A sociedade atual está vivendo uma revolução baseada na informação. A recente explosão da popularidade da internet e desregulamentação dos setores públicos de telefonia em muitos países foram os principais responsáveis pelo aumento da demanda por largura de banda. Adicionalmente, houve um aumento no volume de tráfego em função do oferecimento de serviços multimídia de banda larga. Este cresimento rápido e global por demanda de largura de banda está acelerando o desenvolvimento e a implantação de redes de comunicações ópticas empregnando sistemas multiplexados em comprimento de onda (WDM). O emprego de fibras na transmissão de informações em alta velocidade a grandes distâncias já está consolidado, e sua utilização em redes metropolitanas MAN (Metropolitan Area Network) e redes de acesso está aumentando. Verifica-se que a tendência de evolução das redes ópticas é a aproximação das fibras até o usuário final. A forma mais viável economicamente de aproximar o usuário dos benefícios das redes ópticas com grande capacidade é o emprego de redes ópticas passivas PON (Passive Optical Network) empregando o sistema WDM. Este tipo de rede apresenta a flexibilidade de suportar tanto broadcast, onde o mesmo sinal é distribuído a todos os usuários, como serviços ponto-a-ponto. Um dos requisitos mais importantes para o aumento da capacidade destes sistemas ópticos é o aumento do número de canais (comprimentos de onda). Para viabilizar este aumento do número de canais é necessária a utilização de dispositivos com alta seletividade em comprimento de onda proposto neste trabalho. Este filtro é composto pela combinação de três estruturas distintas formando um único dispositivo: fibra tipo D, duplo refletor de Bragg (DBR), e guia multicamada em configuração ARROW ( Antiresonant Reflecting Optical Waveguide). O DBR é previamente definido na fibra tipo D e esta é então colada sobre a estrutura multicamada. ) A estrutura multicamada, por sua vez, possui a configuração ARROW em virtude de suas características de seletividade em comprimento de onda. A utilização de um fibra tipo D no dispositivo proposto permite que as perdas de inserção deste dispositivo no enlace óptico sejam drasticamente reduzidas. Este trabalho realiza um projeto criterioso de um filtro em questão, buscando valores ótimos para cada um dos parâmetros que compõem este dispositivo, tais como: espessuras de camadas, índices de refração, perda de inserção, isolação entre canais, etc. Os resultados obtidos com as simulações são discutidos e uma configuração ótima para o dispositivo é proposta. / The progress in lightwave propagation based on optical fibers has provided our society with an unprecedented communication capability. The deregulation of public telephone companies together with the increasingly popularity of the internet can be pointed out as the major contributors for this huge bandwidth demand we experience today. Additionally, there was a corresponding growth in traffic volume due to wideband multimedia services. As a consequence of this fast growing demand for bandwidth in a global scale, telecom companies have accelerated the development and implementation of optical communication networks based on wavelength division multiplexing (WDM) technology. The tendency in terms of network evolution is to approximate the optical fibers as close as possible to the end user. The most economically viable way of doing so is by way of the so called Passive Optical Networks (PON) based on WDM systems. This type of network is flexible enough to support broadcast, where the same signal is distributed to all users, and point-to-point services. If more capacity is required for these systems, this can be achieved just by adding new channels (wavelengths) to it. Another important issue regarding WDM systems consists in the extraction of channels from the optical network, which can be achieved by using highly selective dropping filters such as the one proposed in this work. This filter is defined by the combination of three distinct structures defining one single device: a d-type fiber, a double Bragg reflector (DBR), and a multilayer ARROW type waveguide (Antiresonant Reflecting Optical Waveguide). The DBR is defined beforehand on the d-fiber, and then the d-fiber is literally glued on top of the multilayer waveguide. This multilayer waveguide is fabricated directly atop of a p-i-n photodetector in such a way that the three structures (d-fiber, multilayer waveguide and photodetector) now form a rigid structure. The ARROW type configuration for the multilayer waveguide and the DBR structure are chosen by virtue of their wavelength selectivity characteristics, helping to improve the device performance. The d-fiber structure, by its tum, has the great advantage of reducing drastically the insertion loss when compared to rectangular waveguide filters. In the analysis provided in this work parameters such as layer thicknesses, refractive indices, insertion loss, channel isolation, etc, and their influence on the device performance, are extensively investigated for optimization purposes. The simulated results are discussed thoroughly and an optimal structure configuration is proposed.

