"Faster, higher, stronger"---the Olympic motto is being pursued and practised in the design of broadband wireless communication systems. Motivated by the huge demands for fast and reliable communications over wireless channels, broadband communication systems are required to provide faster (low-complexity) data processing, higher data throughput and stronger (lower error rate) performance. In practice, however, broadband communication systems must cope with critical performance-limiting challenges that include time- and frequency-selective fading channels, noise, inter-symbol interference (ISI), intercarrier interference (ICI) as well as power and bandwidth constraints. To address these challenges, this thesis investigates several physical layer aspects of broadband wireless communication systems. / Incorporating OFDM into multiple-input multiple-output (MIMO) system, MIMO-OFDM has been shown to provide larger channel capacity and greater diversity gain. However, current coding schemes for MIMO-OFDM are either space-time coded (STC) OFDM without the guarantee of full diversity gains or space-frequency coding (SFC) with a greater loss of data rate. Furthermore, most existing STC and SFC have focused on quasi-static fading which is not practical for broadband wireless communications. When multi-band OFDM (MB-OFDM) is applied to ultra-wide band (UWB) communications, a high diversity can be obtained, but in the expense of a much lower (close to half) data rate. To address the limitations of existing coding schemes for broadband wireless communication systems, this thesis: (i) proposes a space-time-frequency coding (STFC) that can achieve maximum diversity and maximum symbol rate transmission over MIMO block-fading channels; (ii) derives a high-rate full-diversity SFC from STFC tailored for frequency-selective fading channels; and (iii) proposes a high-rate high-diversity algebraic time-frequency coding (ATFC) for MB-OFDM system. / Orthogonal frequency division multiplexing (OFDM) is an effective technique to eliminate ISI in broadband wireless communications. This thesis studies the problem of training-based OFDM channel estimation and proposes a training method that minimizes the number of pilots employed to achieve a desired bit error rate (BER) performance. A clustered pilot pattern is further proposed to enhance the BER performance. Focusing on OFDM frequency synchronization, this thesis also proposes a clustered pilot tones placement and a novel pilot sequence design for carrier frequency offset (CFO) compensation. The analytical and simulation results show that the root mean square error (RMSE) of the CFO estimate can be greatly reduced. / Zhang Wei. / "July 2005." / Adviser: Pak-Chung Ching. / Source: Dissertation Abstracts International, Volume: 67-01, Section: B, page: 0461. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (p. 126-143). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Orientador: Cleonice Fátima Bracciali / Banca: Messias Meneguette Júnior / Banca: Maurílio Boaventura / Resumo: O objetivo principal deste trabalho é estudar o problema de análise de freqüência, utilizando polinômios ortogonais no intervalo [0,1]. Para isto, vimos os polinômios ortogonais no círculo unitário, conhecidos como polinômios de Szego, suas relações com as frações contínuas de Perron-Carathéodory e polinômios para-ortogonais. Estudamos, também, relacões entre polinômios para-ortogonais e polinômios ortogonais no intervalo [-1,1], e como são utilizados em análise de freqüência. / Abstract: The main purpose of this work is to study the frequency analysis problem using ortho- gonal polynomials on the interval [0,1]. For that, we study the orthogonal polynomials in the unit circle, known as Szeg}o polynomials, relations with the continued fractions of Perron- Carathéodory and para-orthogonal polynomials. We also study the relations between the para-orthogonal polynomials and orthogonal polynomials on the interval [-1,1], and how they are used in the frequency analysis problem. / Mestre

Seja K um grupo de Lie compacto agindo na esfera unitária S&#8319 por isometrias. Mostramos como uma cota superior para as curvaturas principais de uma órbita dessa ação pode ser usada (mas não é suficiente) para encontrar uma cota inferior para o diâmetro do espaço de órbitas S&#8319/K. Em seguida mostramos que existe uma órbita Kp com curvaturas principais majoradas por 4&#8730 14. / Let K be a compact Lie group acting on the unit sphere S&#8319 by isometries. We show how an upper bound on the principal curvatures of one orbit can be used (but is not sufficient) to obtain a lower bound for the diameter of the orbit space S&#8319/K. Then we show that there is an orbit Kp with principal curvatures bounded from above by 4&#8730 14.

The undercutting of toneholes has been practised for centuries with the aim of improving the tuning and playability of woodwind instruments. The influence of undercutting on tuning can be understood in terms of linear acoustic theory. Its effect on other playing characteristics is thought to lie in its reduction of local non-linear flow phenomena (boundary layer separation and the formation of jets and vortices) at the tonehole. Particle Image Velocimetry (PIV) is used to examine the oscillating airflow around a model woodwind tonehole. Velocity and vorticity information is obtained and compared for a square-edged tonehole and an undercut tonehole at a variety of sound levels. The upstream, internal edge of the tonehole is found to be the location of the most significant local non-linear flow behaviour. Undercutting is found to reduce the strength of local non-linear flow phenomena at a given sound level. Microphone measurements carried out in a reverberation chamber show that undercutting the tonehole also reduces the harmonic distortion introduced to the radiated pressure signal by the non-linear flow. Proper Orthogonal Decomposition (POD) is then applied to PIV data of oscillating flow at the end of a tube. It is used to approximately separate the acoustic field from the induced local non-linear flow phenomena. The POD results are then used to approximate the percentage of kinetic energy present in the non-linear flow. POD analysis is applied to the case of flow around the two toneholes. It shows a smaller transfer of kinetic energy to non-linear flow effects around the undercut tonehole at a given sound level. The dependence of the local non-linear flow kinetic energy on Strouhal number is considered.

