As WDM results in an ever-increasing trend of traffic concentration, any failure on a single fiber-link or a single switching node will be catastrophic. Rapid restoration can recover the affected traffic so as to make the network more robust and reliable. The conventional restoration methods are designed mostly for reconfiguring the network topology; they are not immune from service interruption. In this thesis, an effective algorithm called backup-path-wavelength rearrangement scheme is proposed to reduce the connection blocking probability in an all-optical network. The proposed scheme performs wavelength retuning on the backup paths to improve the acceptance probability for new connection requests and introduce zero service interruption to the traffic in the network. The performance evaluation indicates that the connection blocking probability can be decreased greatly by the proposed scheme. The combination of BPWR and traffic grooming can efficiently alleviate the wavelength continuity constraint.

Orientador: Nelson Luis Saldanha da Fonseca / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Computação / Made available in DSpace on 2018-08-26T19:28:46Z (GMT). No. of bitstreams: 1 Oliveira_HelderMayNunesdaSilva_M.pdf: 2081314 bytes, checksum: 14db2ede39b2af222efaee8c560f6bcd (MD5) Previous issue date: 2014 / Resumo: Em redes ópticas, falhas em enlaces e nós causam perda maciça de dados, mesmo que por curtos períodos. Assim sendo, técnicas de proteção têm sido desenvolvidos para lidar com falhas. Essa dissertação apresenta dois estudos em sobrevivência em redes ópticas. O primeiro provê sobrevivência em redes ópticas wavelength-division multiplexing no contexto de rede de múltiplas camadas, IP/MPLS sobre OTN sobre DWDM. Investiga-se em qual camada da rede multicamada, deve-se prover sobrevivência. Utiliza-se um modelo de PLI nessa investigação. Resultados indicam que o uso de sinais mais baixos de proteção na camada OTN reduz o custo total, fazendo com que a proteção da camada OTN seja mais vantajosa do que a IP/MPLS ou camada DWDM. A dissertação apresenta também estudo sobre sobrevivência em redes ópticas elásticas. São propostos algoritmos de sobrevivência baseados em p-cycles FIPP para uma única falha e duas falhas na rede. Resultados apontam que os algoritmos podem fornecer 100% de proteção, com baixo custo para redes com alta conectividade / Abstract: In optical networks, faults in links and nodes cause massive data loss, even for short periods. Thus, protection techniques have been developed to deal with failures. This dissertation present two studies in optical networks for survivability. The first one provides survivability in optical wavelength-division multiplexing networks in the context of multi-layer network. This work have investigated in which layer of the multilayer network should provide survivability. We Used a model of ILP in this investigation. Results indicate that the use of lower signals in the OTN protection layer reduces the overall cost, making the OTN protection layer more advantageous than the IP/MPLS or DWDM layer. The dissertation also presents study on survivability in elastic optical networks. Algorithms based on survivability FIPP p-cycles for a single fault and two faults on the network are proposed. Results indicate that algorithms can provide 100% protection, with low cost for networks with high connectivity / Mestrado / Ciência da Computação / Mestre em Ciência da Computação

Made available in DSpace on 2016-04-04T18:31:40Z (GMT). No. of bitstreams: 1 Paulo Cesar Barreto da Silva.pdf: 3934588 bytes, checksum: cbbb71074cd6c370e06fd00f5d3eabce (MD5) Previous issue date: 2014-02-05 / New optical wavelength division multiplexing (WDM) networks are expected to utilize multiple bit rate channels. Moreover, each individual channels may carry over 200 Gb/s and occupy a bandwidth that exceeds the 50-GHz WDM fixed grid. In this scenario, spectral efficiency becomes an important issue and new channel allocation schemes need to be considered. An attractive solution for this problem is the utilization of a WDM grid with variable channel spacing, in the so-called elastic optical network (EON) approach. The main goal of this work is to propose algorithms to solve