Coreografias de serviços Web representam uma forma mais escalável e flexível de compor serviços do que uma abordagem centralizada como a orquestração, e seu papel na integração e comunicação de sistemas de larga escala é vital para os objetivos da SOC (Computação Orientada a Serviços) e da Internet do Futuro. Atualmente coreografias de serviços Web possuem vários desafios de pesquisa, dos quais a qualidade de serviço (QoS) e o monitoramento de coreografias de serviçosWeb são linhas importantes. O objetivo deste trabalho é propor e implementar um mecanismo de monitoramento não intrusivo de coreografias de serviços Web baseado em SLAs (Acordos de Nível de Serviço) que especificam as restrições de atributos de QoS de maneira probabilística. Esta dissertação propõe um mecanismo para coreografias de serviços Web que: (1) define requisitos de QoS; (2) especifica contratos probabilísticos sobre parâmetros de QoS usando SLA; e (3) realiza um monitoramento não intrusivo de coreografias de serviços Web para detectar violações de SLA. / Web services choreographies are a more scalable and flexible way to compose services than a centralized approach like orchestrations, and its role in the integration and communication of large-scale systems is vital for the goals of SoC (Service Oriented Computing) and Future Internet. Currently,Web services choreographies have several research challenges. From all challenges, quality of service (QoS) and monitoring of Web services choreography are important research lines. The goal of this work is to propose and implement a mechanism for non-intrusive monitoring of Web services choreography based on SLAs (Service Level Agreements) that define constraints of QoS attributes in a probabilistic way. This thesis proposes a mechanism for Web services choreographies that:(1) defines QoS requirements; (2) specifies probabilistic contracts on QoS parameters using SLAs; and (3) monitors, non-intrusively, the enactment of Web services choreographies to detect SLA violations.

Avec l'utilisation grandissante des technologies distribuées sans fil pour la modernisation des services, les déploiements d'infrastructures radio dédiées ne permettent plus de garantir des communications fiables, à grande échelle et pour un bas coût. Cette thèse vise à permettre à un opérateur de déployer une infrastructure de réseau radio pour plusieurs applications clientes de l'Internet des Objets (IoT). Nous étudions la mutualisation d'une architecture pour différents flux de trafic afin de rentabiliser le déploiement du réseau en partageant la capacité des nœuds et une large couverture. Nous devons alors garantir une Qualité de Service (QoS) différenciée pour les flux de chaque application. Nous proposons de spécifier des contrats de QoS nommés Service Level Agreements (SLA) dans le domaine de l'IoT. Ceux-ci définissent les indicateurs clés de performance (KPI) de délai de transit et de taux de livraison pour le trafic provenant d'objets connectés distribués géographiquement. Dans un second temps, nous détaillons les fonctionnalités nécessaires à la mise en œuvre des SLA sur le réseau opéré, sous la forme d'une architecture de gestion de SLA. Nous envisageons l'admission de nouveaux flux, l'analyse des performances courantes et la configuration des relais de l'opérateur. Sur la base d'une technologie robuste, multi-saut, IEEE Std 802.15.4-2015 mode TSCH, nous proposons un mécanisme d'observation de réseau permettant de vérifier les différents KPI. Nous utilisons les trames de données existantes comme support de collecte afin de réduire le surcoût en termes de ressources de communication. Nous comparons différentes stratégies de piggybacking afin de trouver un compromis entre la performance et l'efficacité de l'observation. Puis nous détaillons KAUSA, un algorithme d'allocation de ressources sous contraintes de QoS multi-flux. Nous dédions des ressources temps-fréquences ajustées saut-par-saut pour chaque message. KAUSA prend en compte les interférences, la fiabilité des liens radio et la charge attendue afin d'améliorer la répartition des ressources allouées et ainsi prolonger la durée de vie du réseau. Nous montrons les gains et la validité de nos contributions par simulation, sur la base de scénarios réalistes de trafic et d'exigences. / With the growing use of distributed wireless technologies for modern services, the deployments of dedicated radio infrastructures do not enable to ensure large-scale, low-cost and reliable communications. This PhD research work aims at enabling an operator to deploy a radio network infrastructure for several client applications, hence forming the Internet of Things (IoT). We evaluate the benefits earned by sharing an architecture among different traffic flows, in order to reduce the costs of deployment, obtaining a wide coverage through efficient use of the capacity on the network nodes. We thus need to ensure a differentiated Quality of Service (QoS) for the flows of each application. We propose to specify QoS contracts, namely Service Level Agreements (SLAs), in the context of the IoT. SLAs include specific Key Performance Indicators (KPIs), such as the transit time and the delivery ratio, concerning connected devices that are geographically distributed in the environment. The operator agrees with each client on the sources and amount of traffic for which the performance is guaranteed. Secondly, we describe the features needed to implement SLAs on the operated network, and we organize them into an SLA management architecture. We consider the admission of new flows, the analysis of current performance and the configuration of the operator's relays. Based on a robust, multi-hop technology, IEEE Std 802.15.4-2015 on TSCH mode, we provide two essential elements to implement the SLAs: a mechanism for the monitoring of the KPIs, and KAUSA, a resource allocation algorithm with multi-flow QoS constraints. The former uses existing data frames as a transport medium to reduce the overhead in terms of communication resources. We compare different piggybacking strategies to find a tradeoff between the performance and the efficiency of the monitoring. With the latter, KAUSA, we dedicate adjusted time-frequency resources for each message, hop by hop. KAUSA takes into account the interference, the reliability of radio links and the expected load to improve the distribution of allocated resources and prolong the network lifetime. We show the gains and the validity of our contributions with a simulation based on realistic traffic scenarios and requirements.

