du fait de l'explosion actuelle des télécommunications, les opérateurs sont victimes d'une crise de croissance les obligeant à installer toujours plus de relais, à découper les cellules (zone de couverture d'un relais) en micro-cellules dans les grandes villes, afin de faire face à la demande toujours grandissante de communications. Les concepteurs des nouveaux réseaux de transmission sont donc constamment à la recherche d'une utilisation plus efficace des ressources disponibles

En la última década, la cooperación entre usuarios ha generado un gran interés por la posibilidad de mejorar la velocidad de transmisión en las redes de comunicaciones inalámbricas. El objetivo es formar un array con las antenas de todos los dispositivos y, de esta forma, aplicar técnicas de procesado espacio-temporal. El esquema de cooperación más sencillo es el canal con relays: todos los terminales que escuchen una comunicación entre dos puntos pueden ayudar a la fuente retransmitiendo lo que hayan recibido.En un sistema realista, los relays no disponen de información sobre el canal en trasmisión. En este escenario, los códigos espacio-temporales (STC, del inglés space-time coding) son la alternativa más eficiente para aprovechar la diversidad introducida por los relays. Sin embargo, los STC clásicos están diseñados para un número limitado y fijo de antenas transmisoras y no se adaptan bien a sistemas cooperativos donde el número de relays puede ser elevado y, sobretodo, puede variar en el tiempo, según los usuarios entren o salgan de la red. El problema principal es la necesidad de usar un código nuevo cada vez que cambie la configuración de la red, generando un importante tráfico de señalización.Esta tesis analiza un código espacio-temporal a bloques de dispersión lineal (LD-STBC, del inglés linear-dispersion space-time block coding), aleatorio y distribuido: a cada relay se le asigna una matriz aleatoria que aplica una transformación lineal al vector que contiene los símbolos de la fuente. Cada matriz se genera de forma independiente y sin ninguna relación con el número de usuarios involucrados. De esta manera, el número de nodos puede variar sin necesidad de modificar los códigos existentes.La forma más intuitiva de construir matrices de dispersión lineal independientes es que sus elementos sean variables aleatorias independientes e idénticamente distribuidas (i.i.d.). Por esta razón, se estudia primero la eficiencia espectral obtenida por este tipo de LD-STBC. Es importante remarcar que la eficiencia espectral es una cantidad aleatoria, ya que es una función de los códigos aleatorios anteriormente descritos. Sin embargo, cuando las dimensiones de las matrices crecen infinitamente pero manteniendo constante la tasa del código (relación entre número de símbolos de la fuente sobre el número de símbolos de los relays), la eficiencia espectral converge rápidamente hacia una cantidad determinista. Este resultado se demuestra usando la teoría de las matrices aleatorias. Por esta razón, el sistema se analiza aproximando la eficiencia espectral con su limite. Por ejemplo, la comparación con el canal directo entre fuente y destino permite definir unas condiciones suficientes en donde el sistema con relays es superior a la comunicación punto a punto.Posteriormente se debe analizar la probabilidad de outage, es decir la probabilidad de que, debido a la baja calidad del canal, la eficiencia espectral sea menor que la velocidad de transmisión solicitada por el sistema. Como ya se ha mencionado anteriormente, los relays se introducen para aumentar la diversidad del canal y, con ella, el número de caminos independientes entre la fuente y el receptor, reduciendo la probabilidad de outage. Para los LD-STBC i.i.d. las prestaciones en términos de outage dependen del tipo de relay (amplify and forward o decode and forward) y son función de la tasa del código, que debe ser cuidadosamente elegida para maximizar el orden de diversidad sin desperdiciar demasiados recursos.Finalmente, en el último capítulo de la tesis se considera otro tipo de LD-STBC, distinto del i.i.d. analizado hasta ahora. En este caso, las matrices de dispersión lineal siguen siendo independientes la una de la otra pero se añade la restricción de que cada una tenga columnas (o filas, según la tasa del código) ortogonales. Así, se consigue que el código siga siendo flexible con respecto a las variaciones en el número de usuarios, pero su estructura permite reducir la interferencia generada por cada relay, como se puede notar comparando su eficiencia espectral