Research on heterogeneous seamless handover has become popular, since the wireless networking systems were introduced with mobility. The typical current vertical handover mechanism consists of an architecture built at Layer 3 and implemented to serve for different technologies. Even though it has advantages in terms of simplicity, there still exist drawbacks such as the need of adaptation of the network architecture to different network technologies and systems. Providing transparency to the heterogeneous seamless handover can be provided by conducting the handover process at a higher layer. By that way, efficient handover decisions for vertical handover are made with more number of constraints that will lead to high performance, accessibility and low cost. Making this possible is by providing Quality of Experience (QoE) and obtaining current information of the throughput by measurements in the network. Analyzing and interpreting the statistics collected through measurements are vital in terms of decision making and to decide when to perform vertical handover. This thesis consists of the implementation of measurement module in two different approaches (Payload Dependent Approach and Payload Independent Approach) that will provide these statistics to the storage module in PERIMETER project.

To provide Internet applications with good performance, the transport protocol TCP is designed to optimize the throughput of data transfers. Today, however, more and more applications rely on low latency rather than throughput. Such applications can be referred to as data-limited and are not appropriately supported by TCP. Another emerging problem is associated with the use of novel networking techniques that provide infrastructure-less networking. To improve connectivity and performance in such environments, multi-path routing is often used. This form of routing can cause packets to be reordered, which in turn hurts TCP performance. To address timeliness issues for data-limited traffic, we propose and experimentally evaluate several transport protocol adaptations. For instance, we adapt the loss recovery mechanisms of both TCP and SCTP to perform faster loss detection for data-limited traffic, while preserving the standard behavior for regular traffic. Evaluations show that the proposed mechanisms are able to reduce loss recovery latency with 30-50%. We also suggest modifications to the TCP state caching mechanisms. The caching mechanisms are used to optimize new TCP connections based on the state of old ones, but do not work properly for data-limited flows. Additionally, we design a SCTP mechanism that reduces overhead by bundling several packets into one packet in a more timely fashion than the bundling normally used in SCTP. To address the problem of packet reordering we perform several experimental evaluations, using TCP and state of the art reordering mitigation techniques. Although the studied mitigation techniques are quite good in helping TCP to sustain its performance during pure packet reordering events, they do not help when other impairments like packet loss are present. / <p>Paper V was in manuscript form at the time of the defense.</p>

2015-05-27に本文を差替 / 京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第18481号 / 情博第532号 / 新制||情||94(附属図書館) / 31359 / 京都大学大学院情報学研究科知能情報学専攻 / (主査)教授 黒橋 禎夫, 教授 田中 克己, 教授 河原 達也 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

No description available.

In science and engineering, many applications require the solution of a sequence of linear systems. There are many ways to solve linear systems and we always look for methods that are faster and/or require less storage. In this dissertation, we focus on solving these systems with Krylov subspace methods and how to obtain effective preconditioners inexpensively. We first present an application for electronic structure calculation. A sequence of slowly changing linear systems is produced in the simulation. The linear systems change by rank-one updates. Properties of the system matrix are analyzed. We use Krylov subspace methods to solve these linear systems. Krylov subspace methods need a preconditioner to be efficient and robust. This causes the problem of computing a sequence of preconditioners corresponding to the sequence of linear systems. We use recycling preconditioners, which is to update and reuse existing preconditioner. We investigate and analyze several preconditioners, such as ILU(0), ILUTP, domain decomposition preconditioners, and inexact matrix-vector products with inner-outer iterations. Recycling preconditioners produces cumulative updates to the preconditioner. To reduce the cost of applying the preconditioners, we propose approaches to truncate the cumulative preconditioner updates, which is a low-rank matrix. Two approaches are developed. The first one is to truncate the low-rank matrix using the best approximation given by the singular value decomposition (SVD). This is effective if many singular values are close to zero. If not, based on the ideas underlying GCROT and recycling, we use information from an Arnoldi recurrence to determine which directions to keep. We investigate and analyze their properties. We also prove that both truncation approaches work well under suitable conditions. We apply our truncation approaches on two applications. One is the Quantum Monte Carlo (QMC) method and the other is a nonlinear second order partial differential equation (PDE). For the QMC method, we test both truncation approaches and analyze their results. For the PDE problem, we discretize the equations with finite difference method, solve the nonlinear problem by Newton's method with a line-search, and utilize Krylov subspace methods to solve the linear system in every nonlinear iteration. The preconditioner is updated by Broyden-type rank-one updates, and we truncate the preconditioner updates by using the SVD finally. We demonstrate that the truncation is effective. In the last chapter, we develop a matrix reordering algorithm that improves the diagonal dominance of Slater matrices in the QMC method. If we reorder the entire Slater matrix, we call it global reordering and the cost is O(N^3), which is expensive. As the change is geometrically localized and impacts only one row and a modest number of columns, we propose a local reordering of a submatrix of the Slater matrix. The submatrix has small dimension, which is independent of the size of Slater matrix, and hence the local reordering has constant cost (with respect to the size of Slater matrix). / Ph. D.

