Дисертації з теми "Edge Networking"

Oggi connettersi a Internet è una pratica di uso comune, destinata a crescere notevolmente con l'avvento dell'Internet of Things. Nonostante il grande successo, Internet presenta diversi limiti architetturali che lo rendono un veicolo povero in confronto alla grande mole di contenuti trasmessi. La seconda importante limitazione riguarda la sicurezza che è applicata solamente a livello di host ma non di contenuto. Per promuovere la prototipizzazione di nuovi paradigmi architetturali, nel 2010 la National Science Fondation crea il programma Future Internet Architecture. Tra i progetti nati, il più emergente è Named Data Networking(NDN). Il suo intento è quello di distogliere l'attenzione da "dove" trovare una risorsa per concentrarsi su "cosa" applicazioni e utenti cerchino. Per questo NDN pensa che l'identificazione non debba più riguardare gli host ma i contenuti. Quest'ultimi sono distinti da un nome univoco, godono di immutabilità e racchiudono la strategia di sicurezza. L'obiettivo di questo progetto è quello di dare un piccolo contributo a Named Data Networking, ideando e sviluppando un protocollo di computazione cooperativa, che operi per mezzo del protocollo NDN. L'intento del protocollo è quello di permettere a generici dispositivi wifi di assegnare l'esecuzione di alcuni dei loro task ad un nodo nelle vicinanze. Inoltre, la tesi accenna come il protocollo si inserisca nel tema dell'Edge computing, che assieme al Fog computing pone le basi per l'evoluzione dell'architettura Cloud. Il lavoro ha portato alla realizzazione di un prototipo del protocollo che è possibile installare sui nodi del simulatore di reti ndnSIM. Esso soddisfa i comportamenti base richiesti ed offre buone prestazioni negli scenari statici. L'elaborato si conclude con alcuni test che ne confermano il corretto funzionamento ma allo stesso tempo denotano alcuni aspetti da migliorare negli sviluppi futuri.

Abstract. The involvement of low power Internet of Things (IoT) devices in the Wireless Sensor Networks (WSN) allow enhanced autonomous monitoring capability in many application areas. Recently, the principles of edge computing paradigm have been used to cater onsite processing and managing actions in WSNs. However, WSNs deployed in remote sites require human involvement in data collection process since internet accessibility is still limited to population dense areas. Nowadays, researchers propose UAVs for monitoring applications where human involvement is required frequently. In this thesis work, we introduce an edge-based architecture which create end-to-end secure communication between IoT sensors in a remote WSN and central cloud via UAV, which assist the data collection, processing and managing procedures of the remote WSN. Since power is a limited resource, we propose Bluetooth Low Energy (BLE) as the communication media between UAV and sensors in the WSN, where BLE is considered as an ultra-low power radio access technology. To examine the performance of the system model, we have presented a simulation analysis considering three sensor nodes array types that can realize in the practical environment. The impact of BLE data rate, impact of speed of the UAV, impact of distance between adjacent sensors and impact of data generation rate of the sensor node have been analysed to examine the performance of system. Moreover, to observe the practical functionality of the proposed architecture, prototype implementation is presented using commercially available off-the-shelf devices. The prototype of the system is implemented assuming ideal environment.

Computational offloading advances the deployment of Mobile Edge Computing (MEC) in the next generation communication networks. However, the distributed nature of the mobile users and the complex applications make it challenging to schedule the tasks reasonably among multiple devices. Therefore, by leveraging the idea of Software-Defined Networking (SDN) and Service Composition (SC), we propose a Software-Defined Service Composition model (SDSC). In this model, the SDSC controller is deployed at the edge of the network and composes service in a centralized manner to reduce the latency of the task execution and the traffic on the access links by satisfying the user-specific requirement. We formulate the low latency service composition as a Constraint Satisfaction Problem (CSP) to make it a user-centric approach. With the advent of the SDN, the global view and the control of the entire network are made available to the network controller which is further leveraged by our SDSC approach. Furthermore, the service discovery and the offloading of tasks are designed for MEC environment so that the users can have a complex and robust system. Moreover, this approach performs the task execution in a distributed manner. We also define the QoS model which provides the composition rule that forms the best possible service composition at the time of need. Moreover, we have extended our SDSC model to involve the constant mobility of the mobile devices. To solve the mobility issue, we propose a mobility model and a mobility-aware QoS approach enabled in the SDSC model. The experimental simulation results demonstrate that our approach can obtain better performance than the energy saving greedy algorithm and the random offloading approach in a mobile environment.