A sociedade atual está vivendo uma revolução baseada na informação. A recente explosão da popularidade da internet e desregulamentação dos setores públicos de telefonia em muitos países foram os principais responsáveis pelo aumento da demanda por largura de banda. Adicionalmente, houve um aumento no volume de tráfego em função do oferecimento de serviços multimídia de banda larga. Este cresimento rápido e global por demanda de largura de banda está acelerando o desenvolvimento e a implantação de redes de comunicações ópticas empregnando sistemas multiplexados em comprimento de onda (WDM). O emprego de fibras na transmissão de informações em alta velocidade a grandes distâncias já está consolidado, e sua utilização em redes metropolitanas MAN (Metropolitan Area Network) e redes de acesso está aumentando. Verifica-se que a tendência de evolução das redes ópticas é a aproximação das fibras até o usuário final. A forma mais viável economicamente de aproximar o usuário dos benefícios das redes ópticas com grande capacidade é o emprego de redes ópticas passivas PON (Passive Optical Network) empregando o sistema WDM. Este tipo de rede apresenta a flexibilidade de suportar tanto broadcast, onde o mesmo sinal é distribuído a todos os usuários, como serviços ponto-a-ponto. Um dos requisitos mais importantes para o aumento da capacidade destes sistemas ópticos é o aumento do número de canais (comprimentos de onda). Para viabilizar este aumento do número de canais é necessária a utilização de dispositivos com alta seletividade em comprimento de onda proposto neste trabalho. Este filtro é composto pela combinação de três estruturas distintas formando um único dispositivo: fibra tipo D, duplo refletor de Bragg (DBR), e guia multicamada em configuração ARROW ( Antiresonant Reflecting Optical Waveguide). O DBR é previamente definido na fibra tipo D e esta é então colada sobre a estrutura multicamada. ) A estrutura multicamada, por sua vez, possui a configuração ARROW em virtude de suas características de seletividade em comprimento de onda. A utilização de um fibra tipo D no dispositivo proposto permite que as perdas de inserção deste dispositivo no enlace óptico sejam drasticamente reduzidas. Este trabalho realiza um projeto criterioso de um filtro em questão, buscando valores ótimos para cada um dos parâmetros que compõem este dispositivo, tais como: espessuras de camadas, índices de refração, perda de inserção, isolação entre canais, etc. Os resultados obtidos com as simulações são discutidos e uma configuração ótima para o dispositivo é proposta. / The progress in lightwave propagation based on optical fibers has provided our society with an unprecedented communication capability. The deregulation of public telephone companies together with the increasingly popularity of the internet can be pointed out as the major contributors for this huge bandwidth demand we experience today. Additionally, there was a corresponding growth in traffic volume due to wideband multimedia services. As a consequence of this fast growing demand for bandwidth in a global scale, telecom companies have accelerated the development and implementation of optical communication networks based on wavelength division multiplexing (WDM) technology. The tendency in terms of network evolution is to approximate the optical fibers as close as possible to the end user. The most economically viable way of doing so is by way of the so called Passive Optical Networks (PON) based on WDM systems. This type of network is flexible enough to support broadcast, where the same signal is distributed to all users, and point-to-point services. If more capacity is required for these systems, this can be achieved just by adding new channels (wavelengths) to it. Another important issue regarding WDM systems consists in the extraction of channels from the optical network, which can be achieved by using highly selective dropping filters such as the one proposed in this work. This filter is defined by the combination of three distinct structures defining one single device: a d-type fiber, a double Bragg reflector (DBR), and a multilayer ARROW type waveguide (Antiresonant Reflecting Optical Waveguide). The DBR is defined beforehand on the d-fiber, and then the d-fiber is literally glued on top of the multilayer waveguide. This multilayer waveguide is fabricated directly atop of a p-i-n photodetector in such a way that the three structures (d-fiber, multilayer waveguide and photodetector) now form a rigid structure. The ARROW type configuration for the multilayer waveguide and the DBR structure are chosen by virtue of their wavelength selectivity characteristics, helping to improve the device performance. The d-fiber structure, by its tum, has the great advantage of reducing drastically the insertion loss when compared to rectangular waveguide filters. In the analysis provided in this work parameters such as layer thicknesses, refractive indices, insertion loss, channel isolation, etc, and their influence on the device performance, are extensively investigated for optimization purposes. The simulated results are discussed thoroughly and an optimal structure configuration is proposed.

