Followed two different proofs of van der Waerden's theorem. Found that the two proofs yield important information about arithmetic progressions and the theorem. van der Waerden's theorem explains the occurrence of arithmetic progressions which can be used to explain such things as the Bible Code.

Orientador: Ali Messaoudi / Banca: Milton Edwin Cobo Cortez / Banca: Vanderlei Minori Horita / Resumo: Neste trabalho, estudaremos a dinâmica dos intercâmbio de intervalos. Em particular, mostraremos que se uma aplicação intercâmbio de intervalos é Q-linearmente independente e é irredutível então ela é minimal. Estudaremos também as propriedades dinâmicas da indução de Rauzy-Veech e provaremos que quase todo intercâmbio de intervalos é unicamente ergódico (prova de Boshernitzan) / Abstract: In this work we study dynamic of the map interval exchange. In particular, we show that if the interval exchange is Q-lineally independent and irreducible then it is minimal. We also study some dynamical prprieties of the Rauzy-Veech induction and we prove that almost all interval exchange is uniquely ergodic (proof of Boshernitzan) / Mestre

Soit G le groupe SO°(1, n) (n ≥ 3) ou PU(1, n) (n ≥ 2) et fixons une décomposition d'Iwasawa G = KAN. Soit ɼ un sous-groupe discret de G, que nous supposons Zariski-dense et de mesure de Bowen-Margulis-Sullivan finie. Lorsque G = SO°(1, n), nous étudions la géométrie de la mesure de Bowen-Margulis-Sullivan le long des sous-groupes fermés connexes de N, en lien avec la dichotomie de Mohammadi-Oh. Nous établissons des résultats déterministes sur la dimension des projections de la mesure de Patterson- Sullivan. Lorsque G = PU(1, n), nous relions la géométrie de la mesure de Bowen- Margulis-Sullivan le long du centre du groupe de Heisenberg au problème du calcul de la dimension de Hausdorff de l'ensemble limite relativement à la distance sphérique au bord. Nous calculons cette dimension pour certains groupes de Schottky. / Let G be the group SO° (1,n) (n ≥ 3) or PU(1, n) (n ≥ 2) and fix some Iwasawa decomposition G = KAN. Let ɼ be a discrete subgroup of G.We assume that ɼ is Zariski-dense with finite Bowen-Margulis-Sullivan measure. When G = SO°(1,n), we investigate the geometry of the Bowen-Margulis-Sullivan measure elong connected closed subgroups of N. This is related to the Mohammadi-Oh dichotomy. We then prove deterministic results on the dimension of projections of Patterson-Sullivan measure. When G = PU(1,n), we relate the geometry of Bowen-Margulis-Sullivan measure along the center of Heisenberg group to the problem of computing the Hausdorff dimension of the limit set with respect to the spherical metric on the boudary. We construct some Schottky subgroups for wich we are able to compute this dimension.

Orientador: Vanderlei Minori Horita / Banca: Roberto Markarian / Banca: Paulo Ricardo da Silva / Resumo: O objetivo principal deste trabalho e estudar a dinâmica de uma partícula pontual no interior de subconjuntos do plano. Tais sistemas são conhecidos na literatura como bilhares. Apresentaremos os principais conceitos desses sistemas e veremos que tais sistemas deixam invariante uma medida de probabilidade, o que nos permite aplicar a Teoria Ergódica ao problema do bilhar / Abstract: The main goal of this work is to study the dynamical behavior of a point-like (dimensionless) particle in the interior of planar regions. Such systems are known in the literature as billiards. We're going to present the principal concepts of those systems and we'll see that such system turns the probability measure invariant, which allows us to apply the Ergodic Theory to billiard problems / Mestre