con la eficiencia espectral obtenida por el código i.i.d. Cabe destacar que el análisis asintótico de estos códigos (llamados isométricos) se basa en herramientas matemáticas más sofisticadas que las anteriores y, por lo tanto, es necesario un estudio más profundo para poder entender cómo se comporta en términos de outage. / In the last decade, cooperation among multiple terminals has been seen as one of the more promising strategies to improve transmission speed in wireless communications networks. Basically, the idea is to mimic an antenna array and apply distributed versions of well-known space-diversity techniques. In this context, the simplest cooperative scheme is the relay channel: all the terminals (relays) that overhear a point-to-point communication between a source and a destination may decide to aid the source by forwarding (relaying) its message.In a mobile system, it is common to assume that the relays do not have any information about the channel between them and the destination. Under this hypothesis, the best solution to exploit the diversity offered by multiple transmitting antennas is to use space-time coding (STC). However, classical STC's are designed for systems with a fixed and usually low number of antennas. Thus, they are not suitable for relaying in most mobile communications systems where the number of terminals is potentially large and may vary as users join or leave the network. For each new configuration, a new code has to be chosen and notified to the relays, introducing a set-up overhead of signaling traffic.In this dissertation we will propose and analyze a randomized distributed linear-dispersion space-time block code (LD-STBC): each relay is assigned a specific matrix which linearly transforms (left-multiplies) the column vector of source symbols. Each matrix is independently generated and does not depend on the total number of transmitters, which can thus change without interrupting data transmission for a new code--relay assignment.The more intuitive way to build independent linear-dispersion matrices is to fill them with independent and identically distributed (i.i.d.) random variables. Therefore, we will first consider these i.i.d. codes and characterize the resulting spectral efficiency. In order to analyze the performance achieved by the system, we consider a large-system analysis based on random matrix theory. We will show that the random spectral efficiency (function of the random linear-dispersion matrices) converges almost surely to a deterministic quantity when the dimensions of the code grow indefinitely while keeping constant the coding rate. Since convergence is very fast, the random spectral efficiency will be approximated by the deterministic limit in the subsequent analysis. By comparison with the direct link, sufficient conditions are derived for the superiority of relaying.Next, we will analyze the outage probability of the system, that is the probability that the spectral efficiency falls below a given target rate due to channel fading. The main purpose of diversity techniques is to introduce alternative paths from the source to the destination, so that data transmission does not fail when the direct link undergoes deep fading. We will show that the diversity behavior of LD-STBC relaying mainly depends on both the coding rate and the relaying strategy (amplify and forward or decode and forward). It is then important to choose the coding rate that maximizes the diversity order without wasting too many resources.To conclude the dissertation, we will consider a different code based on independent isometric Haar-distributed random linear-dispersion matrices. Thenew code maintains the flexibility of the previous one with respect to variations in the number of relays. However, the more complex structure of the codes allows a noticeable reduction of the interference generated by the relays. Unfortunately, isometric codes also require more sophisticated mathematical tools for their asymptotic analysis. For this reason, we simply introduce the problem by showing that it is possible to have some spectral-efficiency gain with respect to i.i.d. codes. The outage-probability analysis requires a more thorough understanding and will be the subject of future work.