Cette thèse s'intéresse aux stratégies de transmission de contenu multimédia des réseaux sans fil. L'intérêt d'une stratégie repose sur sa capacité à gérer les ressources selon les objectifs des applications visées. On propose ainsi un schéma de transmission, qui garantit la qualité de service (QdS), quelles que soient les conditions du canal et suivant les spécificités du contenu multimédia. On exploite pour cela les diversités (spatiales, fréquentielles, etc.) du canal radio. A ce titre, l'association MIMO-OFDM apparaît comme une solution idéale pour accroitre les performances du système. On propose ainsi un système MIMO-OFDMA, qui intègre des stratégies inter-couches. Celles-ci se basent sur des techniques d'adaptation de lien pour régler dynamiquement les paramètres du système. On développe en premier un schéma inter-couches PHY 1 APPL, associé à un système MIMO-OFDM mono-utilisateur, qui transmet une vidéo H.264/SVC. L'adaptation de lien permet de définir le jeu de paramètres optimaux, qui minimise la distorsion de la vidéo reçue. On introduit ainsi une politique d'optimisation sous contraintes de puissance et de taux d'erreurs binaires. Pour tenir compte d'aspects de transmission réalistes, on propose un schéma inter-couches PHY/MAC, dédié à un système MIMO-OFDMA, multi-utilisateurs et multi-services. Ce dernier associe un algorithme d'ordonnancement, au paramétrage optimal, pour fournir au maximum d'utilisateurs des débits utiles respectant les exigences des services demandés. Enfin, on combine les deux premières solutions pour définir les paramètres optimaux minimisant la distorsion de la vidéo reçue par utilisateur, tout en garantissant la QdS des autres services. / This thesis focuses on the transmission strategies of the multimedia content in wireless networks. The advantage of a such strategy is its ability to manage the resources, according to the objectives of the targeted applications. We propose a transmission scheme, which guarantees the quality of service (QoS), depending on the channel state information and on the multimedia content specifications. We take advantage of the radio channel diversities (spatial, frequency, etc.). ln fact, the association of MIMO and OFDM techniques appears as effective solution to increase the system performance. For this reason, we propose a MIMO-OFDMA system, which considers cross-layer srtategies based on link adaptation schemes to dynamically adjust the system parameters. Firstly, we develop a PHY/APPL cross-layer strategy, dedicated to a single user of a MIMO-OFDMA system, which transmits an H.264/SVC video. The Iink adaptation scheme allows to define the optimum parameters, which minimize the end-ta-end video distortion by using an optimization algorithm under power and bit error rate constraints. To consider the realistic transmission aspects, we propose a PHY/MAC cross-layer strategy, dedicated to a multi-user, multi-service, MIMOOFDMA system. This scheme combines a scheduling policy with optimal parameterization to provide a satisfied troughput to a maximum number of users, acheiving the required services. Finally, the first two solutions are combined to determine the optimal parameters which minimize the end-ta-end distortion of the received video by user, and also acheive QoS requirements of the other services.

This work deals with the wireless Wi-Fi networks, that used unlicenced ISM frequency band 2,4 GHz. These wireless networks are very extended in these days and used mainly for the Internet connection. With the multimedia data transfers expanding it’s necessary to ensure the specific quality of service QoS in wireless networks mainly for applications, which are sensitive to delay or packets lost. The main aim of this work is to describe Wi-Fi networks based on IEEE 802.11b/g/n standards, description of QoS techniques according to IEEE 802.11e standard and description of parameters describing QoS for the VoIP services. The practical part is divided on a two parts. The problems of measuring the radio link quality and the creation of simple program for quality of wireless connection analysis are solved in the first part. The second part incluledes measuring of the lost packets number in Wi-Fi network with the commercial application AirMagnet Laptop Wireless LAN Analyzer and measuring of the signal noise influence with the help of spectrum analyzer and vector signal generator.