The current internet architecture is evolving from a simple carrier of bits to a platform able to provide multiple complex services running across the entire Network Service Provider (NSP) infrastructure. This calls for increased flexibility in resource management and allocation to provide dedicated, on-demand network services, leveraging a distributed infrastructure consisting of heterogeneous devices. More specifically, NSPs rely on a plethora of low-cost Customer Premise Equipment (CPE), as well as more powerful appliances at the edge of the network and in dedicated data-centers. Currently a great research effort is spent to provide this flexibility through Fog computing, Network Functions Virtualization (NFV), and data plane programmability. Fog computing or Edge computing extends the compute and storage capabilities to the edge of the network, closer to the rapidly growing number of connected devices and applications that consume cloud services and generate massive amounts of data. A complementary technology is NFV, a network architecture concept targeting the execution of software Network Functions (NFs) in isolated Virtual Machines (VMs), potentially sharing a pool of general-purpose hosts, rather than running on dedicated hardware (i.e., appliances). Such a solution enables virtual network appliances (i.e., VMs executing network functions) to be provisioned, allocated a different amount of resources, and possibly moved across data centers in little time, which is key in ensuring that the network can keep up with the flexibility in the provisioning and deployment of virtual hosts in today’s virtualized data centers. Moreover, recent advances in networking hardware have introduced new programmable network devices that can efficiently execute complex operations at line rate. As a result, NFs can be (partially or entirely) folded into the network, speeding up the execution of distributed services. The work described in this Ph.D. thesis aims at showing how various network services can be deployed throughout the NSP infrastructure, accommodating to the different hardware capabilities of various appliances, by applying and extending the above-mentioned solutions. First, we consider a data center environment and the deployment of (virtualized) NFs. In this scenario, we introduce a novel methodology for the modelization of different NFs aimed at estimating their performance on different execution platforms. Moreover, we propose to extend the traditional NFV deployment outside of the data center to leverage the entire NSP infrastructure. This can be achieved by integrating native NFs, commonly available in low-cost CPEs, with an existing NFV framework. This facilitates the provision of services that require NFs close to the end user (e.g., IPsec terminator). On the other hand, resource-hungry virtualized NFs are run in the NSP data center, where they can take advantage of the superior computing and storage capabilities. As an application, we also present a novel technique to deploy a distributed service, specifically a web filter, to leverage both the low latency of a CPE and the computational power of a data center. We then show that also the core network, today dedicated solely to packet routing, can be exploited to provide useful services. In particular, we propose a novel method to provide distributed network services in core network devices by means of task distribution and a seamless coordination among the peers involved. The aim is to transform existing network nodes (e.g., routers, switches, access points) into a highly distributed data acquisition and processing platform, which will significantly reduce the storage requirements at the Network Operations Center and the packet duplication overhead. Finally, we propose to use new programmable network devices in data center networks to provide much needed services to distributed applications. By offloading part of the computation directly to the networking hardware, we show that it is possible to reduce both the network traffic and the overall job completion time.