the spectral efficiency issue in emerging optical WDM networks. Such proposals are divided in two classes. The first one consists of allocating different spectral blocks for channels with different bit rates; this is named as the spectrum block division (SBD) scheme. The second class of our proposals is based on EON schemes. In this case, not only the enhancement of a previously reported algorithm, the Maximize Total Link Spectrum Consecutiveness (MTLSC) algorithm, is considered but also a new algorithm, the Shortest Path with Maximum number of Free Frequency Slot Units (SPMFF) is proposed. Another contribution of this work is the development of an EON simulator, called EONSim, based on JAVA programming language. This simulator was properly tested and reproduced the results of literature papers within a very good accuracy. All of our results were obtained with the aid of EONSim and suggest that the proposed algorithms yield a bandwidth occupation gain that varies from 7 to 18% higher than the one provided by traditional first-fit (FF) algorithms. Such algorithms also provide a blocking probability that is 2 to 8% lower than in FF strategy. On the other hand, it is found that algorithms with higher spectral efficiency use an average number of hops that is up to 16% higher than those necessary for algorithms that do not take bandwidth into account in channel allocation. / Novas redes ?pticas de multiplexa??o por divis?o de comprimento de onda (Wavelength Division Multiplexing, WDM) podem utilizar v?rios canais com taxas diferentes de bits. Al?m disso, cada um dos canais individuais pode transportar mais de 200 Gb/s e ocupar uma largura de banda que excede a grade fixa de 50 GHz da rede WDM. Neste cen?rio, a efici?ncia espectral torna-se uma quest?o importante e novos esquemas de aloca??o de canais precisam ser considerados. Uma solu??o atrativa para este problema ? a utiliza??o de uma rede WDM com espa?amento de canal vari?vel, na abordagem chamada rede ?ptica el?stica (EON). O principal objetivo do presente trabalho ? propor algoritmos para resolver a quest?o da efici?ncia espectral em redes ?pticas WDM emergentes. Tais propostas s?o divididas em duas classes. A primeira consiste em alocar diferentes blocos de espectro para canais com diferentes taxas de bits, o que ? apontado como esquema de divis?o de blocos de espectro (Spectrum Block Division, SBD). A segunda classe ? baseada em esquemas de EON. Neste caso, n?o s? o aperfei?oamento de um algoritmo previamente descrito por WANG (2012), o algoritmo Maximize Total Link Spectrum Consecutiveness (MTLSC), ? considerado, mas tamb?m um novo algoritmo, o Shortest Path with Maximum number of Free Frequency Slot Units (SPMFF) ? proposto. Outra contribui??o deste trabalho ? o desenvolvimento de um simulador de EON, chamado EONSim, com base na linguagem de programa??o JAVA. Este simulador foi devidamente testado e foram reproduzidos os resultados de WANG (2012) dentro de uma precis?o muito boa. Todos os resultados foram obtidos com a ajuda de EONSim e sugerem que os algoritmos propostos produzem um ganho de ocupa??o de banda, que varia de 7 a 18% mais elevada do que a fornecida pelo tradicional algoritmo First Fit (FF). Tais algoritmos tamb?m proporcionam uma probabilidade de bloqueio, que ? de 2 a 8% mais baixa do que na estrat?gia FF. Por outro lado, verifica-se que os algoritmos de melhor ocupa??o espectral utilizam um n?mero m?dio de saltos at? 16% mais elevado do que os necess?rios para os algoritmos de menor caminho, que n?o levam em conta a largura de banda de atribui??o de canal.

In this study, a review of research literature has been carried out to investigate Hybrid Time and wavelength Division Multiplexing techniques with respect to reach, split ratio, and speed of the network. This was done to overcome challenges presented by Passive Optical Networks, such as the reach and the high split ratio. A comparison of different methods clarifies that Hybrid PassiveOptical Networks have a high power-budget that ensure efficient signal transmission. The Optisys software has been used to analyze the performance of the different techniques.

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.

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.