The thesis described the proposal for a wireless network problems. There are recorded and compared the characteristics of different wireless technologies. The possibility of voice communication securing is outlined and solutions implemented in the form of ZRTP protocol. The requirements for quality of service are discussed and the solutions to this issue are proposed. Other chapters are devoted to the actual design of specific equipment, including antennas and advance resource. The whole network is practically implemented in the mountainous countryside of eastern Wallachia. In conclusion, this work is network mate in terms of QoS (quality of service) and tested in terms of prioritization of voice communication.

This master’s thesis deals abou the transmission of voice in communications networks. The theoretical part describes criteria for optimizing voice, such as quality of service, type of service, level of service, service type, and mean opinion score. Next I describe the Internet Protocol, comparing IPv4 and IPv6, VoIP, including security, protocols and parameters necessary for transmission. Other part is about neural networks. There are basically described the neural network, Hopfield neural network and Kohenen neural network. The research is based on a comparison of the network without ensuring the quality of service and with ensuring quality of service. Then, there are compared two types of switches. Classical switch-controlled sequentially, and switch controlled by neural networks. The overall simulation program is implemented in Opnet Modeler. The conclusion deals with the creation of laboratory tasks in this program to compare the different systems of ensuring quality of service.

[ES] La presente tesis aborda el problema del encaminamiento en las redes definidas por software (SDN). Específicamente, aborda el problema del diseño de un protocolo de encaminamiento basado en inteligencia artificial (AI) para garantizar la calidad de servicio (QoS) en transmisiones multimedia. En la primera parte del trabajo, el concepto de SDN es introducido. Su arquitectura, protocolos y ventajas son comentados. A continuación, el estado del arte es presentado, donde diversos trabajos acerca de QoS, encaminamiento, SDN y AI son detallados. En el siguiente capítulo, el controlador SDN, el cual juega un papel central en la arquitectura propuesta, es presentado. Se detalla el diseño del controlador y se compara su rendimiento con otro controlador comúnmente utilizado. Más tarde, se describe las propuestas de encaminamiento. Primero, se aborda la modificación de un protocolo de encaminamiento tradicional. Esta modificación tiene como objetivo adaptar el protocolo de encaminamiento tradicional a las redes SDN, centrado en las transmisiones multimedia. A continuación, la propuesta final es descrita. Sus mensajes, arquitectura y algoritmos son mostrados. Referente a la AI, el capítulo 5 detalla el módulo de la arquitectura que la implementa, junto con los métodos inteligentes usados en la propuesta de encaminamiento. Además, el algoritmo inteligente de decisión de rutas es descrito y la propuesta es comparada con el protocolo de encaminamiento tradicional y con su adaptación a las redes SDN, mostrando un incremento de la calidad final de la transmisión. Finalmente, se muestra y se describe algunas aplicaciones basadas en la propuesta. Las aplicaciones son presentadas para demostrar que la solución presentada en la tesis está diseñada para trabajar en redes heterogéneas. / [CA] La present tesi tracta el problema de l'encaminament en les xarxes definides per programari (SDN). Específicament, tracta el problema del disseny d'un protocol d'encaminament basat en intel·ligència artificial (AI) per a garantir la qualitat de servici (QoS) en les transmissions multimèdia. En la primera part del treball, s'introdueix les xarxes SDN. Es comenten la seva arquitectura, els protocols i els avantatges. A continuació, l'estat de l'art és presentat, on es detellen els diversos treballs al voltant de QoS, encaminament, SDN i AI. Al següent capítol, el controlador SDN, el qual juga un paper central a l'arquitectura proposta, és presentat. Es detalla el disseny del controlador i es compara el seu rendiment amb altre controlador utilitzat comunament. Més endavant, es descriuen les propostes d'encaminament. Primer, s'aborda la modificació d'un protocol d'encaminament tradicional. Aquesta modificació té com a objectiu adaptar el protocol d'encaminament tradicional a les xarxes SDN, centrat a les transmissions multimèdia. A continuació, la proposta final és descrita. Els seus missatges, arquitectura i algoritmes són mostrats. Pel que fa a l'AI, el capítol 5 detalla el mòdul de l'arquitectura que la implementa, junt amb els mètodes intel·ligents usats en la proposta d'encaminament. A més a més, l'algoritme intel·ligent de decisió de rutes és descrit i la proposta és comparada amb el protocol d'encaminament tradicional i amb la seva adaptació a les xarxes SDN, mostrant un increment de la qualitat final de la transmissió. Finalment, es mostra i es descriuen algunes aplicacions basades en la proposta. Les aplicacions són presentades per a demostrar que la solució presentada en la tesi és dissenyada per a treballar en xarxes heterogènies. / [EN] This thesis addresses the problem of routing in Software Defined Networks (SDN). Specifically, the problem of designing a routing protocol based on Artificial Intelligence (AI) for ensuring Quality of Service (QoS) in multimedia transmissions. In the first part of the work, SDN is introduced. Its architecture, protocols and advantages are discussed. Then, the state of the art is presented, where several works regarding QoS, routing, SDN and AI are detailed. In the next chapter, the SDN controller, which plays the central role in the proposed architecture, is presented. The design of the controller is detailed and its performance compared to another common controller. Later, the routing proposals are described. First, a modification of a traditional routing protocol is discussed. This modification intends to adapt a traditional routing protocol to SDN, focused on multimedia transmissions. Then, the final proposal is described. Its messages, architecture and algorithms are depicted. As regards AI, chapter 5 details the module of the architecture that implements it, along with all the intelligent methods used in the routing proposal. Furthermore, the intelligent route decision algorithm is described and the final proposal is compared to the traditional routing protocol and its adaptation to SDN, showing an increment of the end quality of the transmission. Finally, some applications based on the routing proposal are described. The applications are presented to demonstrate that the proposed solution can work with heterogeneous networks. / Rego Máñez, A. (2020). Intelligent multimedia flow transmission through heterogeneous networks using cognitive software defined networks [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/160483 / TESIS