Τη σημερινή εποχή γινόμαστε καθημερινά μάρτυρες μίας ταχέως αναπτυσσόμενης αγοράς, που δεν είναι άλλη από αυτή των κινητών πολυμεσικών εφαρμογών, όπως του Mobile TV και του Mobile Streaming. Υπηρεσίες όπως αυτές έχουν ή αναμένεται να έχουν υψηλή διείσδυση στη βιομηχανία της κινητής πολυμεσικής επικοινωνίας. Για να ικανοποιηθούν οι απαιτήσεις αυτών των υπηρεσιών για υψηλές ταχύτητες μετάδοσης, ο οργανισμός 3rd Generation Partnership Project (3GPP) ανέπτυξε το Long Term Evolution Advanced (LTE-A), μία τεχνολογία η οποία αποτελεί την εξέλιξη των κινητών τηλεπικοινωνιακών τεχνολογιών 3ης γενιάς. Το LTE-A χρησιμοποιεί την τεχνολογία Orthogonal Frequency Division Multiple Access (OFDMA). Η συγκεκριμένη τεχνολογία μπορεί να προσφέρει νέες υψηλής χωρητικότητας ευρυζωνικές εφαρμογές και υπηρεσίες, ενώ παρέχει αποτελεσματική, από πλευράς κόστους, καθολική κάλυψη. Επιπλέον, ο οργανισμός 3GPP εισήγαγε την τεχνολογία Multimedia Broadcast / Multicast Service (MBMS), ως μέσο πανεκπομπής και πολυεκπομπής πληροφοριών στους χρήστες κινητών, με το Mobile TV να είναι η κύρια υπηρεσία που παρέχεται. Η υποδομή του LTE-A προσφέρει στο MBMS την επιλογή να χρησιμοποιήσει ένα uplink κανάλι για την αλληλεπίδραση μεταξύ της υπηρεσίας και του χρήστη, η οποία στα συνήθη δίκτυα πανεκπομπής δεν είναι απλό θέμα. Στο πλαίσιο των LTE-A συστημάτων, το MBMS έχει εξελιχθεί σε e-MBMS (το "e-" αντιστοιχεί στη λέξη evolved, δηλαδή εξελιγμένο). Αυτό θα επιτευχθεί μέσα από την αυξημένη απόδοση της ασύρματης διεπαφής που περιλαμβάνει μία νέα τεχνολογία μετάδοσης που ονομάζεται MBMS over Single Frequency Network (MBSFN). Κατά τη λειτουργία του MBSFN, τα MBMS δεδομένα μεταδίδονται ταυτόχρονα μέσω του αέρα από πολλαπλά κελιά τα οποία είναι αυστηρά χρονο-συγχρονισμένα. Το σύνολο των κελιών που λαμβάνουν αυτά τα δεδομένα, καλείται MBSFN περιοχή. Δεδομένου ότι οι MBSFN μεταδόσεις ενισχύουν σημαντικά το λόγο σήματος προς παρεμβολή και θόρυβο, μπορούν να οδηγήσουν σε σημαντικές βελτιώσεις στη φασματική απόδοση σε σύγκριση με την πολυεκπομπή μέσω των συστημάτων 3ης γενιάς. Αυτό είναι εξαιρετικά επωφελές στα όρια των κελιών, όπου οι μεταδόσεις (που στα συστήματα 3ης γενιάς, όπως το Universal Mobile Telecommunications System - UMTS, θεωρούνται ως παρεμβολή) μεταφράζονται σε χρήσιμη ενέργεια σήματος και ως εκ τούτου η ισχύς του λαμβανόμενου σήματος είναι αυξημένη, ενώ την ίδια στιγμή η ισχύς παρεμβολής μειώνεται σε μεγάλο βαθμό. Για να αξιοποιηθούν πλήρως τα πλεονεκτήματα της τεχνολογίας MBSFN και να βελτιωθεί η φασματική απόδοση, θα πρέπει να επιλεχθεί με προσοχή το σχήμα διαμόρφωσης και κωδικοποίησης για τη μετάδοση των δεδομένων. Η σχέση μεταξύ της απόδοσης του MBSFN και της επιλογής του σχήματος διαμόρφωσης και κωδικοποίησης έχει μελετηθεί διεξοδικά σε προηγούμενες ερευνητικές εργασίες. Ωστόσο οι περισσότερες (αν όχι όλες) από τις εργασίες αυτές επικεντρώνονται μόνο στην πλευρά των χρηστών και ως εκ τούτου δεν μπορούν να θεωρηθούν πλήρεις. Μερικές φορές ο στόχος του παρόχου μπορεί να είναι η μεγιστοποίηση της φασματικής απόδοσης σε όλους τους χρήστες της τοπολογίας ή η παροχή της υπηρεσίας σε όλους τους χρήστες ανεξάρτητα από τις συνθήκες που βιώνουν. Επίσης, οι περισσότερες από αυτές τις εργασίες καθορίζουν το σχήμα διαμόρφωσης και κωδικοποίησης κατά τις MBSFN μεταδόσεις εξετάζοντας μόνο την περίπτωση της μετάδοσης από ένα πομπό σε ένα δέκτη και δεν εξετάζουν τα οφέλη που μπορούν να προσφέρουν οι τεχνικές Multiple Input Multiple Output (MIMO) στη συνολική απόδοση του συστήματος. Ο στόχος της παρούσας εργασίας είναι να επεκτείνει τις προηγούμενες ερευνητικές εργασίες και, επιπλέον, να προτείνει μια λύση στο πρόβλημα της επιλογής του σχήματος διαμόρφωσης και κωδικοποίησης. Προς την κατεύθυνση αυτή, αναλύουμε πρώτα μία διαδικασία τριών βημάτων η οποία επιλέγει το σχήμα διαμόρφωσης και κωδικοποίησης και υπολογίζει τη φασματική απόδοση στην περίπτωση ενός μόνο χρήστη. Στη συνέχεια, ακολουθεί η γενίκευση της υπόθεσης ενός χρήστη και προτείνονται τρεις προσεγγίσεις που επιλέγουν το σχήμα κωδικοποίησης για την μετάδοση των MBSFN δεδομένων σε σενάρια πολλαπλών χρηστών. Οι προσεγγίσεις αξιολογούνται για τρεις διαφορετικούς τρόπους μετάδοσης, έτσι ώστε να εξεταστεί η επίδραση των τεχνικών MIMO στην επιλογή σχήματος διαμόρφωσης για διαφορετικές κατανομές χρηστών. Τα αποτελέσματα της αξιολόγησης δείχνουν ότι, ανάλογα με το στόχο που έχει θέσει ο πάροχος (π.