Soit (X, μ) un espace de probabilité standard et Γ un groupe dénombrable agissant sur X de manière à préserver la mesure de probabilité (p.m.p.). La partition de l’espace X en orbites induite par l’action de Γ est entièrement encodée par le groupe plein de l’action, constitué de l’ensemble des bijections boréliennes de l’espace qui agissent par permutation sur chaque orbite. Plus précisément, le théorème de reconstruction de H. Dye stipule que deux actions p.m.p. sont orbitalement équivalentes (i.e. induisent la même partition à une bijection p.m.p. près) si et seulement si leurs groupes pleins sont isomorphes.Le sujet de cette thèse est grandement motivé par ce théorème de reconstruction, puisqu’il s’agit de voir comment des invariants d’équivalence orbitale, qui portent donc sur la partition de l’espace en orbites, se traduisent en des propriétés algébriques ou topologiques du groupe plein associé.Le résultat majeur porte sur le rang topologique des groupes pleins, c’est-à-dire le nombre minimum d’éléments nécessaires pour engendrer un sous-groupe dense. Il se trouve être fortement relié a un invariant fondamental d’équivalence orbitale : le coût. Plus précisément, nous avons montré que le rang topologique était, dans le cas ergodique, égal à la partie entière du coût de l’action plus un. Le cas non ergodique a également été étudié, et on a obtenu des résultats complémentaires sur la généricité de l’ensemble des générateurs topologiques.Enfin, on a caractérisé les actions dont toutes les orbites sont infinies : ce sont exac- tement celles dont le groupe plein n’admet aucun morphisme non trivial à valeurs dans Z/2Z. / Let (X,μ) be a standard probability space and Γ a countable group acting on X in a measure preserving way. The partition of the space X into Γ-orbits is entirely encoded by the full group of the action, consisting of all the Borel bijections of X which act by permutation on every orbit. To be more precise, Dye’s reconstruction theorem states that two measure preserving actions are orbit equivalent (i.e. they induce the same partition up to a measure preserving bijection of (X, μ)) if and only if their full groups are isomorphic.The reconstruction theorem is the main motivation for this thesis, in which we try to understand how exactly orbit equivalence invariants of measure preserving actions translate into algebraic or topological properties of the associated full group.The main result deals with the topological rank of full groups, that is the minimal number of elements needed to generate a dense subgroup. It happens to be deeply linked to a fundamental invariant of orbit equivalence : the cost. To be more precise, we have shown that the topological rank is, in the ergodic case, equal to the integer part of the cost of the action plus one. The non-ergodic case was also treated, and we obtained some genericity results for the set of topological generators.We also obtained a characterization of the measure preserving actions having only infinite orbits : these are the ones whose full group has non nontrivial morphism into Z/2Z.

Orientador: Nancy Lopes Garcia / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática, Estatística e Computação Científica / Made available in DSpace on 2018-08-20T13:18:28Z (GMT). No. of bitstreams: 1 Fadel_DesireeFaria_M.pdf: 12260685 bytes, checksum: 0187ee6ba6eb07c46ce97bd741e51a6c (MD5) Previous issue date: 2012 / Resumo: Neste trabalho, iremos considerar modelos autorregressivos com memória variável estacionários (AR-MV). Em particular, consideraremos modelos cuja memória depende do valor do primeiro antecessor, Yt-1, pertencer a uma partição da reta determinada por um parâmetro 'ALFA' (escalar ou vetorial), chamado de parâmetro limiar. O objetivo deste trabalho é estimar o parâmetro limiar 'ALFA' através de uma adaptação do método proposto por Hansen (2000). A ideia do método é minimizar a soma dos quadrados dos erros estimando, sequencialmente, os coeficientes 'BETA' do modelo autorregressivo (AR) supondo primeiramente que 'ALFA' é conhecido, e depois estimar o parâmetro 'ALFA' utilizando o valor estimado '^BETA' até atingir a convergência. A comparação dos modelos AR-MV com os respectivos AR foi feita através da capacidade de previsão de cada um deles / Abstract: In this paper, we consider stationary autoregressive models with variable memory (AR-MV). In particular, we consider models whose memory depends on the value of the first ancestor, Yt-1, to belong to a partition of the line determined by a parameter 'ALPHA' (scalar or vector), called the threshold parameter. The objective of this study is to estimate the threshold parameter 'ALPHA' by adapting the method proposed by Hansen (2000). The idea of the method is to minimize the sum of squared errors by estimating sequentially the coefficients 'BETA' of the autoregressive model (AR) assuming first that 'ALPHA' is known, and then estimate the parameter 'ALPHA' using the estimated value of '^BETA' until convergence is achieved. The comparison of the AR-MV with the respective AR was performed by the ability to predict each / Mestrado / Estatistica / Mestre em Estatística