L’évolution rapide des services et applications multimédias (Internet haut débit, 3G, LTE) a entrainé un besoin chez les clients qui contraint les opérateurs à augmenter le débit de tous les segments du réseau, y compris le réseau d’accès. Les solutions utilisant la fibre optique tendent à remplacer progressivement les liaisons câblées (cuivre ou coaxial) afin de garantir des capacités de transfert plus importantes. La fibre optique est un medium très attractif car son atténuation linéique est très faible et sa bande passante importante. Cependant la dispersion chromatique de la fibre associée au chirp des sources optiques limite la montée en débit dans les futurs réseaux d’accès optiques (débits au-delà de 10 Gb/s) NG-PON (Next Generation Passive Optical Network). Dans ce contexte, des formats de modulation à efficacité spectrale meilleure que le NRZ pourraient être retenus. L’OFDM est une solution pour accroître l’efficacité spectrale, tout en garantissant une meilleure performance et une grande robustesse face aux canaux sélectifs en fréquence comme la fibre optique. Dans ce travail de thèse, nous avons proposé de nouvelles techniques d’implémentation de l’OFDM pour le NG-PON et évalué leurs performances dans un canal IM/DD. Nous avons montré par des simulations système dans un canal optique réaliste, que les techniques New DCO, New INC-ACO et DC-ACO sont capables d’augmenter les limitations de distances de transmission imposées par la modulation NRZ-OOK (Non-Return to Zero On-Off Keying) avec l’utilisation de composants bas coût. Ainsi, nous avons montré qu’avec les méthodes «MET (Minimization E-Tight)» et Levin-Campello, les techniques New DCO et DC-ACO permettent de réaliser des débits de 10 Gb/s sur une distance de 70 km en New DCO et 55 km en DC-ACO avec un taux de partage de 1×64. Cela permet d’affirmer que l’approche New AMOFDM serait un bon candidat pour le déploiement de réseaux d’accès optiques dans les métropoles de pays en développement. / The rapid development of multimedia services and applications such as broadband Internet, 3G, LTE, has led customers to force operators to increase throughput of all network segments, including the access network. Solutions using optical fiber tend to gradually replace cable based-copper or coaxial communications to ensure larger transfer capacity. The optical fiber is a very attractive medium because its linear attenuation is very low and its bandwidth very high. However, the chromatic dispersion of the fiber associated with the chirp of the optical sources limit the rise in flowrate in future optical access networks (beyond rates of 10 Gb/s) NG-PON (Next Generation Passive Optical Network). In this context, modulation formats with higher spectral efficiency than NRZ-OOK could be selected. OFDM is a solution to increase the spectral efficiency, while ensuring a better performance and high robustness against frequency selective channel such as fiber optics. In this thesis, we proposed a new OFDM techniques implementation for NG-PON and evaluated their performance in an IM/DD channel. We showed by simulations system of a realistic optical channel, that New DCO, New INC-ACO and DC-ACO OFDM techniques are able to increase the limited transmission distances imposed by the NRZ-OOK modulation with the use of low-cost components. Thus, we showed that using the “Minimization and E-Tight (MET)”or the Levin-Campello algorithm, the New DCO and DC-ACO techniques permit to achieve data rates of 10 Gb/s with a split ratio of 1 × 64 over a distance of 70 km with New DCO and 55 km for DC-ACO. Then we conclude that the New AMOFDM approach is a good choice for the deployment of optical access networks in metropolitan cities of developing countries.