Innovations in optical communication are expected to transform the landscape of global communications, internet and datacenter networks. This dissertation investigates several important issues in optical communication such as fairness, throughput, blocking probability and differentiated quality of service (QoS). Novel algorithms and new approaches have been presented to improve the performance of optical circuit switching (OCS) and optical burst switching (OBS) for long haul, and datacenter networks. Extensive simulations tests have been conducted to evaluate the effectiveness of the proposed algorithms. These simulation tests were performed over a number of network topologies such as ring, mesh and U.S. Long-Haul, some high processing computing (HPC) topologies such as 2D and 6D mesh torus topologies and modern datacenter topologies such as FatTree and BCube. Two new schemes are proposed for long haul networks to improve throughput and hop count fairness in OBS networks. The idea is motivated by the observation that providing a slightly more priority to longer bursts over short bursts can significantly improve the throughput of the OBS networks without adversely affecting hop-count fairness. The results of extensive performance tests have shown that proposed schemes improve the throughput of optical OBS networks and enhance the hop-count fairness. Another contribution of this dissertation is the research work on developing routing and wavelength assignment schemes in multimode fiber networks. Two additional schemes for long haul networks are presented and evaluated over multimode fiber networks. First for alleviating the fairness problem in OBS networks using wavelength-division multiplexing as well as mode-division multiplexing while the second scheme for achieving higher throughput without sacrificing hop count fairness. We have also shown the significant benefits of using both mode division multiplexing and wavelength division multiplexing in real-life short-distance optical networks such as the optical circuit switching networks used in the hybrid electronic-optical switching architectures for datacenters. We evaluated four mode and wavelength assignment heuristics and compared their throughput performance. We also included preliminary results of impact of the cascaded mode conversion constraint on network throughput. Datacenter and high performance computing networks share a number of common performance goals. Another highly efficient adaptive mode wavelength- routing algorithm is presented over OBS networks to improve throughput of these networks. The effectiveness of the proposed model has been validated by extensive simulation results. In order to optimize bandwidth and maximize throughput of datacenters, an extension of TCP called multipath-TCP (MPTCP) has been evaluated over an OBS network using dense interconnect datacenter topologies. We have proposed a service differentiation scheme using MPTCP over OBS for datacenter traffic. The scheme is evaluated over mixed workload traffic model of datacenters and is shown to provide tangible service differentiation between flows of different priority levels. An adaptive QoS differentiation architecture is proposed for software defined optical datacenter networks using MPTCP over OBS. This scheme prioritizes flows based on current network state.