This thesis report presents background knowledge about cognitive radio network (CRN) and investigates performance of underlay cognitive radio networks based on an adaptive power allocation policy of secondary transmitter (SU-Tx). In particular, it has been assumed that SU-Tx and primary user transmitter (PU-Tx) are equipped with a single antenna, while the corresponding receivers are equipped with multiple antennas. Additionally, SU-Tx operates under the joint constraint of its peak transmission power and outage constraint of the primary network. The probability density function (PDF) and cumulative density function (CDF) of the signal to interference and noise ratio (SINR) of SU over Rayleigh fading channel are derived. Using these two functions, a closed-form expression for the outage probability and an approximate expression for ergodic capacity of the considered system are obtained. Matlab simulation results are provided to validate the correctness of the analyses. The results show that simulation and analytical results closely match. The results show that the performance of SU increases as power of PU increases, but behaves the opposite after SU-Tx reaches its peak transmission power. Furthermore, the results reveal that as the number of antennas at the receivers (both SU and PU receivers) increases, the performance of the SU network increases. / maarig2000@gmail.com, marshed18@hotmail.com

Cette thèse est consacrée à l'étude des différents problèmes liés aux structures des cubes , en théorie ergodique et en dynamique topologique. Elle est composée de six chapitres. La présentation générale nous permet de présenter certains résultats généraux en théorie ergodique et dynamique topologique. Ces résultats, qui sont associés d'une certaine façon aux structures des cubes, sont la motivation principale de cette thèse. Nous commençons par les structures de cube introduites en théorie ergodique par Host et Kra (2005) pour prouver la convergence dans $L^2 $ de moyennes ergodiques multiples. Ensuite, nous présentons la notion correspondante en dynamique topologique. Cette théorie, développée par Host, Kra et Maass (2010), offre des outils pour comprendre la structure topologique des systèmes dynamiques topologiques. En dernier lieu, nous présentons les principales implications et extensions dérivées de l'étude de ces structures. Ceci nous permet de motiver les nouveaux objets introduits dans la présente thèse, afin d'expliquer l'objet de notre contribution. Dans le Chapitre 1, nous nous attachons au contexte général en théorie ergodique et dynamique topologique, en mettant l'accent sur l'étude de certains facteurs spéciaux. Les Chapitres 2, 3, 4 et 5 nous permettent de développer les contributions de cette thèse. Chaque chapitre est consacré à un thème particulier et aux questions qui s'y rapportent, en théorie ergodique ou en dynamique topologique, et est associé à un article scientifique. Les structures de cube mentionnées plus haut sont toutes définies pour un espace muni d'une unique transformation. Dans le Chapitre 2, nous introduisons une nouvelle structure de cube liée à l'action de deux transformations S et T qui commutent sur un espace métrique compact X. Nous étudions les propriétés topologiques et dynamiques de cette structure et nous l'utilisons pour caractériser les systèmes qui sont des produits ou des facteurs de produits. Nous présentons également plusieurs applications, comme la construction des facteurs spéciaux. Le Chapitre 3 utilise la nouvelle structure de cube définie dans le Chapitre 2 dans une question de théorie ergodique mesurée. Nous montrons la convergence ponctuelle d'une moyenne cubique dans un système muni deux transformations qui commutent. Dans le Chapitre 4, nous étudions le semigroupe enveloppant d'une classe très importante des systèmes dynamiques, les nilsystèmes. Nous utilisons les structures des cubes pour montrer des liens entre propriétés algébriques du semigroupe enveloppant et les propriétés topologiques et dynamiques du système. En particulier, nous caractérisons les nilsystèmes d'ordre 2 par une propriété portant sur leur semigroupe enveloppant. Dans le Chapitre 5, nous étudions les groupes d'automorphismes des espaces symboliques unidimensionnels et bidimensionnels. Nous considérons en premier lieu des systèmes symboliques de faible complexité et utilisons des facteurs spéciaux, dont certains liés aux structures de cube, pour étudier le groupe de leurs automorphismes. Notre résultat principal indique que, pour un système minimal de complexité sous-linéaire, le groupe d'automorphismes est engendré par l'action du shift et un ensemble fini. Par ailleurs, en utilisant les facteurs associés aux structures de cube introduites dans le Chapitre 2, nous étudions le groupe d'automorphismes d'un système de pavages représentatif. La bibliographie, commune à l'ensemble de la thèse, se trouve en fin document / This thesis is devoted to the study of different problems in ergodic theory and topological dynamics related to og cube structures fg. It consists of six chapters. In the General Presentation we review some general results in ergodic theory and topological dynamics associated in some way to cubes structures which motivates this thesis. We start by the cube structures introduced in ergodic theory by Host and Kra (2005) to prove the convergence in $L^2$ of multiple ergodic averages. Then we present its extension to topological dynamics developed by Host, Kra and Maass (2010), which gives tools to understand the topological structure of topological dynamical systems. Finally we present the main implications and extensions derived of studying these structures, we motivate the new objects introduced in the thesis and sketch out our contributions. In Chapter 1 we give a general background in ergodic theory and topological dynamics given emphasis to the treatment of special factors. % We give basic definitions and describe special factors associated to a From Chapter 2 to Chapter 5 we develop the contributions of this thesis. Each one is devoted to a different topic and related questions, both in ergodic theory and topological dynamics. Each one is associated to a scientific article. In Chapter 2 we introduce a novel cube structure to study the actions of two commuting transformations $S$ and $T$ on a compact metric space $X$. In the same chapter we study the topological and dynamical properties of such structure and we use it to characterize products systems and their factors. We also provide some applications, like the construction of special factors. In the same topic, in Chapter 3 we use the new cube structure to prove the pointwise convergence of a cubic average in a system with two commuting transformations. In Chapter 4, we study the enveloping semigroup of a very important class of dynamical systems, the nilsystems. We use cube structures to show connexions between algebraic properties of the enveloping semigroup and the geometry and dynamics of the system. In particular, we characterize nilsystems of order 2 by its enveloping semigroup. In Chapter 5 we study automorphism groups of one-dimensional and two-dimensional symbolic spaces. First, we consider low complexity symbolic systems and use special factors, some related to the introduced cube structures, to study the group of automorphisms. Our main result states that for minimal systems with sublinear complexity such groups are spanned by the shift action and a finite set. Also, using factors associated to the cube structures introduced in Chapter 2 we study the automorphism group of a representative tiling system. The bibliography is defer to the end of this document