The availability of high performance computing clusters has allowed scientists and engineers to study more challenging problems. However, new algorithms need to be developed to take advantage of the new computer architecture (in particular, distributed memory clusters). Since the solution of linear systems still demands most of the computational effort in many problems (such as the approximation of partial differential equation models) iterative methods and, in particular, efficient preconditioners need to be developed. In this study, we consider application of incomplete LU (ILU) preconditioners for finite element models to partial differential equations. Since finite elements lead to large, sparse systems, reordering the node numbers can have a substantial influence on the effectiveness of these preconditioners. We study two implementations of the ILU preconditioner: a stucturebased method and a threshold-based method. The main emphasis of the thesis is to test a variety of breadth-first ordering strategies on the convergence properties of the preconditioned systems. These include conventional Cuthill-McKee (CM) and Reverse Cuthill-McKee (RCM) orderings as well as strategies related to the physical distance between nodes and post-processing methods based on relative sizes of associated matrix entries. Although the success of these methods were problem dependent, a number of tendencies emerged from which we could make recommendations. Finally, we perform a preliminary study of the multi-processor case and observe the importance of partitioning quality and the parallel ILU reordering strategy. / Master of Science

"Efficient querying over streaming data is a critical technology which requires the ability to handle numerous and possibly similar queries in real time dynamic environments such as the stock market and medical devices. Existing DBMS technology is not well suited for this domain since it was developed for static historical data. Queries over streams often contain relative window predicates such as in the query: ``Heart rate decreased to fifty-two beats per second within four seconds after the patient's temperature started rising." Relative window predicates are a specific type of join between streams that is based on the tuple's timestamp. In our operator, called Juggler, predicates are classified into three types: attribute, join, and window. Attribute predicates are stream values compared to a constant. Join predicates are stream values compared to another stream's values. Window predicates are join predicates where the streams' timestamp values are compared. Juggler's composite operator incorporates the processing of similar though not identical, query functionalities as one complex computation process. This execution strategy handles multi-way joins for multiple selection and join predicates. It adaptively orders the execution of predicates by their selectivity to efficiently process multiple continuous queries based on stream characteristics. In Juggler, all similar predicates are grouped into lists. These indices are represented by a collection of bits. Every tuple contains the bit structure representation of the predicate lists which encodes tuple predicate evaluation history. Every query also contains a similar bit structure to encode the predicate's relationship to the registered queries. The tuple's and query's bit structures are compared to assess if the tuple has satisfied a query. Juggler is designed and implemented in Java. Experiments were conducted to verify correctness and to assess the performance of Juggler's three features. Its adaptivity of reordering the evaluation of predicate types performed as well as the most selective predicate ordering. Its ability to exploit similar predicates in multiple queries showed reduction in number of comparisons. Its effectiveness when multiple queries are combined in a single Juggler operator indicated potential performance improvements after optimization of Juggler's data structures."