Les travaux de cette thèse s'inscrivent dans le contexte général des réseaux logiciels, dits "Software-Defined Networking" (SDN). Ce paradigme récent est l'une des initiatives les plus notables pour rendre les réseaux actuels programmables ou, en d'autres termes, plus simple à configurer, à tester, à corriger et à faire évoluer. Dans un écosystème SDN, l'interface nord (Northbound API) est utilisée par l'administrateur réseaux pour définir ses politiques et programmer le plan de contrôle, elle représente donc un enjeu majeur. Idéalement, cette interface nord devrait permettre aux administrateurs de décrire, le plus simplement possible, des services réseaux et leurs interactions, plutôt que de spécifier comment et sur quels équipements physiques ils doivent être déployés. Des travaux existants montrent que cela peut être notamment réalisé grâce à des solutions de virtualisation de réseaux et des langages de programmation dédiés de haut niveau. L'objectif de ce travail de thèse est de proposer une nouvelle interface nord qui, d'une part, exploiterait la virtualisation de réseau et, d'autre part, exposerait ses services sous la forme d'un langage de programmation dédié. Actuellement, plusieurs langages intégrant des solutions de virtualisation de réseau existent. Néanmoins, nous pensons que les modèles d'abstraction qu'ils utilisent pour construire des réseaux virtuels restent inappropriés pour assurer des critères de simplicité, modularité et flexibilité des topologies virtuelles et des programmes de contrôle. Dans ce contexte, nous proposons un nouveau langage de contrôle de réseaux nommé AirNet. Ce dernier intègre un modèle d'abstraction dont la principale caractéristique est d'offrir une séparation nette entre les équipements de bordure (Edge) et de cœur de réseau (Fabric). Cette idée est bien connue et acceptée dans le domaine des architectures réseaux. L'originalité de notre contribution étant de faire remonter ce concept au niveau du plan de contrôle virtuel et non de le restreindre au seul plan physique. Ainsi, des frontières logiques entre les différents types de politiques existeront (fonctions de contrôle et de données vs. fonctions de transport), garantissant ainsi la modularité et la réutilisabilité de tout ou partie du programme de contrôle. De plus, dans l'approche proposée, la définition du réseau virtuel et des politiques peut être totalement dissociée de l'infrastructure physique cible, favorisant ainsi la portabilité des applications de contrôle. Une implémentation du langage AirNet a également été réalisée. Ce prototype inclut en particulier une bibliothèque des primitives et opérateurs du langage, ainsi qu'un hyperviseur qui assure la composition des politiques de contrôle sur un réseau virtuel, et leur transposition (mapping) sur l'infrastructure physique. Afin de s'appuyer sur des contrôleurs SDN existants, l'hyperviseur inclut des modules d'intégration des contrôleurs POX et RYU. Une validation expérimentale a été menée sur différents cas d'étude (filtrage, répartition de charge, authentification dynamique, limitation de bande passante, etc.) dont les résultats attestent de la faisabilité de la solution. Enfin, des mesures de performances ont montré que le surcoût apporté par cette nouvelle couche d'abstraction est parfaitement acceptable
The work of this thesis falls within the general context of software-defined networking (SDN). This new paradigm is one of the most significant initiatives to enable networks programmability or, in other words, to make current networks easier to configure, test, debug and evolve. Within an SDN ecosystem, the Northbound interface is used by network administrators to define policies and to program the control plane, it thus represents a major challenge. Ideally, this northbound interface should allow administrators to describe, as simply as possible, network services and their interactions, rather than specifying how and on what physical device they need to be deployed. Current related works show that this can be partly achieved through virtualization solutions and high-level domain specific languages (DSL). The objective of this thesis is to propose a new Northbound interface which will, on the one hand, rely on network virtualization and, on the other hand, expose its services as a domain specific programming language. Currently, several languages that include network virtualization solutions exist. Nevertheless, we believe that the abstract models they are using to build virtual networks remain inadequate to ensure simplicity, modularity and flexibility of virtual topologies and control programs. In this context, we propose a new network control language named AirNet. Our language is built on top of an abstraction model whose main feature is to provide a clear separation between edge and core network devices. This concept is a well-known and accepted idea within the network designer community. The originality of our contribution is to lift up this concept at the virtual control plane, not limiting it solely at the physical plane. Thus, logical boundaries between different types of policies will exist (control and data functions vs. transport functions), ensuring modularity and reusability of the control program. Moreover, in the proposed approach, the definition of the virtual network and policies is totally dissociated from the target physical infrastructure, promoting the portability of control applications. An implementation of the AirNet language has also been done. This prototype includes in particular a library that implements the primitives and operators of the language, and a hypervisor that achieves the composition of the control policies on the virtual network, and their mapping on the physical infrastructure. In order to rely on existing SDN controllers, the hypervisor includes integration modules for the POX and RYU controllers. An experimental validation has been also conducted on different use cases (filtering, load balancing, dynamic authentication, bandwidth throttling, etc.), whose results demonstrate the feasibility of our solution. Finally, performance measurements have shown that the additional cost brought by this new abstraction layer is perfectly acceptable