χ. μεγιστοποίηση της φασματικής απόδοσης ή επίτευξη μίας συγκεκριμένης τιμής φασματικής απόδοσης) κάθε προσέγγιση θα μπορούσε να οδηγήσει σε βελτιωμένη απόδοση. / Today we are witnesses of a rapidly increasing market for mobile multimedia applications, such as Mobile TV and Mobile Streaming. Services like these have or are expected to have high penetration in the mobile multimedia communications industry. In order to confront such high requirements for services that demand higher data rates, the 3rd Generation Partnership Project (3GPP) developed the Long Term Evolution Advanced (LTE-A) technology which constitutes the evolution of the 3rd Generation (3G) mobile telecommunications technologies. LTE-A utilizes Orthogonal Frequency Division Multiple Access (OFDMA). This radio technology is optimized to enhance networks by enabling new high capacity mobile broadband applications and services, while providing cost efficient ubiquitous mobile coverage. In addition, 3GPP has introduced the Multimedia Broadcast/Multicast Service (MBMS) as a means to broadcast and multicast information to mobile users, with Mobile TV being the main service offered. LTE-A infrastructure offers to MBMS an option to use an uplink channel for interaction between the service and the user, which is not a straightforward issue in usual broadcast networks. In the context of LTE-A systems, the MBMS will evolve into e-MBMS (“e-” stands for evolved). This will be achieved through the increased performance of the air interface that will include a new transmission scheme called MBMS over Single Frequency Network (MBSFN). In MBSFN operation, MBMS data are transmitted simultaneously over the air from multiple tightly time-synchronized cells. A group of those cells, which are targeted to receive these data, is called MBSFN area. Since the MBSFN transmission greatly enhances the Signal to Interference plus Noise Ratio (SINR), the MBSFN transmission mode leads to significant improvements in Spectral Efficiency (SE) in comparison to multicasting over 3G systems. This is extremely beneficial at the cell edge, where transmissions (which in 3G systems, like Universal Mobile Telecommunications System - UMTS, are considered as inter-cell interference) are translated into useful signal energy and hence the received signal strength is increased, while at the same time the interference power is largely reduced. In order to fully exploit the benefits of MBSFN and to improve its performance in terms of SE, the Modulation and Coding Scheme (MCS) for the transmission of the data should be carefully selected. The relationship between MBSFN performance and MCS selection has been thoroughly studied in previous research works; however most (if not all) of these works focus only on the users’ side and therefore may not be sufficient. Sometimes the operator’s goal may be the maximization of the SE over all users of the topology or the provision of the service to all the users irrespectively of the conditions that they experience. In addition, most of these works determine the MCS scheme for MBSFN considering only the case of single antenna transmissions and they do not examine the benefits that Multiple Input Multiple Output (MIMO) transmissions may offer on the overall performance. The goal of this thesis is to extend the previous research works and, furthermore, to tackle the problems addressed. To this direction, we first analyze a 3-step procedure that selects the MCS and calculates the SE in the case of a single user. Then, we generalize the single-user case and we propose three approaches that select the MCS for the delivery of the MBSFN data in multiple-users scenarios. The approaches are evaluated for three different transmission modes, so as to examine the impact of multiple antennas techniques on the MCS selection, and for different users’ distributions. The evaluation results indicate that depending on the target that the operator may set (i.e. SE maximization or achievement of a specific SE) each approach could lead to improved performance.

In this study, a dynamic resource management and subscriber profile prediction scheme for mobile communication systems is presented. The aim is to achieve a high level of QoS for both handoff calls and new calls, while at the same time to improve the utilization of wireless network resources. The simultaneous satisfaction of these two actually conflicting interests will be thanks to two major key features. First, it will be due to the individual subscriber profile based prior information about handoff reservation requests that are provided by the mobile terminals. This information is based on the cell transition probabilities calculated by the mobile itself using collected information during past operations. Second, it will be due to a two-way approach implemented in the resource management processes. The two-way approach controls both the amount of reserved radio channels and the new call admission in a dynamic way, depending on the subscriber mobility and network traffic conditions.