No description available.

Avec le développement de nouveaux dispositifs apparait le besoin d’être capable de contrôler la chimique de surfaces de substrats multi-échelle et multi-matériaux. Plusieurs techniques font appel à de la chimie localisée via différentes technologies. Une approche consiste à exploiter les différences de réactivités chimiques entre les différents matériaux du substrat et différents groupements chimiques de manière à fonctionnaliser sélectivement chaque matériaux pour former des couches minces organiques de type « Self Assembled Monolayer »: Ce principe proposée par Pr. G. M. Whitesides est appelé chimie orthogonale. Dans le cadre de cette thèse, le but ultime était de réaliser la fonctionnalisation chimique orthogonale de substrats dont la surface était composée de SiO2/Au/TiW La première étape de ce travail a été de déterminer pour la première fois la fonction chimique la plus adaptée pour la fonctionnalisation de TiW. Pour se faire nous avons comparé la chimie des silanes, des acides phosphoniques et des catéchols. Après caractérisations (XPS, ToF-SIMS, IR) des différentes couches, la voie des acides phosphoniques semblait être celle donnant lieu à la couche la plus stable. Ensuite nous avons étudié l’orthogonalité sur de substrats bi-matériaux (SiO2/TiW ou Au/TiW), et enfin sur substrat dont la surface était constituée de Au/SiO2/TiW. / The development of nanotechnologies makes it possible to manufacture the micro or nanometric-sized patterns with various materials (dielectrics, metals, semiconductors). These heterogeneous surfaces are commonly used in the electronics industry for the production of nanoelectronic structures and components: transistors, memories or sensors. The concept of orthogonal chemical functionalization was first proposed by George M. Whitesides to modify the surfaces composed of different materials at the macroscopic scale. In this context, this PhD work aimed at exploring the orthogonal chemical functionalization approach on a predefined patterned titanium tungsten (TiW) surface by lithography producing. Pattern materials (Au, SiO2) are chosen to have different chemical properties, which can be functionalized with completely independent reactions. To achieve this aim, we have studied three different chemical groups for the formation of organolayers (silane, catechol, phosphonic acid) on TiW for the first time. The three layers were characterized (XPS, IR, ToF-SIMS) and the stability of the formed organolayers was also addressed. Then we developed and ascertained the orthogonal chemical functionalization of patterned Au/TiW and Au/SiO2/TiW surfaces. It proposes a novel strategy for the orthogonal functionalization on a triple-material patterned surface. In addition, the capture of nanoparticles by electrostatic interaction at specific location on Au/TiW patterned substrate was successfully implemented to prove the interest of such method for colloids trapping.

In this thesis we study the application of trace maps over Galois fields and Galois rings in the construction of non-binary linear and non-linear codes and mutually unbiased bases. Properties of the trace map over the Galois fields and Galois rings has been used very successfully in the construction of cocyclic Hadamard, complex Hadamard and Butson Hadamard matrices and consequently to construct linear codes over integers modulo prime and prime powers. These results provide motivation to extend this work to construct codes over integers modulo . The prime factorization of integers paved the way to focus our attention on the direct product of Galois rings and Galois fields of the same degree. We define a new map over the direct product of Galois rings and Galois fields by using the usual trace maps. We study the fundamental properties of the this map and notice that these are very similar to that of the trace map over Galois rings and Galois fields. As such this map called the trace-like map and is used to construct cocyclic Butson Hadamard matrices and consequently to construct linear codes over integers modulo . We notice that the codes construct in this way over the integers modulo 6 is simplex code of type . A further generalization of the trace-like map called the weighted-trace map is defined over the direct product of Galois rings and Galois fields of different degrees. We use the weighted-trace map to construct some non-linear codes and mutually unbiased bases of odd integer dimensions. Further more we study the distribution of over the Galois fields of degree 2 and use it to construct 2-dimensional, two-weight, self-orthogonal codes and constant weight codes over integers modulo prime.