The diploma thesis is focused on design of new laboratory exercises on the topic of gigabit passive optical networks. The theoretical analysis is devoted to acquaintance with basic problematics of PON and analysis of individual PON standards. Special attention, besides GPON, XG-PON and EPON standards, is devoted to security issues in PON. The goal of practical section of the thesis is the design of four laboratory exercises. Two exercises are focused on configuring GPON and XG-PON networks in a console environment with the help of exemplary console commands. The goal of the third exercise is to hook up and configure an EPON network using web configuration interface. The fourth exercise is designed for a realization of DoS attack against a fully functioning laboratory GPON network. The objective is to jam the upstream communication bandwidth with laser transmission causing a desynchronization between active network elements.

Initialement poussées par le marché résidentiel, les évolutions du réseau d’accès optique sont aujourd’hui également stimulées par l’expansion du réseau mobile. Comme le montre le premier chapitre de ce document, l’introduction d’un multiplexage en longueur d’onde dense constitue l’une des solutions privilégiées pour permettre la montée en débit dans les réseaux d’accès optique. Dans cette thèse, l’impact de l’introduction du multiplexage en longueur d’onde dense est étudié sous trois axes :• Une prochaine étape de l’évolution des technologies pour les réseaux d’accès passerait par une hybridation entre d’une part, un multiplexage temporel (hérité des précédentes générations), et d’autre part, un multiplexage en longueur d’onde dense. Cette technologie, appelée NGPON2-TWDM, permet aujourd’hui d’envisager des débits de 40Gb/s à 80Gb/s grâce à 4 ou 8 canaux. Les difficultés liées à la stabilité de la longueur d’onde lors de l’émission de données en mode paquet dans le sens montant du lien, ainsi que les solutions associées, sont étudiées dans le second chapitre.• L’importance du marché que représente le réseau d’accès optique (aussi bien pour les clients résidentiels que pour les réseaux mobiles), induit la nécessité pour les différentes générations de technologies de coexister au sein d’une même infrastructure. Du fait des fortes puissances optiques en jeu et des plages spectrales allouées à chaque technologie, cette coexistence peut induire des interactions entre technologies par émission Raman stimulée, dont le principe et les impacts sont décrits dans le troisième chapitre.• Enfin, la quatrième partie de ce document est dédiée à l’étude des limites et potentialités de la technologie self-seeded pour le multiplexage en longueur d’onde dense en bande O, capable de stabiliser automatiquement et passivement la longueur d’onde d’émission de chacun des émetteurs du système. / Initially led by the residential market, today’s optical access network evolutions are stimulated by mobile network expansion. As shown in the first chapter of this document, dense wavelength division multiplexing is one of the favorite solutions in order to increase optical access networks throughput. In this thesis, we propose a study of dense wavelength division multiplexing introduction according to three main topics :• Service providers and equipment suppliers have decided that the next step in residential market evolution will consist in a hybridization between, on one hand, a legacy time division multiplexing, and on the other hand, a dense wavelength division multiplexing. Named NG-PON2, this technology allows today 40Gb/s to 80Gb/s thanks to 4 to 8 channel pairs. Wavelength stability of the upstream emitter under burst mode operation, and related solutions, are studied in the second chapter.• Market importance (for both residential market and mobile networks) requires the different technologies generations to coexist on the same infrastructure. Due to the high optical power and the wavelength spans allocated to each technology, this coexistence can lead to technologies interactions by stimulated Raman scattering, as described in the third chapter.• Finally, the fourth part of this document describes the limits and potentialities of the self-seeded emitter technology for O-band dense wavelength division multiplexing, able to automatically and passively self-stabilize the wavelength of each emitter.