Along with playing an ever-increasing role in the integration of other communication networks and expanding in application diversities, the current Internet suffers from serious overuse and congestion bottlenecks. Efficient congestion control is fundamental to ensure the Internet reliability, satisfy the specified Quality-of-Service (QoS) constraints and achieve desirable performance in response to varying application scenarios. Active Queue Management (AQM) is a promising scheme to support end-to-end Transmission Control Protocol (TCP) congestion control because it enables the sender to react appropriately to the real network situation. Analytical performance models are powerful tools which can be adopted to investigate optimal setting of AQM parameters. Among the existing research efforts in this field, however, there is a current lack of analytical models that can be viewed as a cost-effective performance evaluation tool for AQM in the presence of heterogeneous traffic, generated by various network applications. This thesis aims to provide a generic and extensible analytical framework for analyzing AQM congestion control for various traffic types, such as non-bursty Poisson and bursty Markov-Modulated Poisson Process (MMPP) traffic. Specifically, the Markov analytical models are developed for AQM congestion control scheme coupled with queue thresholds and then are adopted to derive expressions for important QoS metrics. The main contributions of this thesis are listed as follows: iii ¿ Study the queueing systems for modeling AQM scheme subject to single-class and multiple-classes Poisson traffic, respectively. Analyze the effects of the varying threshold, mean traffic arrival rate, service rate and buffer capacity on the key performance metrics. ¿ Propose an analytical model for AQM scheme with single class bursty traffic and investigate how burstiness and correlations affect the performance metrics. The analytical results reveal that high burstiness and correlation can result in significant degradation of AQM performance, such as increased queueing delay and packet loss probability, and reduced throughput and utlization. ¿ Develop an analytical model for a single server queueing system with AQM in the presence of heterogeneous traffic and evaluate the aggregate and marginal performance subject to different threshold values, burstiness degree and correlation. ¿ Conduct stochastic analysis of a single-server system with single-queue and multiple-queues, respectively, for AQM scheme in the presence of multiple priority traffic classes scheduled by the Priority Resume (PR) policy. ¿ Carry out the performance comparison of AQM with PR and First-In First-Out (FIFO) scheme and compare the performance of AQM with single PR priority queue and multiple priority queues, respectively.

In order to efficiently utilize the scarce wireless resource as well as keep up with the ever-increasing demand for Quality-of-Service (QoS) of multimedia applications, wireless networks are undergoing rapid development and dramatic changes in the underlying technologies and protocols. The Medium Access Control (MAC) protocol, which coordinates the channel access and data transmission of wireless stations, plays a pivotal role in wireless networks. Performance modelling and analysis has been and continues to be of great theoretical and practical importance in the design and development of wireless networks. This research is devoted to developing efficient and cost-effective analytical tools for the performance analysis and enhancement of MAC protocols in Wireless Local Area Networks (WLANs) under heterogeneous multimedia traffic. To support the MAC-layer QoS in WLANs, the IEEE 802.11e Enhanced Distributed Channel Access (EDCA) protocol has proposed three QoS differentiation schemes in terms of Arbitrary Inter-Frame Space (AIFS), Contention Window (CW), and Transmission Opportunity (TXOP). This research starts with the development of new analytical models for the TXOP scheme specified in the EDCA protocol under Poisson traffic. A dynamic TXOP scheme is then proposed to adjust the TXOP limits according to the status of the transmission queue. Theoretical analysis and simulation experiments show that the proposed dynamic scheme largely improves the performance of TXOP. To evaluate the TXOP scheme in the presence of ii heterogeneous traffic, a versatile analytical model is developed to capture the traffic heterogeneity and model the features of burst transmission. The performance results highlight the importance of taking into account the heterogeneous traffic for the accurate evaluation of the TXOP scheme in wireless multimedia networks. To obtain a thorough and deep understanding of the performance attributes of the EDCA protocol, a comprehensive analytical model is then proposed to accommodate the integration of the three QoS schemes of EDCA in terms of AIFS, CW, and TXOP under Poisson traffic. The performance results show that the TXOP scheme can not only support service differentiation but also improve the network performance, whereas the AIFS and CW schemes provide QoS differentiation only. Moreover, the results demonstrate that the MAC buffer size has considerable impact on the QoS performance of EDCA under Poisson traffic. To investigate the performance of EDCA in wireless multimedia networks, an analytical model is further developed for EDCA under heterogeneous traffic. The performance results demonstrate the significant effects of heterogeneous traffic on the total delay and frame losses of EDCA with different buffer sizes. Finally, an efficient admission control scheme is presented for the IEEE 802.11e WLANs based on analytical modelling and a game-theoretical approach. The admission control scheme can maintain the system operation at an optimal point where the utility of the Access Point (AP) is maximized with the QoS constraints of various users.