Orientador: Edson Moschim / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-06T19:40:07Z (GMT). No. of bitstreams: 1 Dante_ReinaldoGolmia_D.pdf: 5560458 bytes, checksum: a0ff22b15404f7232c1e6c38c9668fdc (MD5) Previous issue date: 2006 / Resumo: Neste trabalho, apresentam-se algumas contribuições para o estudo de redes ópticas inteligentes e transparentes, por exemplo: Redes Ópticas Comutáveis Automaticamente (ASON), no que tange aos protocolos de roteamento e atribuição de comprimentos de onda para o estabelecimento de conexões fim-a-fim em ambientes distribuídos. Os resultados deste trabalho foram obtidos através da simulação dos algoritmos de roteamento e atribuição de comprimentos de onda desenvolvidos em C++. Para que o estudo pudesse ser feito, foram adotadas algumas topologias de rede comerciais e hipotéticas para a comparação de desempenho entre os algoritmos de roteamento e atribuição de comprimentos de onda propostos e os clássicos de modo a identificar as diferenças de comportamento destes algoritmos nas redes ópticas em diversas situações, incluindo as condições de restrição de alcance dos canais ópticos e a continuidade do comprimento de onda no estabelecimento de uma conexão fim-a-fim. Os resultados mostraram que os algoritmos de roteamento e de atribuição de comprimentos de onda propostos são eficientes para o funcionamento destas redes, provendo desempenhos iguais ou superiores em relação aos demais algoritmos de roteamento e atribuição de comprimentos de onda tradicionais medidos em termos da probabilidade de bloqueio dos pedidos de conexão. Como conseqüência, estes algoritmos propostos demonstraram que são capazes de aperfeiçoar o QoS óptico nas redes ópticas inteligentes e transparentes / Abstract: In this work, we present contributions to the routing and wavelength assignment protocols in transparent and intelligent optical networks such as Automatically Switched Optical Networks (ASON) for end-to-end connection establishment on the distributed environments. The results of this work were obtained through simulation of the routing and wavelengh assignment algorithms which were developed in C++. For this study, some commercially and hypothetical network topologies were adopted to compare the performance of the proposed routing and wavelengh assignment algorithms with traditional ones in order to identify the differences of those algorithm performance. Studies were carried out in optical networks under several conditions such as the limit of optical transmission reach and wavelength-continuity constraints for end-to-end connection establishment. The results have shown that the proposed routing and wavelengh assignment algorithms are efficient for use in these networks, providing equal or higher performance than traditional routing and wavelengh assignment algorithms measured in terms of blocking probability of the connection requests. Consequently, the proposed algorithms showed that they are able to improve the optical QoS in transparent and intelligent optical networks / Doutorado / Telecomunicações e Telemática / Doutor em Engenharia Elétrica

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.

Orientador: Nelson Luis Saldanha da Fonseca / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Computação / Made available in DSpace on 2018-08-27T20:17:10Z (GMT). No. of bitstreams: 1 Moura_PedroMesquita_M.pdf: 2113385 bytes, checksum: 1ba529be35f0f2fbcb95f91c01acfa29 (MD5) Previous issue date: 2015 / Resumo: As redes ópticas com multiplexação por comprimento de onda empregam uma grade fixa de divisão do espectro, dividindo-o em grandes faixas com alta capacidade de transmissão. Apesar de este esquema atingir altas velocidades de até 100Gb/s atualmente, a demanda de tráfego está cada vez maior e novas soluções são propostas como futuro das redes ópticas. A divisão do espectro em grandes faixas pode gerar problemas de falta de flexibilidade, onde requisições com baixas demandas de tráfego subutilizam comprimentos de onda. Nesse contexto as redes ópticas elásticas emergem, buscando flexibilizar a alocação do espectro utilizando alta granularidade na divisão do espectro, de modo que as conexões utilizem tipicamente vários slots, que são a unidade de alocação de redes ópticas elásticas. Utilizando-se da tecnologia de Multiplexação por Divisão de Frequências Ortogonais (OFDM), é possível fazer com que os slots adjacentes se sobreponham ortogonalmente, sem interferência, atingindo alta eficiência de utilização do espectro. O roteamento e alocação de espectro surge neste