Ces travaux de recherches visent à étudier et à proposer de nouvelles architectures dans le contexte d'une Nouvelle Génération de Réseau d'Accès Optique Passif (NG-PON). Cette nouvelle génération d'équipements se justifie par l'augmentation des besoins en débit (descendant et/ou montant), ou d'éventuelles futures obligations réglementaires. Elle a pour fil conducteur le souci d'optimiser les coûts de déploiement de l'accès FTTH. Depuis 2010, l'organisme de pré-normalisation, le FSAN (associé à l'ITU) a prévu une action sur le thème du NG-PON qui se décline en XG-PON1 pour la première vague (débit 10 Gbit/s descendant et 2.5 Gbit/s montant) puis en un deuxième volet appelé NG-PON2 (pour une offre viable commerciale basée sur un débit de l'ordre de 1 Gbit/s) pour succéder au G-PON et au XGPONl. Des évaluations ont été menées sur la transmission optique, plus exactement au niveau de la couche physique (mesures de taux d'erreur, budget optique, portée) pour en déduire par la suite la faisabilité et la performance des systèmes. Deux thématiques ont structuré le travail de thèse sur la base de ce périmètre d'analyse : i) la recherche de solutions techniques pour optimiser le nombre de centraux " opticalisés ", ii) la recherche de solutions dans le modem client pour permettre de se connecter à des solutions de type une longueur d'onde par client. Un combineur optique sans pertes répondant au meilleur remplissage des cartes OLTs (Optical Line Termination) au Central Optique a été démontré. Ce combineur optique permet la réduction des sites NRO (Noeud de Raccordement Optique). Cette thèse s'intéresse aussi à l'ONU (Optical Network Unit), l'équipement du côté du client qui est un élément fondamental qui impacte le modèle de coût dans les réseaux d'accès. Les ONU achromatiques (dont les émetteurs sont identiques quelque soit la longueur d'onde d'émission) étudiés sont soit à base de source large à découpage spectral (" Spectrum Slicing ") soit à base de lasers accordables. Ils offrent l'avantage de la simplicité dans une architecture dont le multiplexage et l'accès multiples sont basés sur la longueur d'onde. Le format de modulation avancé de type OFDM et l'emploi de la technique de traitement de signal comme l'égaliseur sont également mis en relief dans ces travaux de thèse. Ces deux aspects correspondent à des mises en oeuvre d'outils de l'état de l'art le plus récent pour compenser les pénalités d'une transmission optoélectronique.

Fault localization is the process of realizing the true source of a failure from a set of collected failure notifications. Isolating failure recovery within the network optical domain is necessary to resolve alarm storm problems. The introduction of the monitoring trail (m-trail) has been proven to deliver better performance by employing monitoring resources in a form of optical trails - a monitoring framework that generalizes all the previously reported counterparts. In this dissertation, the m-trail design is explored and a focus is given to the analysis on using m-trails with established lightpaths to achieve fault localization. This process saves network resources by reducing the number of the m-trails required for fault localization and therefore the number of wavelengths used in the network. A novel approach based on Geographic Midpoint Technique, an adapted version of the Chinese Postman’s Problem (CPP) solution and an adapted version of the Traveling Salesman’s Problem (TSP) solution algorithms is introduced. The desirable features of network architectures and the enabling of innovative technologies for delivering future millimeter-waveband (mm-WB) Radio-over-Fiber (RoF) systems for wireless services integrated in a Dense Wavelength Division Multiplexing (DWDM) is proposed in this dissertation. For the conceptual illustration, a DWDM RoF system with channel spacing of 12.5 GHz is considered. The mm-WB Radio Frequency (RF) signal is obtained at each Optical Network Unit (ONU) by simultaneously using optical heterodyning photo detection between two optical carriers. The generated RF modulated signal has a frequency of 12.5 GHz. This RoF system is easy, cost-effective, resistant to laser phase noise and also reduces maintenance needs, in principle. A revision of related RoF network proposals and experiments is also included. A number of models for Passive Optical Networks (PON)/ RoF-PON that combine both innovative and existing ideas along with a number of solutions for m-trail design problem of these models are proposed. The comparison between these models uses the expected survivability function which proved that these models are liable to be implemented in the new and existing PON/ RoF-PON systems. This dissertation is followed by recommendation of possible directions for future research in this area.