Les travaux présentés dans cette thèse portent sur la classification et l'analyse des manipulateurs 3R à axes orthogonaux. Le but final est de proposer une classification exhaustive de ces manipulateurs selon leur topologie d'espace de travail. On définit une topologie d'espace de travail par un couple (nombre de points cusps, nombre de nœuds). Ces points particuliers apparaissent sur les surfaces de singularités. Notre étude se décompose en 4 parties. La première partie de ce travail est consacrée à une étude bibliographique de quelques notions liées à la cinématique et à la conception des manipulateurs sériels et plus particulièrement leurs singularités. Dans la deuxième partie de ce travail, nous établissons une classification exhaustive de toutes les topologies d'espace de travail d'une famille de manipulateurs 3R à axes orthogonaux telle que r3 = 0 en fonction des paramètres de DHm. Pour chaque topologie d'espace de travail, nous donnons toutes les propriétés cinématiques décrites dans la première partie. Une nouvelle condition nécessaire et suffisante pour qu'un manipulateur soit cuspidal (peut changer de posture sans franchir une singularité) a été établie sous forme d'une relation entre les paramètres de DHm. La troisième partie a été consacrée à étendre la classification réalisée dans la partie précédente à une famille de manipulateurs orthogonaux plus large (r3 <> 0). La quatrième partie a été consacrée à l'analyse des manipulateurs selon deux critères de performance. Le premier est relatif aux conditionnements maximum et moyen. Le second permet de calculer les proportions des régions à 2 et 4 solutions au MGI ainsi que tout le domaine accessible par rapport à une sphère centrée sur l'origine et englobant l'espace de travail. Les résultats proposés dans ce travail constituent une aide intéressante pour la conception de mécanismes innovants. En effet, en utilisant ce travail, nous pouvons choisir la topologie d'espace de travail la mieux adaptée à la tâche souhaitée. Nous venons ensuite ajuster les paramètres de DHm du manipulateur satisfaisant les critères proposés.

Multiple-input multiple-output (MIMO) antennas can be exploited to provide high data rate using a limited bandwidth through multiplexing gain. MIMO combined with orthogonal frequency division multiplexing (OFDM) could potentially provide high data rate and high spectral efficiency in frequency-selective fading channels. MIMO-OFDM technology has been widely employed in modern communication systems, such as Wireless Local Area Network (WLAN), Long Term Evolution (LTE) and Worldwide Interoperability for Microwave Access (WiMAX). However, most of the conventional schemes either are computationally prohibitive or underutilize the full performance gain provided by the inherent merits of MIMO and OFDM techniques. In the first part of this dissertation, we firstly study the channel matrix inversion which is commonly required in various MIMO detection schemes. An algorithm that exploits second-order extrapolation in the time domain is proposed to efficiently reduce the computational complexity. This algorithm can be applied to both linear detection and non-linear detection such as ordered successive interference cancellation (OSIC) while maintaining the system performance. Secondly, we study the complexity reduction for Lattice Reduction Aided Detection (LRAD) of MIMO-OFDM systems. We propose an algorithm that exploits the inherent feature of unimodular transformation matrix that remains the same for relatively highly correlated frequency components. This algorithm effectively eliminates the redundant brute-force lattice reduction iterations among adjacent subcarriers. Thirdly, we analyze the impact of channel coherence bandwidth on two LRAD algorithms. Analytical and simulation results demonstrate that carefully setting the initial calculation interval according to the coherence bandwidth is essential for both algorithms. The second part of this dissertation focuses on efficient multi-user (MU) scheduling and coordination for the uplink of WLAN that uses MIMO-OFDM techniques. On one hand, conventional MU-MIMO medium access control (MAC) protocols require large overhead, which lowers the performance gain of concurrent transmissions rendered by the multi-packet reception (MPR) capability of MIMO systems. Therefore, an efficient MU-MIMO uplink MAC scheduling scheme is proposed for future WLAN. On the other hand, single-user (SU) MIMO achieves multiplexing gain in the physical (PHY) layer and MU-MIMO achieves multiplexing gain in the MAC layer. In addition, the average throughput of the system varies depending on the number of antennas and users, average payload sizes, and signal-to-noise-ratios (SNRs). A comparison on the performance between SU-MIMO and MU-MIMO schemes for WLAN uplink is hence conducted. Simulation results indicate that a dynamic switch between the SU-MIMO and MU-MIMO is of significance for higher network throughput of WLAN uplink. / Graduation date: 2013

This research develops and applies nonparametric estimation tools in two sectors of interest of financial econometrics: risk management and market microstructure. In the first part we address the problem of estimating conditional quantiles in financial and economic time series. Research in this field received great impulse since quantile based risk measures such as Value at Risk (VaR) have become essential tools to assess the riskiness of trading activities. The great amounts of data available in financial time series allows building nonparametric estimators that are not subject to the risk of specification error of parametric models. A wavelet based estimator is developed. With this approach, minimum regularity conditions of the underlying process are required. Moreover the specific choice of the wavelets in this work leads to the constructions of shape preserving estimators of probability functions. In other words, estimates of probability functions, both densities and cumulative distribution functions, are probability functions themselves. This method is compared with competing methods through simulations and applications to real data. In the second part we carry out a nonparametric analysis of financial durations, that is of the waiting times between particular financial events, such as trades, quote updates, volume accumulation, that happen in financial markets. These data display very peculiar stylized facts one has to take into account when attempting to model them. We make use of an existing algorithm to describe nonparametrically the dynamics of the process in terms of its lagged realizations and of a latent variable, its conditional mean. The estimation devices needed to effectively apply the algorithm to our dataset are presented in this part of the work.