contexto com o objetivo de alocar rotas nas redes ópticas elásticas, necessitando caminhos na rede que possuam espectro suficiente para acomodar a demanda de tráfego, e a fim de manter o sinal no domínio óptico e evitar a custosa operação de conversão opto eletrônica, é necessário manter a mesma porção do espectro alocada em todos os enlaces do caminho, problema denominado de restrição de continuidade do espectro. Os slots devem ser também adjacentes para que estes se sobreponham utilizando OFDM, problema chamado de restrição de contiguidade do espectro. Esta dissertação investiga o problema roteamento e alocação de espectro e propõe algoritmos que melhoram características da rede, como qualidade de serviço, custo operacional e eficiência energética / Abstract: Wavelength division multiplexing optical networks employ fixed grid for spectrum, with high capacity transmission slots. Although this division allows high speeds of up to 100Gbps nowadays, the traffic demand grows each year and new solutions are needed in optical networks. The high capacity fixed grid can produce problems like the sub utilization of wavelengths by requests with lower traffic demand than their capacity. In this context the elastic optical networks emerged, allowing flexible division of spectrum, in a way that connections allocate several slots, the unit of spectrum of elastic optical networks. Together with Orthogonal Frequency Division Multiplexing (OFDM), it is possible to orthogonally overlap adjacent slots, without interference, achieving higher spectrum efficiency. The routing and spectrum assignment problem aims to allocate routes and spectrum in elastic optical networks, finding for paths with enough spectrum to accommodate the traffic demand. In order to avoid the costly optoelectronic signal conversion, it is necessary to allocate the same portion of spectrum in each link of the path, problem called spectrum continuity constraint. The slots must also be allocated contiguously, in order to the overlapping with OFDM be effective, problem called spectrum contiguity constraint. This work investigate the routing and spectrum assignment problem and proposes algorithms to improve network characteristics such as quality of service, operational expenditure and energy efficiency / Mestrado / Ciência da Computação / Mestre em Ciência da Computação

The energy consumption of communication networks is continuously growing. Many energy saving approaches have been proposed at the device, system, and network level. The most promising way to address this problem is to utilize photonic technologies as much as possible thanks to their low energy consumption per bit performance. Moreover, several approaches have been proposed to further reduce the energy consumption in optical networks. One popular technique exploits low power modes (e.g., sleep or doze mode) for devices that are not used. However, sleep mode based approaches may affect the way optical connections (i.e., lightpaths) are routed, or alter the characteristics of some devices. This in turn may have a detrimental impact on crucial network/device performance parameters. In other words a green approach may introduce additional delay, change the level of resource utilization in the network, or even impact the lifetime of a device, resulting in increased network operational cost. This thesis provides a study that carefully assesses, in both access and core networks, the trade-off between the benefits of sleep-based energy-efficient schemes and their possible side-effects. In fiber access networks putting a device into sleep mode and waking it up can introduce a significant energy overhead. Already proposed energy-efficient approaches reduce this overhead by aggregating as much as possible the traffic before a transmission. However, aggregating data may cause an additional delay that in some cases might not be acceptable. This thesis investigates the trade-off between energy saving and additional packet delay in the case of a LTE backhaul network based on wavelength division multiplexing passive optical network (WDM-PON). The thesis proposes a novel energy-efficient approach based on the dozing concept able to precisely control when a transmitter needs to wake up in order to maximize the time spent in sleep mode, while assuring that packet transmissions are completed before a given deadline. The proposed scheme is also able to exploit possibly diverse traffic delay requirements to further improve energy saving performance. In optical core networks, one way to decrease