Em um futuro próximo, “dispositivos inteligentes” serão massivamente empregados em campos de batalha. Essa já é uma realidade, porém, o número de dispositivos utilizados em campos de batalha tende a aumentar em ordens de magnitude. As redes de comunicação de dados serão essenciais para transmitir os dados que esses dispositivos coletam e transformá-los em informações valiosas utilizadas como suporte à atuação humana. O suporte à tomada de decisão, ou mesmo níveis de autonomia, permitindo que estes dispositivos coordenem outros dispositivos, exigem comunicação contínua. Desafios relacionados à comunicação surgirão devido à dinamicidade do ambiente. A configuração da rede deve refletir decisões superiores automaticamente. A grande escala das redes conectando os altos escalões, tropas, veículos e sensores, aliada à falta de padronização dos dispositivos, tornará a integração destes desafiadora. Em um ambiente tão heterogêneo, muitos protocolos e tecnologias coexistirão. As redes de campo de batalha são um elemento de suma importância nas operações militares modernas e conceito de guerra centrada em rede é uma tendência sem volta e influencia desde os altos escalões até o controle de tropas Embora estudos tenham sido realizados nessa área, a maioria deles aborda redes estratégicas de alto nível e portanto não levam em conta as “redes táticas de última milha” (TEN), que compreendem dispositivos de comunicação com recursos limitados, como sensores ou ainda pequenos veículos aéreos não tripulados. Em uma tentativa de preencher esta lacuna, esse trabalho propõe uma arquitetura que combina conceitos dos paradigmas de redes definidas por software (SDN) juntamente com redes tolerantes à atraso/disrupçoes (DTN), para aplicação em redes táticas de última milha. O uso de SDN em cenários com nodos móveis é avaliado considerando uma aplicação de vigilância que utiliza streaming de vídeo e medidas de Qualidade de Experiência (QoE) de usuário são coletadas. Com base nos resultados obtidos, uma aplicação em conjunto dos conceitos de SDN e DTN é proposta, além disso abordamos a escolha do nodo que atuará como controlador SDN na rede. Os experimentos foram executados utilizando um emulador de redes. Apesar de pesquisas adicionais serem necessárias – considerado requisitos de segurança, por exemplo – os resultados foram promissores e demonstram a aplicabilidade destes conceitos no cenários das TENs.
The future battlefield tends to be populated by a plethora of “intelligent things”. In some ways, this is already a reality, but in future battlefields, the number of deployed things should be orders of magnitude higher. Networked communication is essential to take real advantage of the deployed devices on the battlefield, and to transform the data collected by them into information valuable for the human warfighters. Support for human decision making and even a level of autonomy, allowing devices to coordinate and interact with each other to execute their activities in a collaborative way require continuous communication. Challenges regarding communication will arise from the high dynamics of the environment. The network adaption and management should occur autonomously, and it should reflect upper-level decisions. The large scale of the network connecting high-level echelons, troops on the field, and sensors of many types, beside the lack of communication standards turn the integration of the devices more challenging. In such a heterogeneous environment, many protocols and communication technologies coexist. This way, battlefield networks is an element of paramount importance in modern military operations Additionally, a change of paradigm regarding levels of autonomy and cooperation between humans and machines is in course and the concept of network-centric warfare is a no way back trend. Although new studies have been carried out in this area, most of these concern higher-level strategic networks, with abundant resources. Thus, these studies fail to take into account the “last-mile Tactical Edge Network (TEN) level,” which comprises resource constrained communication devices carried by troopers, sensor nodes deployed on the field or small unmanned aerial vehicles. In an attempt to fill this gap, this work proposes an architecture combining concepts from software-defined networking (SDN) paradigm and the delay-tolerant approach to support applications in the last-mile TEN. First, the use of SDN in dynamic scenarios regarding node positioning is evaluated through a surveillance application using video streaming and Quality of Experience (QoE) measures are captured on the video player. We also explore the election of nodes to act as SDN Controllers in the TEN environment. The experiments use emulator for SDN with support to wireless networks. Further investigation is required, for example, considering security requirements, however the results are promising and demonstrate the applicability of this architecture in the TEN network scenario.