This thesis is focused on the analysis and development of Ensembles of Neural Networks. An ensemble is a system in which a set of heterogeneous Artificial Neural Networks are generated in order to outperform the Single network based classifiers. However, this proposed thesis differs from others related to ensembles of neural networks [1, 2, 3, 4, 5, 6, 7] since it is organized as follows. In this thesis, firstly, an ensemble methods comparison has been introduced in order to provide a rank-based list of the best ensemble methods existing in the bibliography. This comparison has been split into two researches which represents two chapters of the thesis. Moreover, there is another important step related to the ensembles of neural networks which is how to combine the information provided by the neural networks in the ensemble. In the bibliography, there are some alternatives to apply in order to get an accurate combination of the information provided by the heterogeneous set of networks. For this reason, a combiner comparison has also been introduced in this thesis. Furthermore, Ensembles of Neural Networks is only a kind of Multiple Classifier System based on neural networks. However, there are other alternatives to generate MCS based on neural networks which are quite different to Ensembles. The most important systems are Stacked Generalization and Mixture of Experts. These two systems will be also analysed in this thesis and new alternatives are proposed. One of the results of the comparative research developed is a deep understanding of the field of ensembles. So new ensemble methods and combiners can be designed after analyzing the results provided by the research performed. Concretely, two new ensemble methods, a new ensemble methodology called Cross-Validated Boosting and two reordering algorithms are proposed in this thesis. The best overall results are obtained by the ensemble methods proposed. Finally, all the experiments done have been carried out on a common experimental setup. The experiments have been repeated ten times on nineteen different datasets from the UCI repository in order to validate the results. Moreover, the procedure applied to set up specific parameters is quite similar in all the experiments performed. It is important to conclude by remarking that the main contributions are: 1) An experimental setup to prepare the experiments which can be applied for further comparisons. 2) A guide to select the most appropriate methods to build and combine ensembles and multiple classifiers systems. 3) New methods proposed to build ensembles and other multiple classifier systems.

[pt] Esta dissertação tem por objetivo analisar o processo de reordenamento das entidades da Rede Socioassistencial privada, que integram o Conselho Municipal de Assistência Social do Rio de Janeiro (CMAS/RJ), frente aos Marcos Legais de operacionalização da Política Nacional de Assistência Social (PNAS), no período de 2013-2015. Desde 2004, a Assistência Social vem definindo um conjunto de reformulações significativas, sejam elas políticas, e/ou normativas e estruturais de execução e gestão, que caracterizam o processo de reordenamento institucional, demarcando uma trajetória pautada em permanências e rupturas, caracterizadoras de transições históricas que vão da caridade ao direito social. O percurso metodológico envolve pesquisa documental, relativas às deliberações das normativas de inscrição e regularização das entidades de assistência social, serviços, programas e projetos; pesquisa qualitativa e entrevistas semiestruturadas, junto a Conselheiros da rede socioassistencial privada do CMAS/RJ. Destaca-se também a observação participante no acompanhamento das reuniões do CMAS/RJ; bem como, através do assessoramento realizado com as 29 entidades filantrópicas. A abordagem permite concluir que apesar da importância em adequar as ações institucionais aos marcos legais, temos um Estado que ora considera estas entidades como fundamentais na execução da política pública, ora concorre com elas, na busca de financiamento privado. Neste sentido, procuramos demonstrar a existência de dificuldades do Estado na condução e execução da política pública, assim como das entidades socioassistenciais em relação ao seu papel complementar de execução; falta maior interlocução entre ambos para garantir a efetivação da política de assistência social, enquanto direito social. Em relação ao CMAS/RJ, verifica-se baixa discursão política quanto ao seu papel neste processo, prevalecendo uma adequação burocrática na efetivação desses marcos legais. / [en] This thesis aims to analyze the reorganization process of the entities of private social assistance network, comprising the Municipal Council of Social Welfare of Rio de Janeiro (CMAS / RJ), compared to the Legal Frameworks of operation of the National Social Assistance Policy (PNAS) in the period 2013-2015. Since 2004, the Social Assistance has defined a set of significant reformulations, whether political, and / or regulatory and structural implementation and management, which characterize the process of institutional reorganization, marking a path guided by continuities and ruptures, characterizing historical transitions ranging from charity to social rights. The methodological approach involves documentary research relating to the deliberations of the registration regulations and regularization of social assistance entities, services, programs and projects; qualitative research and semi-structured interviews with the directors of private social-network CMAS / RJ. Also noteworthy participant observation in monitoring the meetings of the CMAS / RJ; as well as through advisory services offered to the 29 charities. The approach shows that despite the importance of adequate institutional actions to the legal framework, we have a state that now considers these entities as fundamental in the implementation of public policy, sometimes competing with them in the search for private financing. In this sense, we seek to demonstrate the existence of State difficulties in conducting and execution of public policy and the social assistance entities in relation to its complementary role of enforcement; lack greater dialogue between them to ensure the effectiveness of social welfare policy, as a social right. Regarding the CMAS / RJ, there is low political increasing discussion about their role in this process, prevailing bureaucratic adequacy in terms of these legal frameworks.