Increasing energy consumption is a result of the rapid growth in cellular communication technologies and a massive increase in the number of mobile terminals (MTs) and communication sites. In cellular communication networks, energy efficiency (EE) and spectral efficiency (SE) are two of the most important criteria employed to evaluate the performance of networks. A compromise between these two conflicting criteria is therefore required, in order to achieve the best cellular network performance. Fractional frequency reuse (FFR), classed as either strict FFR or soft frequency reuse (SFR), is an intercell interference coordination (ICIC) technique applied to manage interference when more spectrum is used, and to enhance the EE. A conventional cellular model's downlink is designed as a reference in the presence of inter-cell interference (ICI) and a general fading environment. Energy-efficient cellular models,such as cell zooming, cooperative BSs and relaying models are designed, analysed and compared with the reference model, in order to reduce network energy consumption without degrading the SE. New mathematical models are derived herein to design a distributed antenna system (DAS), in order to enhance the system's EE and SE. DAS is designed in the presence of ICI and composite fading and shadowing with FFR. A coordinate multi-point (CoMP) technique is applied, using maximum ratio transmission (MRT) to serve the mobile terminal (MT), with all distributed antenna elements (DAEs), transmit antenna selection (TAS) being applied to select the best DAE and general selection combining (GSC) being applied to select more than one DAE. Furthermore, a Cloud radio access network (C-RAN) is designed and analysed with two different schemes, using the high-power node (HPN) and a remote radio head (RRH), in order to improve the EE and SE of the system. Finally, a trade-off between the two conflicting criteria, EE and SE, is handled carefully in this thesis, in order to ensure a green cellular communication network.

Cette thèse se concentre sur l’amélioration conjointe de l'efficacité spectrale et de l'efficacité en puissance de schémas de transmission par satellite. L’émergence de nouveaux services et l'augmentation du nombre d’acteurs dans le domaine nécessitent de disposer de débits de plus en plus importants avec des ressources de plus en plus limitées. Les progrès réalisés ces dernières années sur la technologie embarquée et dans le domaine des communications numériques permettent de considérer des schémas de transmission à plus haute efficacité spectrale et en puissance. Cependant, l’enjeu majeur des schémas de transmission proposes actuellement reste de rentabiliser les ressources disponibles. L’étude développée dans cette thèse explore les possibilités d’amélioration conjointe de l’efficacité spectrale et de l’efficacité en puissance en proposant la combinaison de la modulation Cyclic Code-Shift-Keying (CCSK), dont l’efficacité en puissance augmente avec l’élévation du degré de la modulation, avec une technique de multiplexage par codage de type Code-Division Multiplexing (CDM) pour pallier la dégradation de l’efficacité spectrale liée à l’étalement du spectre induit par la modulation CCSK. Deux approches basées sur l’utilisation de séquences de Gold de longueur N sont définies: Une approche multi-flux avec un décodeur sphérique optimal en réception. La complexité liée à l’optimalité du décodeur conduit à des valeurs d'efficacité spectrale limitées mais l’étude analytique des performances, vérifiée par des simulations, montre une augmentation de l'efficacité en puissance avec l'efficacité spectrale. Une approche mono-flux justifiée par l’apparition de redondance dans les motifs résultant du multiplexage des séquences. L’approche mono-flux propose des valeurs d’efficacité spectrale équivalente aux schémas retenus dans le standard DVB-S2 avec une amélioration de l’efficacité en puissance à partir d’un certain seuil de rapport signal à bruit par rapport à ces schémas. Par la suite, l'étude porte sur la transposition de plusieurs symboles de modulation sur les porteuses d’un système OFDM et sur les bénéfices et avantages d’une telle approche. Elle se conclut sur l’apport d’un codage canal basé sur des codes par bloc non binaires Reed-Solomon et LDPC. La forme d’onde proposée offre des points de fonctionnement à haute efficacité spectrale et haute efficacité en puissance avec des perspectives intéressantes. Dans le contexte actuel, son application reste limitée par