the energy consumption is to minimize the number of used active devices by aggregating the lightpaths on the lowest possible number of active fiber links. Routing strategies based on this intuition are beneficial in terms of energy saving, but on the other hand may impact the network performance (e.g., blocking probability) by affecting length of the lightpaths and link occupancy distribution. This trade-off is evaluated in the thesis with the help of a specially designed routing and wavelength assignment (RWA) strategy referred to as weighted power aware lightpath routing (WPA-LR). The WPA-LR strategy permits the fine tuning between the minimization of two objectives: energy consumption and network resource (i.e., wavelength) utilization. Evaluation results confirm that energy efficiency and network performance are conflicting objectives. However, the proposed WPA-LR strategy offers energy minimization with acceptable impact on the network performance. The thesis also investigates the impact that sleep-based energy-efficient strategies have on the lifetime of a number of optical network devices, in both access and core networks. In fact, utilizing a sleep mode functionality may change the operational conditions of the device which can impact the device lifetime. This is a crucial aspect to consider because it may directly affect the network operational cost related to fault management. The thesis provides a methodology to assess under which conditions and for which devices an energy-efficient scheme may lead to overall cost benefit vs. a (possible) increase of reparation cost. It was found that in access networks and with business customers a small lifetime variation in optical line terminals (OLTs) or in optical network units (ONUs) can lead to significant cost increase that cannot be covered by the profits coming from the energy saving. In core networks erbium doped fiber amplifiers (EDFAs) are the most vulnerable devices in terms of impact on their lifetime. For this reason it was found that the usage of green routing algorithms based on putting EDFAs into sleep mode may not always be economically beneficial. In conclusion this thesis provides a different perspective on sleep mode based energy-efficient algorithms where the potential benefit in terms of energy saving is weighted against the impact of a possible degradation of the network performance and devices lifetime. On the other hand these performance degradations can be controlled and limited by the proposed algorithms. / Energikonsumtionen av kommunikationsnätverk, växer kontinuerligt. Många energibesparande åtgärder har föreslagits, såväl på komponentnivå, systemnivå och nätverksnivå. Det mest lovande sättet att hantera energibehov i kommunikationsnätverk är att utnyttja optisk teknologi så mycket som möjligt då denna har potential att ge låg energiförbrukning per bit. Det finns också ett antal föreslagna metoder för att ytterligare reducera energibehovet i optiska nätverk. En av de mest använda teknikerna bygger på att låta enheter som ej används gå ner på låg effekt (”sovläge”). Om detta används i allt för hög utsträckning kan det dock påverka hur optiska förbindelser (sk lightpaths) sätts upp eller ge förändrade egenskaper hos de aktuella komponenterna. Detta kan i sin tur ha en skadlig inverkan på de centrala nätverks- och komponentegenskaper vilket påverkar prestandan. Med andra ord kan en sådan ”grön” ansats baserad på sovläge leda till ökad fördröjning, förändring av resursutnyttjandet i nätet och till och med påverka risken för att det uppstår fel i komponenterna vilket ökar driftskostnaden för nätet. Denna avhandling fokuserar på dessa aspekter och visar upp resultat som belyser avvägningen mellan sovlägesbaserade energieffektiva strategier och deras eventuella bieffekter. Att slå av och väcka upp komponenter i optiska access-nätverk kan vara förknippat med en signifikant extra energikostnad. Tidigare föreslagna energieffektiva strategier försöker reducera denna ökade energikostnad genom att samla så mycket trafik som möjligt innan överföringen sker. Dock kan detta leda till ökad fördröjning som i vissa lägen inte är acceptabel. I avhandlingen undersöker vi avvägningen mellan energibesparingar och ökad paketfördröjning i fallet med LTE distributionsnät baserade på våglängsmultiplexerade passiva optiska nätverk (WDM-PON). Vi föreslår en ny energieffektiv ansats baserad på att enheter försätts i ”slummer-läge”. Detta gör det möjligt att med god precision bestämma när en sändare behöver väckas upp i syfte att maximera