La thèse étudie le pouvoir génératif de la structuration des connaissances pour la conception innovante. Partant d’une analyse de la générativité des structures de connaissances non-verbales, et notamment du dessin d’architecte, elle identifie un nouveau mode de conception en montrant comment une revisite des rapports entre connaissances favorise la génération de nouveaux concepts. En particulier, une restructuration des connaissances visant à rendre la structure splitting, c’est-à-dire, sans rapports modulaires ou déterministes entre connaissances, possède un fort pouvoir génératif. En session collective, les médias non-verbaux, tels que le dessin, les images ou encore le prototypage 3D, se révèlent ainsi particulièrement favorables à la restructuration des connaissances : trois conditions nécessaires présentées dans la thèse doivent cependant être réunies afin de garantir la performance du non-verbal en session collective. Par ailleurs, ce phénomène de restructuration générative - aussi appelée K-preordering – diffère de l’exploration classique par génération de concepts alternatifs. Les deux modes de conception ne sont pas pour autant incompatibles : la thèse propose ainsi un processus de conception alternant phases de K-preordering et phases de génération de concepts. Ce processus est notamment testé à travers deux cas de recherche-intervention visant à provoquer la réorganisation des structures de connaissances stabilisées et partagées à l’échelle de l’organisation. Ces deux cas montrent en particulier comment un tel processus favorise la conception de programmes de recherche transdisciplinaires. / The thesis investigates the generative power of knowledge structuration for innovative design. Analyzing the generativity of non-verbal knowledge structures - especially, architectural drawings -, this work identifies a new design method by showing that the transformation of links between knowledge bases fosters concept generation. In particular, a restructuration that aims to design a splitting structure - a structure avoiding modular and deterministic links between knowledge bases - has a strong generative power. During collective creativity sessions, non-verbal tools such as sketches, pictures or 3D-printing, can enhance knowledge restructuration: however, three necessary conditions, which are presented in the thesis, have to be met in order to ensure performance of the non-verbal tools used during the session. Moreover, this generative restructuration - also called K-preordering - differs from the classic design exploration by alternative concepts generation. But the two design modes are not mutually exclusive: the thesis proposes a design process that alternates K-preordering and concept generation. This process is applied in two studies, which were led as action research and intended to foster the reorganization of knowledge structures that were both stabilized and shared at the organizational level. In particular, these studies show how such a process helps enhancing the design of transdisciplinary research programs.