ses fluctuations d’amplitude mais est envisageable dans un contexte de transmission multiporteuse, comme attendu dans les années à venir. / This thesis focuses on jointly improving the spectral efficiency and the power efficiency of satellite transmission schemes. The emergence of new services and the increasing number of actors in this field involve higher transmission rates with increasingly limited resources. Recent progress in the embedded technologies and in digital communications offered to consider transmission schemes with higher spectral and power efficiency. Nevertheless, the major current challenge consists in making efficient use of resources. The study developed in this thesis explores the possibilities of jointly improving the spectral and power efficiency by offering a combination of the Cyclic-Code-Shift Keying modulation (CCSK), which power efficiency increases with the degree of modulation, with a multiplexing technique such as Code-Division Multiplexing (CDM) to offset the deterioration on the spectral efficiency due to the spread spectrum induced by CCSK. Two approaches based on the use of Gold sequences of length N are defined : A multi-stream approach with an optimal receiver implemented through sphere decoding. The complexity due to the receiver optimality leads to limited spectral efficiencies but the study of performance, confirmed by simulations, shows an increase in power efficiency with spectral efficiency. A single-stream approach justified by the appearance of redundancy in the patterns following the sequences multiplexing. The single-stream approach offers spectral efficiencies equivalent to the adopted schemes in the DVB-S2 standard, with improved power efficiency from a certain level of signal to noise ratio compared to those schemes. Subsequently, the study focuses on the implementation of several modulation symbols on the subcarriers of an OFDM modulator and the benefits and advantages of such an approach. It concludes with the contribution of channel coding based on nonbinary block codes such as Reed-Solomon and LDPC codes. The proposed waveform offers operating points with high spectral efficiency and high power efficiency with attractive perspectives. In the current context, its application is limited by its amplitude fluctuations but is possible in a multicarrier transmission context, as expected in the years to come.

Le travail de recherche présenté dans cette thèse est basé sur l’étude du protocole de communication par satellite appelé « Enhanced Spread Spectrum Aloha » (E-SSA) et créé par l’Agence Spatiale Européenne (ESA) en 2009. Il s’agit d’une technique qui permet à un terminal d’envoyer des messages courts par satellite de manière totalement asynchrone. L’intérêt est de maximiser l’efficacité spectrale tout en gardant une complexité minimale du terminal grâce aux techniques d’étalement de spectre et d’annulation d’interférences (SIC). Alors qu’au départ le protocole E-SSA avait pour but d’apporter un support physique pour des applications destinées à des terminaux mobiles en bande S, les développements se sont récemment étendus à une gamme de terminaux fixes dans des bandes de fréquence différentes (C, Ku ou Ka). À partir de ces deux modèles d’utilisation de l’E-SSA, ce travail de recherche se concentre sur la mise en place de solutions techniques permettant d’augmenter significativement la capacité globale du système. Cela est possible à travers une optimisation des différents paramètres du protocole et la conception d’algorithmes de gestion des puissances des terminaux. / The work presented in this thesis focuses on the satellite communication protocol named “Enhanced Spread Spectrum Aloha” (E-SSA) and developed in 2009 by the European Space Agency (ESA). The E-SSA structure, based on fully asynchronous random access, perfectly copes with low-duty cycle bursty transmissions adapted to satellite messaging. High performance in terms of spectral efficiency is possible by exploiting iterative successive interference cancellation (SIC) while keeping minimal complexity at the terminal side. If the first exploitation of the protocol concerned mobile applications in S-band, it has been extended to fleets of fixed terminals in other frequency bandwidths (C, Ku or Ka). Starting from these two scenarios, this thesis proposes different innovations to further boost the E-SSA global system capacity. These technical solutions particularly include a fine optimization of protocol parameters and new power control algorithms.