Afin de réaliser la convergence de l'infrastructure optique des réseaux d'accès fixes et mobiles, l'objectif de cette thèse est d'étudier les solutions pour la distribution de porteuses radio natives à travers des réseaux d'accès optiques typiques. La première partie décrit les contextes et les principales propriétés physiques des réseaux d'accès optiques et radio : depuis les systèmes Fiber To The Home (FTTH) actuellement déployés et leurs évolutions attendues, en passant par l'actuel système radio Universal Mobile Telecommunications System (UMTS) jusqu'aux exigences des futurs systèmes modernes de téléphonie mobile. Ceci permet de fixer le cadre de l'environnement optique dans lequel les fonctionnalités Radio over Fiber (RoF) devront être intégrées, et d'anticiper les figures de mérite, se rapportant aux systèmes radios, auxquelles une attention particulière devra être protée durant la mise en oeuvre. La seconde partie montre l'intérêt et les possibilités de réutiliser l'infrastructure optique des réseaux d'accès fixes pour les systèmes de radio distribuée. Ensuite est proposé un inventaire des techniques RoF analogiques et numériques, ainsi qu'une discussion quant à leur intégrabilité dans des systèmes FTTH courants. La troisième partie traite du calcul formel et de simulations d'un système de transmission RoF analogique, où la liaison optique est soit passive ou optiquement pré-amplifiée, et voire amplifiée à l'émission. L'objectif est de fournir des résultats numériques en complément aux résultats pratiques (seconde moitié de la quatrième partie) pour lesquels le rapport signal à bruit est le souci majeur. Pour cela nous utilisons successivement des expressions formelles, et des résultats numériques obtenus pour des signaux sinusoïdaux et pour des signaux UMTS plus réalistes. La quatrième et dernière partie concerne les résultats expérimentaux : d'une part ceux concernant les architectures à base de récepteur APD et à portée étendue grâce à un SOA exploité de manière bidirectionnelle, et ayant pour figure de mérite principale l'EVM ; d'autre part les résultats ayant pour figure de mérite l'ACLR, qui pour les systèmes radio dans le sens descendant, est un paramètre critique, et s'avérant être plus important que l'EVM. C'est pourquoi sont présentées des architectures RoF compatibles avec les budgets PON, permettant de surmonter les non-linéarités d'ordre 3 induites par la dispersion chromatique et le chirp du laser, qui dégradent les performances de ACLR / In order to realize the convergence of the optical infrastructure of fixed and mobile access networks, the objective of this thesis is to study the solutions for distributing native radio carriers through typical optical access networks.The first Part describes the contexts and the main physical properties of the optical and radio access networks: from nowadays deployed Fiber To The Home (FTTH) systems, and their expected evolutions, to the current radio system Universal Mobile Telecommunications System (UMTS) towards the expected requirements of modern mobile radio systems. This allows to settle the optical environment in which the Radio over Fiber(RoF)-functionalities will have to be integrated, and to know on which radio systems'figures of merits to focus on when implementing it. The second Part shows the benefit and possibilities of re-using the optical infrastructure of the fixed access networks for distributed radio systems. Then a review of the analog and digital RoF techniques is proposed, and their feasibility of integration into legacy FTTH systems is discussed. The third part deals with the computing and simulations of an analog RoF-system where the optical link is either passive or optically pre-amplified, and even boosted. The goal is to provide numerical results to the practical lab. results of the second half of the fourth part where the Signal to Noise Ratio (SNR) matters. Therefore successively formal expressions, numerical results for simple 2-tone signals and more realistic UMTS signals are considered. The fourth and last part deals with the obtained practical results. These can be split into two main categories : Error Vector Magnitude (EVM)-oriented results where an Avalanche Photo-Detector (APD) is used for legacy and extended-reach PON architectures using a direction shared Semiconductor Optical Amplifier (SOA) ; and an Adjacent/Alternate Channel power Leakage Ratio (ACLR)-driven part where the focus is seton a very critical figure of merit of radio systems, especially in the downlink. The latter part turned out to be mandatory and prevailing over the initially considered EVM concerns. Hence several RoF architectures, compatible with PONs, are introduced in order to overcome the non-linearities undergone by the RoF-signals, induced by the chromatic dispersion of the PON's fiber and the laser chirp, and degrading the ACLR performances