tiden i sovläget och att försäkra sig om att paketen kommer fram inom avsedd tid. Strategin är även kapabel att utnyttja differentierad fördröjning för att ytterligare förbättra energibesparingen. Ett sätt att minska energikonsumtionen i optiska distributionsnät är att minimera antalet aktiva enheter genom att exempelvis samla optiska förbindelser till ett minimalt antal aktiva fiberlänkar. Routingstrategier som utgår från denna princip är fördelaktiga ur energisynpunkt men kan å andra sidan skada nätverksprestanda (exempelvis blockeringssannolikhet) genom påverkan på förbindelselängder och annorlunda belastning av länkarna. Denna avvägning utvärderas i avhandlingen med hjälp av en specifikt utformad strategi för routing och våglängstilldelning (RWA) som vi benämner ”viktad effektmedveten optisk förbindelserouting” (WPA-LR). Denna strategi möjliggör noggrann avvägning mellan minimeringen av två kriterier: å ena sidan energikonsumtion, å andra sidan utnyttjandet av nätverksresurser (speciellt väglängdsutnyttjandet). Vår utvärdering bekräftar att energieffektivitet och nätverksprestanda står i motsatsförhållande till varandra. Dock erbjuder WPA-LR strategin minimering av energin med en acceptabel påverkan på nätverksprestanda. Slutligen undersöks i avhandlingen den påverkan som sovlägesbaserade energieffektiva strategier har på livslängden för optiska nätverkskomponenter, både i access- och i distributionsnät. Användning av sovlägesfunktion kan påverka arbetsförhållandena för en komponent, något som i sin tur kan påverka livslängden. Detta är en kritisk aspekt att ta i beaktande då det direkt kan påverka driftskostnaden kopplad till nätunderhållet. En metod ges för att utvärdera under vilka förhållanden och för vilka enheter en energieffektiv strategi kan leda till en total kostnadsfördel jämfört med en (möjlig) ökning av reparationskostnaderna. Ett resultat är att, i accessnät och för företagsanvändare, så kan även en liten variation i feluppkomst i optiska linjeterminaler (OLTs) eller optiska nätverksenheter (ONUs) leda till signifikanta kostnadsförluster vilka inte kan kompenseras genom de vinster som kan åstadkommas med energibesparingar. I distributionsnät är erbium-dopade fiberförstärkare (EDFAs) de mest utsatta enheterna vad gäller inverkan på livslängd. Genom att studera routingstrategier (ex.vis WPA-LR) har vi funnit att användningen av ”gröna” routingalgoritmer baserade på att lägga EDFAs i sovläge inte alltid är ekonomiskt fördelaktigt. Denna avhandling ger ett perspektiv på sovlägesbaserade energieffektiviseringsalgoritmer där de potentiella fördelarna vad gäller minskade driftskostnader ställs mot möjliga försämringar av nätverksprestanda och komponenters livslängd. Å andra sidan kan dessa försämringar hållas under kontroll och begränsas av den föreslagna algoritmen. / La consommation d'énergie des réseaux de communication ne cesse de croître. Ce problème fait l’objet de nombreuses approches orientées vers les économies d'énergie (écoénergétiques) au niveau des appareils (équipements) des systèmes et des réseaux. La façon la plus prometteuse de limiter l’augmentation de consommation évoquée est d'utiliser autant que possible des technologies photoniques, vu leur faible consommation d'énergie par bit. Plusieurs autres approches ont été proposées pour réduire davantage encore la consommation d'énergie dans des réseaux optiques. Une technique populaire exploite les modes de faible puissance (par exemple le mode veille) pour les appareils qui ne sont pas utilisés. Cependant, les approches basées sur le mode de veille peuvent affecter la manière dont les liaisons optiques (circuits optiques) sont acheminées, ou modifier les caractéristiques de certains appareils. Cela peut avoir un impact négatif sur les paramètres de performance des réseaux/équipements cruciaux. En d'autres termes, une approche écoénergétique peut introduire un retard supplémentaire, changer le niveau d'utilisation des ressources dans le réseau, ou même avoir un impact sur le taux d'échec d'un équipement, entraînant une augmentation des coûts d'exploitation du réseau. Cette thèse évalue attentivement, à la fois dans le réseau d'accès mais aussi dans le cœur du réseau, le compromis entre les avantages des régimes économes en énergie utilisant le mode veille et leurs effets secondaires possibles. Dans les réseaux d'accès optiques, mettre un équipement en mode veille et le réactiver peut introduire une surcharge d'énergie significative. Les approches d'économie d'énergie déjà proposées réduisent cette surcharge en