Este trabalho tem como cenário as redes ópticas de próxima geração, por onde trafegarão supercanais flexíveis, sincronizados e modulados a taxas variáveis que podem chegar à ordem de Terabit por segundo. Mais especificamente, focaliza o supercanal óptico gerado a partir de um único laser (laser semente) composto por portadoras ortogonais entre si, travadas em frequência e moduladas de maneira síncrona. Tal arranjo constitui um sistema de transmissão conhecido como OFDM óptico (optical orthogonal frequency division multiplexing, O-OFDM). Este esquema não requer banda de guarda entre canais, o que proporciona uma melhor eficiência espectral, se mantidas as condições de ortogonalidade mútua, quando comparado à técnica Nyquist WDM (Nyquist wavelength division multiplexing, N-WDM), usualmente reconhecida como base para os sistemas de próxima geração. Muitos são os desafios a serem vencidos antes que a técnica O-OFDM possa ser efetivamente implantada comercialmente e esta tese busca, através de um estudo de seus princípios de funcionamento e módulos constituintes, elencar os principais obstáculos e as possíveis maneira de superá-los. Sem pretender ser exaustivo em termos de tecnologias disponíveis para alcançar este fim, o objetivo é propor novas configurações de subsistemas e arquitetura de nó para o transmissor, o nó intermediário e o receptor coerente, capazes de executar, de forma relativamente simples no domínio óptico, as principais funcionalidades de uma transmissão ponto a ponto com nós deriva/insere intermediários. Através de simulações sistêmicas e demonstrações experimentais, como prova de conceito, dois cenários são abordados: análise do desempenho numa transmissão ponto a ponto, e operação em rede, com derivação e inserção de canal em nós intermediários. Ao final, através de uma análise qualitativa, é feita uma estimativa de componentes e subsistemas necessários para tornar a transmissão de sinais O-OFDM implementável por tecnologias de fotônica integrada que atenda, com eficiência espectral e economia de energia, a sempre crescente demanda de capacidade em sistemas de transmissão óptica. / This work focuses on the scenario of next generation optical networking, where flexible optical superchannels will propagate modulated at variable rates that can reach terabits per second. More specifically, it focuses on the optical superchannel generated from a single laser (seed laser) composed of orthogonal carriers, which are frequency-locked and synchronously modulated. Such arrangement constitutes a transmission system known as optical orthogonal frequency division multiplexing (O-OFDM). This scheme does not require guard band between channels, which provides a better spectral efficiency, if the conditions of mutual orthogonality are maintained, when compared to the Nyquist wavelength dividing multiplexing (N-WDM) technique, usually recognized as the basis for the next generation systems. There are many challenges to overcome before O-OFDM technique can be effectively deployed commercially and this thesis seeks, through a study of its operating principles and constituent modules, to identify the main obstacles and the possible ways of overcoming them. Without intending to be exhaustive in terms of available technologies to achieve this aim, the objective is to propose new configurations of subsystems and node architecture for the transmitter, the intermediate node and the coherent receiver, able to perform in the optical domain, in a relatively simple way, the main features a point-to-point transmission with nodes drifting/inserting intermediates. Through systemic simulations and some experimental demonstrations, as proof of concept, two scenarios are addressed: performance analysis in a point-to-point transmission, and network operation, with channel derivation and insertion at intermediate nodes. At the end, through a qualitative analysis, an estimate of components and subsystems is made to make the transmission of O-OFDM signals implementable by integrated photonics technologies that meet, with spectral efficiency and energy savings, the ever increasing capacity demand in optical transmission systems.

La radio cognitive (CR) est une radio entièrement reconfigurable qui permet de changer intelligemment ses paramètres de communication en réponse à l’activité des autres réseaux radios et demandes d’utilisateur. L’objectif ultime de la CR est de permettre à l’utilisateur secondaire (SU) d’utiliser la ressource de spectre disponible sans interférer sur l’utilisateur primaire (PU) en utilisant des trous de spectre. Par conséquent, la détection du PU est l’un des défis principaux dans le développement de la CR. Par rapport aux systèmes conventionnels de communication sans fil, le système CR introduit de nouveaux problèmes d’allocation de ressource (RA) en raison de l’interférence des canaux adjacents utilisés par le SU et le PU. Dans le contexte de la CR, la plupart des efforts ont été menés sur les systèmes de CR basés sur le multiplexage par division de fréquences orthogonales (OFDM). Toutefois, la technique de l’OFDM montre quelques points faibles dans l’application à cause des remontées significatives du spectre. Les modulations multiporteuses à base de bancs de filtre (FBMC) ont été récemment proposées pour des applications de CR. Dans cette thèse, trois points importants pour le développement d’un système de CR basé sur le FBMC sont discutés.Les trois points