in urban areas are of special importance due to the high urban occupation, which are exacerbated due to the proximity of the mined areas and populated areas. Some solutions to wastedisposal have the potential to significantly reduce the environmental risks and liabilities, but represent higher costs in the stages of deployment and operation. The addition of mining waste as raw material in the development of commercial products reduces the environmental impacts, transforming the waste into a positive element in the generation of employment and income. This thesis studies the incorporation of waste iron ore in two clays, one from the ceramic industry of the City of Natal and the other from the ceramic industry of the Serid? Region, both in the State of Rio Grande do Norte, Brazil. Percentages of iron ore waste of 5%, 10% , 15%, 20%, 25% and 30% were used in the tested ceramic matrix. The two clays and the iron ore waste used as part of this investigation were characterized by X-ray diffraction tests, X-ray fluorescence tests, differential thermal analysis, thermogravimetric analysis and dilatometric analysis. The samples were sintered under temperatures of 850 ?C, 950 ?C and 1050?C at a heating rate of 5 ?C/min with isotherms of two hours. The following tests were performed with the samples: linear shrinkage, water absorption, apparent porosity, apparent density, mass loss in fire and bending resistance in order to obtain their physical and mechanical properties. An amount of 5% of waste iron ore in the matrix clay at a temperature of 850 0C resulted in na increase of about 65% in the tensile strength of the clay samples from the Natal ceramic industry. A linear shrinkage of only 0.12% was observed for the samples, which indicates that the physical properties of the final product were not influenced by the addition of the waste / A ind?stria extrativa mineral ? respons?vel pela gera??o de res?duos provenientes do seu processo natural de extra??o. Os impactos da minera??o em ?rea urbana revestem-se de especial import?ncia devido ao alto grau de ocupa??o urbana, que s?o agravados face ? proximidade entre as ?reas mineradas e as ?reas habitadas. Algumas solu??es de disposi??o de res?duos t?m potencial de reduzir significativamente os riscos e o passivo ambiental, por?m representam custos elevados nas etapas de implanta??o e de opera??o. A adi??o de res?duos de min?rio de ferro como mat?ria-prima na elabora??o de produtos que s?o utilizados comercialmente tem como objetivo diminuir os impactos ambientais, transformando esses res?duos em elementos positivos na gera??o de trabalho e renda. O presente trabalho estuda a incorpora??o de res?duo de min?rio de ferro em duas argilas, uma do p?lo cer?mico da Grande Natal e outra do p?lo cer?mico da Regi?o do Serid?, ambas do Estado do Rio Grande do Norte, em percentuais de 5%, 10%, 15%, 20%, 25% e 30%, com o intuito de verificar a potencialidade da incorpora??o do res?duo em matriz cer?mica. As mat?rias-primas foram caracterizadas atrav?s dos ensaios de difra??o de raios X, fluoresc?ncia de raios X, an?lise t?rmica diferencial, an?lise termogravim?trica e an?lise dilatom?trica. Os corpos-de-prova foram conformados e em seguida sinterizados nas temperaturas de 850 ?C, 950 ?C e 1050 ?C, ? taxa de aquecimento de 5 ?C/min, com iso terma de duas horas. Foram realizados os ensaios tecnol?gicos de retra??o linear de queima, absor??o de ?gua, porosidade aparente, massa espec?fica aparente, perda demassa ao fogo e tens?o de ruptura ? flex?o, com o intuito de se obter suas propriedades f?sicas e mec?nica. A argila do p?lo cer?mico da Grande Natal apresentou um aumento em torno de 65% na tens?o de ruptura ? flex?o, com um percentual ?nfimo de 0,12% na sua retra??o linear de queima, quando adicionados 5% de res?duo de min?rio de ferro na matriz de argila ? temperatura de 850 ?C, n?o comprometendo assim as p ropriedades f?sicas do produto final

xv, 322 p. : ill. A print copy of this thesis is available through the UO Libraries. Search the library catalog for the location and call number. / This dissertation examines formal and popular campaigns against collaborators during the second Sino-Japanese war of 1937-1945, considering the role of these campaigns in the political struggles of the Nationalist (Guomindang) government, the interplay between discourses of law and morality, and the interactions of legal professionals, intellectuals, and commoners in the development of Chinese nationalism and anti-imperialism. During the Sino-Japanese war, the Japanese army occupied vast areas in China and sponsored puppet regimes at central and local levels in areas under its occupation. These regimes variously attracted, persuaded, or forced a large number of Chinese officials, intellectuals, and local elites to work in their administrations. The Nationalist government under Chiang Kai-shek, which was the central Chinese government since 1928, retreated to the inland city of Chongqing to organize resistance against Japan. The Nationalist government labeled collaborators as hanjian , "traitors to the Han." The word became widely used in legal regulations, popular literature, and newspapers and became the most derogatory and politically disastrous title possible for a Chinese citizen. Individuals designated hanjian were exposed to public humiliation, confiscation of land and property, and the threat of assassination. Chiang Kai-shek's government also called for the common people to expose hanjian . Most such accusations were then transformed into legal procedures. These accusations resulted in varying and often unfair sentences. Designed by the Nationalist government to harness the force of popular nationalism and to restore justice, the anti- hanjian campaigns instead inadvertently exposed the corruption and incompetence of the Nationalist government and damaged the post-war construction effort. / Committee in charge: Bryna Goodman, Chairperson, History; Andrew Goble, Member, History; Ina Asim, Member, History; Tze-lan Sang, Outside Member, East Asian Languages & Literature