A disponibilização de fontes ópticas multi-comprimento de onda de baixo custo certamente é um dos aspectos fundamentais para viabilizar a implantação de redes WDMPON (Wavelength Division Multiplexing-Passive Optical Network). Na literatura, são encontradas propostas de algumas tecnologias para a construção de tais dispositivos como a utilização de fontes ópticas sintonizáveis, uso de um vetor de lasers DFB (Distributed Feedback Laser) ou, até mesmo, o fatiamento espectral de uma fonte de banda larga. Contudo, essas propostas apresentam custo elevado, ou limitações de operação, tornando-se inapropriadas para a aplicação de interesse. Uma alternativa que vem sendo amplamente utilizada é o travamento óptico de diodos laser de Fabry-Pérot (FP-LD - Fabry-Pérot laser diode), realizado por meio da injeção de sinal óptico externo provindo do fatiamento espectral de uma fonte de banda larga. O travamento óptico possibilita o controle espectral do FP-LD, e o seu baixo custo torna essa fonte óptica atrativa na implantação de redes WDM. O entendimento do mecanismo de travamento torna-se, então, crucial na descrição da dinâmica de tais fontes ópticas. Desta forma, nesta dissertação, foi desenvolvido o estudo dos modelos matemáticos que descrevem a dinâmica de lasers travados, assim como foi investigado o desempenho de redes WDM-PON empregando estas fontes ópticas. / The availability of low cost multi-wavelength optical sources is certainly one of the fundamental aspects to allow the deployment of WDM-PON networks. Several technologies for the implementation of such optical sources are discussed in the literature, including tunable lasers, DFB arrays and spectral slicing of a broadband light source. However, those propositions suffer from high cost or operational limitations. One alternative is the optical injection locking of a Fabry Pérot laser diode (FPLD), in which the external optical injection signal is obtained from the spectral slicing of a broadband light source. This option has been widely investigated because the optical injection locking process allows for the FP-LD optical spectral control and wavelength selection. In addition, the low cost of such device makes this optical source an attractive solution for WDM-PON widespread deployment. In this context, the understanding of the locking mechanism becomes important in the optimization of such optical sources. Thus, in this dissertation, we conducted a study of mathematical models that describe the dynamics of locked lasers as well as investigated the performance of WDM-PON based on those optical sources.

Redes ópticas passivas são redes de acesso que utilizam fibras ópticas interligadas na topologia estrela e na configuração ponto-multiponto, sendo constituídas apenas por componentes ópticos passivos entre o terminal de linha óptica (optical line terminal - OLT) e a unidade de rede óptica (optical network unit - ONU). Como estratégia de evolução tecnológica, podem ser utilizadas para aproveitar a base instalada de cabos de cobre e cabos coaxiais na rede de acesso, servindo como cabo alimentador para uso das tecnologias digital subscriber line (DSL), cable modem e local multipoint distribution system (LMDS), e, no futuro, para disponibilizar a fibra óptica até às instalações do usuário. O principal mérito da arquitetura das redes ópticas passivas está na redução dos custos de implantação e de manutenção, pela ampliação da largura de banda disponível sem a necessidade de aumento no número de componentes ópticos ativos na rede, como lasers e amplificadores ópticos. Neste contexto, este trabalho define as redes ópticas passivas, apresenta o histórico de seu desenvolvimento, seu princípio de funcionamento e suas aplicações como forma de acesso a serviços de banda larga, identifica fornecedores visando à implementação de um teste de campo e discute as perspectivas para o futuro dessa tecnologia / Passive optical networks (PONs) are access networks which use interconnected optical fibers configured in a point to multipoint star topology having only passive optical components between the optical line terminal (OLT) and the optical network unit (ONU). As a strategy for technological evolution they can be used to take advantage of copper and coax base cables installed in the access network, serving as a feeder cable for the use of DSL, cable modem and LMDS technologies, in order to allow optical fibers to reach customers' facilities. The main merit of the passive optical networks architecture consists on the cost reduction involved on its implantation and maintenance, offering bandwidth enlargement without increasing the number of active optical components, such as lasers and optical amplifiers, in the network. In this context, this text defines PONs, presents its development history, operation principle and their applications as a way to provide broadband access services, identifies suppliers for field test implementation and discusses the perspectives for the future