principaux peuvent être résumés ainsi: nous examinons premièrement les problèmes de détection de spectre des signaux OFDM et FBMC en employant le détecteur de signature de cyclostationnarité (CS). En outre, nous proposons une architecture de détection multi-bande basée sur le banc de filtre polyphasé (PFB), et montrons son avantage; deuxièmement, la comparaison entre l’OFDM et le FBMC du point de vue de l’efficacité spectrale est discutée; et enfin, nous proposons un algorithme stratégique d’allocation de ressource pour les systèmes cognitifs multi-cellulaires et multi-utilisateurs.Les algorithmes proposés dans cette thèse ont été testés par simulation. Les résultats numériques prouvent que le FBMC, par opposition à l’OFDM, pourrait réaliser une efficacité spectrale plus élevée et offre un avantage attrayant dans la détection de spectre. Les contributions de cette thèse ont accru l’intérêt d’appliquer FBMC dans les systèmes de CR à l’avenir. / Cognitive Radio (CR) is a fully reconfigurable radio that can intelligently change its communicationvariables in response to network and user demands. The ultimate goal of CR is to allowthe Secondary User (SU) to utilize the available spectrum resource on a non-interfering basis to thePrimary User (PU) by sensing the existence of spectrum holes. Therefore, the detection of PU isone of the main challenges in the development of the CR technology. Moreover, compared to conventionalwireless communication systems, CR system poses new challenges to Resource Allocation(RA) problems because of the Cross-Channel Interference (CCI) from the adjacent channels used bySU to PU. In the CR context, most past efforts have been spent on Orthogonal Frequency DivisionMultiplexing (OFDM) based CR systems. However, OFDM technique exhibits some shortcomingsin application due to its significant spectrum leakage. Filter Bank based Multi-Carrier (FBMC), asanother promising Multi-Carrier Modulation (MCM) candidate, has been recently proposed for CRapplications. In this dissertation, three important issues in developing a FBMC based CR system arediscussed.The three prime issues can be summarized: we firstly survey the spectrum sensing problemsof OFDM and FBMC signals by using Cyclostationary Signature (CS) detector. Furthermore, wepropose a Polyphase Filter Bank (PFB) based multi-band sensing architecture, and argue for its advantage;secondly, the comparison of OFDM and FBMC from the spectral efficiency point of viewis discussed; and lastly, our emphasis is placed on the strategic resource allocation algorithms fornon-cooperative multi-cell CR systems.The overall proposed algorithms have been verified by simulation. Numerical results show thatFBMC, as opposed to OFDM, could achieve higher spectrum efficiency and attractive benefit inspectrum sensing. The contributions of this dissertation have heighten the interest in applying FBMCin the future CR systems.

Les modulations à phase continue (CPMs) sont des méthodes de modulations robuste à la noncohérence du canal de propagation. Dans un contexte spatial, les CPM sont utilisées dans la chaîne de transmission de télémesure de la fusée. Depuis les années 70, la modulation la plus usitée dans les systèmes de télémesures est la modulation CPFSK continuous phase frequency shift keying filtrée. Historiquement, ce type de modulation est concaténée avec un code ReedSolomon (RS) afin d'améliorer le processus de décodage. Côté récepteur, les séquences CPM non-cohérentes sont démodulées par un détecteur Viterbi à sortie dure et un décodeur RS. Néanmoins, le gain du code RS n'est pas aussi satisfaisant que des techniques de codage moderne capables d'atteindre la limite de Shannon. Actualiser la chaîne de communication avec des codes atteignant la limite de Shannon tels que les codes en graphe creux, implique deremanier l’architecture du récepteur usuel pour un détecteur à sortie souple. Ainsi, on propose dans cette étude d' élaborer un détecteur treillis à sortie souple pour démoduler les séquences CPM non-cohérentes. Dans un deuxième temps, on concevra des schémas de pré-codages améliorant le comportement asymptotique du récepteur non-cohérent et dans une dernière étape on élabora des codes de parité à faible densité (LDPC) approchant la limite de Shannon. / Continuous phase modulations (CPM) are modulation methods robust to the non-coherency of propagation channels. In a space context, CPMs are used in the communication link between the rocket and the base stations. Since the 70's, the most popular telemetry modulation is the filtered continuous phase frequency shift keying (CPFSK). Traditionally, the CPFSK scheme isconcatenated with a Reed-Solomon (RS) code to enhance the decoding process. At the receiver side, the non-coherent CPM sequences are demodulated through a hard Viterbi detector and a RS decoder. However, the RS's coding gain is no more satisfactory when directly compared to modern coding schemes enable to reach the Shannon limit. Updating the communication link to capacity achieving codes, as sparse graph codes, implies to redesign the receiver architecture to soft detector. In that respect, we propose in this study to design a trellis-based soft detector to demodulate non-coherent CPM sequences. In a second part, we will elaborate precoding schemes to improve the asymptotic behaviour of the non-coherent receiver and in a last step we will build low density parity check codes approaching the Shannon limit.