regroupant autant que possible le trafic avant sa transmission. Toutefois, les données d'agrégation peuvent provoquer un retard supplémentaire qui peut ne pas être acceptable dans certains cas. Cette thèse étudie le compromis entre les économies d'énergie et un retard supplémentaire des paquets dans le cas d'un réseau backhaul LTE basé sur réseau optique passif à multiplexage en longueur d'onde (WDM-PON). La thèse propose une nouvelle approche éco énergétique. Elle développe un concept au travers duquel il est possible de contrôler avec précision quand un émetteur doit se réactiver, afin de maximiser le temps passé en mode veille tout en veillant à ce que les transmissions de paquets soient terminées en temps voulu. Le schéma proposé est également capable d’exploiter les (éventuelles) exigences diverses de retard de trafic pour améliorer encore les économies d'énergie. Dans le cœur des réseaux optiques, on peut diminuer la consommation d'énergie en minimisant le nombre d’équipements actifs utilisés pour l’acheminement des circuits optiques et le nombre de liens actifs à fibres optiques. Les stratégies de routage basées sur ce principe sont bénéfiques en termes d'économie d'énergie, mais peuvent affecter les performances du réseau (par exemple, la probabilité de blocage) en affectant la longueur des circuits optiques et la distribution d’occupation des liens. Ce compromis est évalué dans la thèse avec l'aide d’une stratégie de routage et affectation de longueur d'onde (RWA) appelée routage des circuits optiques conscient de la puissance (WPA-LR). La stratégie WPA-LR permet le réglage fin entre deux objectifs: minimiser la consommation d'énergie et minimiser l’utilisation des ressources réseau (i.e. longueur d'onde). Les résultats de l'évaluation confirment que l'efficacité énergétique et les performances du réseau ont des objectifs contradictoires. Cependant, la stratégie WPA-LR proposée permet la minimisation de l'énergie avec un impact acceptable sur les performances du réseau. La thèse étudie également l'impact que les stratégies d’économie d’énergie basées sur le mode veille ont sur la durée de vie d'un certain nombre d’équipements de réseau optique, dans les deux réseaux d'accès et de base. L'utilisation du mode veille peut en effet modifier les conditions de fonctionnement de l’équipement, ce qui peut influer sur la durée de vie de l'appareil. Ceci est un aspect crucial à considérer, car il peut affecter directement le coût opérationnel du réseau lié à la gestion des pannes. La thèse propose une méthodologie pour évaluer dans quelles conditions et pour quels dispositifs un système économe en énergie peut conduire à des avantages de coûts globaux par rapport à une (possible) augmentation des coûts de maintenance. Dans les réseaux d'accès et auprès de clients commerciaux, il a été constaté qu’une petite variation de taux d'échec dans les terminaux de ligne optique (OLT) ou dans les unités de terminaison de réseau optique (ONUs) peut conduire à des pertes financières importantes qui ne peuvent être compensées par les bénéfices provenant des économies d'énergie. Dans les cœurs de réseaux les amplificateurs à fibre dopée en erbium (EDFA) sont les équipements les plus vulnérables en termes d'impact sur leur durée de vie. Pour cette raison, l'utilisation d'algorithmes de routage écoénergétiques basé sur la mise en mode veille des EDFA peut par conséquent n’être pas toujours économiquement avantageuse. En conclusion, cette thèse fournit une perspective différente sur des algorithmes économes en énergie basés sur l’utilisation du mode veille. Leur bénéfice potentiel en termes d'économie d'énergie est comparé à l'impact d'une éventuelle dégradation d’une part de la performance du réseau et d’autre part de la durée de vie des équipements. Ces dégradations de performances peuvent être contrôlées et limitées par les algorithmes proposés. / Zużycie energii elektrycznej w sieciach komunikacyjnych stale rośnie. Do tej pory zostało zaproponowanych wiele metod oszczędzania energii na poziomie urządzeń, systemów i sieci. Najbardziej obiecującym podejściem do tego problemu jest wykorzystanie technologii optycznych, z uwagi na ich niskie zużycie energii „per bit”. Ponadto wiele różnych metod przeznaczonych dla sieci optycznych zostało przedstawionych w literaturze. Jedna z popularnych technik wykorzystuje tryb niskiego poboru energii (uśpienia) w urządzeniach, które nie są używane. Jednakże techniki wykorzystujące tryb uśpienia mogą mieć wpływ na kierowanie optycznych połączeń sieciowych (lightpaths) lub