In-band full-duplex (FD) communications have been optimistically promoted to improve the spectrum utilization and efficiency. However, the penetration of FD communications to the cellular networks domain is challenging due to the imposed uplink/downlink interference. This thesis presents a tractable framework, based on stochastic geometry, to study FD communications in multi-tier cellular networks. Particularly, we assess the FD communications effect on the network performance and quantify the associated gains. The study proves the vulnerability of the uplink to the downlink interference and shows that the improved FD rate gains harvested in the downlink (up to 97%) comes at the expense of a significant degradation in the uplink rate (up to 94%). Therefore, we propose a novel fine-grained duplexing scheme, denoted as α-duplex scheme, which allows a partial overlap between the uplink and the downlink frequency bands. We derive the required conditions to harvest rate gains from the α-duplex scheme and show its superiority to both the FD and half-duplex (HD) schemes. In particular, we show that the α-duplex scheme provides a simultaneous improvement of 28% for the downlink rate and 56% for the uplink rate. We also show that the amount of the overlap can be optimized based on the network design objective. Moreover, backward compatibility is an essential ingredient for the success of new technologies. In the context of in-band FD communication, FD base stations (BSs) should support HD users' equipment (UEs) without sacrificing the foreseen FD gains. The results show that FD-UEs are not necessarily required to harvest rate gains from FD-BSs. In particular, the results show that adding FD-UEs to FD-BSs offers a maximum of 5% rate gain over FD-BSs and HD-UEs case, which is a marginal gain compared to the burden required to implement FD transceivers at the UEs' side. To this end, we shed light on practical scenarios where HD-UEs operation with FD-BSs outperforms the operation when both the BSs and UEs are FD and we find a closed form expression for the critical value of the self-interference cancellation power required for the FD UEs to outperform HD UEs.

Eine Verallgemeinerung der klassischen Resultate von Von Neumann und Birkhoff ist die Frage nach gewichteten Versionen ihrer Theoreme. Eine zentrale Antwort auf diese Fragestellung lieferte Jean Bourgain 1988 mit seinem Rückkehrzeitentheorem. Aufbauend auf dem Beweis von Bourgain, Furstenberg, Katznelson und Ornstein aus dem Jahr 1989 sowie dem Buch von Assani präsentieren wir einen ausführlichen und vollständigen Beweis und besprechen insbesondere den Fall von dynamischen Systemen mit rein atomarer invarianter sigma-Algebra. / In this diploma thesis we present a detailed proof of Bourgain's Return Times Theorem due to Bourgain, Furstenberg, Katznelson and Ornstein following their paper as well as the book by Assani. In particular, we discuss the case of systems with a purely atomic invariant sigma-algebra in all details.