zmieniać właściwości urządzeń. Natomiast to może mieć negatywny wpływ na kluczowe parametry wydajności sieci czy urządzeń sieciowych. Innymi słowy algorytmy oszczędzające energię mogą wprowadzić dodatkowe opóźnienia, zmienić wykorzystanie zasobów sieciowych, a nawet wpływać na awaryjność urządzeń zwiekszając tym samym koszt eksploatacji sieci. Praca ta przedstawia i analizuje kompromis pomiędzy korzyściami płynącymi z energooszczędnych algorytmów opartych na trybie uśpienia, a ich ewentualnymi skutkami ubocznymi, zarówno w sieciach dostępowych, jak i szkieletowych. W przypadku optycznych sieci dostępowych proces wprowadzenia urządzenia w tryb uśpienia i jego wybudzenia może spowodować znaczący narzut energetyczny. Proponowane sposoby zmniejszenia tego narzutu agregują ruch sieciowy przed jego transmisją, Jednakże taka agregacja powoduje dodatkowe opóźnienia transmisji, które w niektórych przypadkach mogą być niedopuszczalne. Praca ta analizuje kompromis pomiędzy oszczędzaniem energii, a dodatkowymi opóźnieniami transmisji w przypadku sieci LTE-backhaul, bazowanej na technologii pasywnych sieci optycznych, opartych na multipleksowaniu z podziałem długości fali WDM-PON (Wavelength Division Multiplexing Passive Optical Network). Niniejsza praca proponuje nowatorską metodę oszczędzania energii, opartą na koncepcji drzemki (dozing), która precyzyjnie kontroluje czas wybudzania nadajnika, tak aby zmaksymalizować czas spędzony w trybie drzemki, przy zapewnieniu, że transmisja danych zostanie zakończona przed upływem wymaganego czasu. Proponowana metoda wykorzystuje również zróżnicowane wymagania maksymalnych opóźnień transmitowanych danych do dalszej poprawy wydajności energetycznej. Jednym ze sposobów zmniejszania zużycia energii w światłowodowych sieciach szkieletowych jest zredukowanie liczby aktywnych urządzeń, poprzez umiejętne kierowanie optycznych połączeń sieciowych przy użyciu już aktywnych łączy światłowodowych. Kierowanie ruchu sieciowego oparte na tym pomyśle jest korzystne z punktu widzenia oszczędzania energii, choć z drugiej strony może mieć wpływ na parametry wydajnościowe sieci (np. zwiększenie prawdopodobieństwa blokady połączeń) poprzez oddziaływanie na długość połączeń, czy zajętość łączy. Przytoczony problem jest analizowany w tej pracy za pomocą specjalnie zaprojektowanego algorytmu routingu i przypisania długości fali RWA (Routing and Wavelength Assignement), nazwanego WPA-LR (Weighted Power Aware Lightpath Routing). Algorytm WPA-LR pozwala na precyzyjną regulację pomiędzy redukcją zużycia energii i optymalizacją wykorzystania zasobów sieciowych. Wyniki wykonanej analizy problemu potwierdzają, że efektywność energetyczna i wydajność sieci to cele ze sobą sprzeczne. Jednakże proponowana strategia (WPA-LR) umożliwia kontrolę i osiągnięcie kompromisu pomiędzy zmniejszeniem zużycia energii, a pogorszeniem wydajności sieci. Praca ta bada również wpływ energooszczędnych strategii, opartych o tryb uśpienia, na trwałość optycznych urządzeń sieciowych, zarówno w sieciach dostępowych, jak i szkieletowych. Używanie trybu uśpienia może zmienić warunki pracy urządzenia, które z kolei mogą mieć wpływ na jego trwałość. Natomiast zmniejszenie trwałości urządzenia może bezpośrednio oddziaływać na koszt eksploatacji sieci związany z zarządzaniem awariami. Niniejsza praca proponuje metodologię oceny, na jakich warunkach i w przypadku których urządzeń, używanie algorytmów oszczędzania energii może prowadzić do ogólnych korzyści finansowych lub strat związanych ze wzrostem kosztu eksploatacji sieci. W pracy stwierdzono, że w sieciach dostępowych, w szczególności obsługujących klientów biznesowych, mały wpływ na awaryjność optycznych terminali ONU (Optical Network Unit) lub optycznych urządzeń dystrybucyjnych OLT (Optical Line Terminal) może prowadzić do znacznego zwiększenia kosztów, które mogą przekroczyć zyski związane z oszczędzaniem energii. W sieciach szkieletowych wzmacniacze światłowodowe EDFA (Erbium Doped Fiber Amplifier) są najbardziej wrażliwymi urządzeniami pod względem wpływu na ich awaryjność. W pracy dowiedziono, że użycie energooszczędnych algorytmów kierowania połączeń światłowodowych, opartych na wprowadzaniu EDFA w tryb uśpienia, nie zawsze jest korzystne ekonomicznie. Niniejsza praca przedstawia nowatorskie spojrzenie na energooszczędne algorytmy oparte na wprowadzaniu urządzeń w tryb uśpienia, gdzie potencjalne korzyści w zakresie oszczędzania energii są porównane ze stratami związanymi z degradacją wydajności sieci lub żywotności urządzeń sieciowych. / <p>QC 20160509</p>