La maturation de la provision et de la gestion des infrastructures de cloud computing et d'edge computing a engendré le Cloud Edge Computing Continuum (CECC), facilitant le déploiement et la migration d'applications entre les infrastructures cloud centralisées et les infrastructures edge décentralisées. Cette transition a donné vie à des nouveaux cas d'utilisation dans des secteurs tels que l'Internet Industriel des Objets (IIoT), les véhicules autonomes et la réalité augmentée, tous profitant de cette architecture distribuée. Ces domaines d'application requièrent à la fois la scalabilité des centres de données massifs du cloud traditionnel et la faible latence des infrastructures edge computing. Les progrès technologiques, tels que la virtualisation, la conteneurisation et les réseaux 5G, ont facilité le développement d'applications CECC, passant des architectures monolithiques aux microservices. L'orchestration efficace des applications CECC est essentielle pour garantir la performance et optimiser l'utilisation des ressources. Cette thèse propose des solutions pour la gestion automatisée du CECC, se concentrant sur la collecte de données de supervision, la compréhension du fonctionnement des applications et de l'infrastructure, et la prise de décisions sur le placement la migration et l'adaptation des ressources des applications. Dans la première contribution de cette thèse, nous proposons un framework de supervision de services multi-domaines de bout en bout. De tels services consistent en des applications et des fonctions réseau qui s'étendent sur plusieurs domaines technologiques, chacun présentant ses propres intrications uniques. Le système de supervision proposé utilise une structure unifiée des Key Performance Indicators (KPIs), abstrayant efficacement toutes les complexités sous-jacentes. Des tests approfondis dans différents scénarios valident la scalabilité du framework et sa capacité à superviser un grand nombre de services simultanément. La prochaine étape de la thèse impliquait le profilage des applications, dans lequel nous avons mené une étude expérimentale pour explorer le comportement de différents types d'applications dans des environnements cloud-native. Cette étude met en évidence l'incapacité des propriétaires d'applications à configurer les ressources appropriées pour leurs applications afin de fonctionner de manière optimale sans entraîner de gaspillage des ressources d'infrastructure. Ensuite, nous avons utilisé des techniques de machine learning et d'eXplainable AI (XAI) pour construire des modèles capables de prédire la dégradation des performances des applications, en utilisant des ensembles de données générés à partir de notre étude. Lorsque le modèle prédit un déclin des performances de l'application, le module XAI fournit des explications pour la sortie du modèle, facilitant l'identification de la cause profonde de la dégradation du service. Cette cause profonde est ensuite traitée par le gestionnaire d'application.Le parcours de cette thèse s'est conclu par la proposition d'une architecture pour la gestion du cycle de vie des applications CECC. Cette architecture utilise des profils d'application et d'infrastructure pour déployer et migrer des applications tout en tenant compte de l'empreinte carbone du déploiement CECC.Le principal défi réside dans la représentation concrète du profil d'application de manière à pouvoir spécifier les exigences actuelles et futures de l'application. Ce défi a été relevé en représentant efficacement le profil d'application pour faciliter la dérivation des exigences actuelles et futures de l'application
The maturation of cloud computing and edge computing infrastructure provisioning and management has led to the emergence of Cloud Edge Computing Continuum (CECC). CECC enables seamless deployment and migration of applications between centralized cloud infrastructures and decentralized edge infrastructure. This evolution has driven new use cases across industries, including Industrial Internet of Things (IIoT), autonomous vehicles, and augmented reality, all benefiting from this distributed architecture.These use cases require scalability and storage from massive data centers typical of traditional cloud computing, as well as the low latency and high bandwidth offered by edge computing infrastructures. Several factors enable the development and deployment of applications to fully leverage CECC : advancements in application deployment technologies like virtualization and containerization, a shift in application architecture and development methodology towards microservices architectures, and innovations in networking technologies such as 5G mobile networks. Efficiently orchestrating applications within the CECC framework is crucial for meeting performance requirements and optimizing infrastructure resource utilization. This thesis proposes solutions for zero-touch management of CECC, focusing on automated monitoring, profiling, and decision-making processes. These solutions aim to automate application management, facilitating seamless orchestration and resource optimization.In the first contribution, a novel monitoring system for multi-domain services is proposed, utilizing a unified structure for Key Performance Indicators (KPIs) to abstract underlying technologies. This scalable system monitors end-to-end network slices, including Radio Access Network (RAN), Core Network (CN), and Cloud/Edge domains.The second contribution presents results from an experimental study aiming to detect if a tenant's configuration allows running its service optimally. The study provides insights on detecting and correcting performance degradation due to misconfiguration of service resources.Moving towards decision-making of a CECC manager, the third contribution proposes a Zero-Touch Service Management (ZSM) framework featuring a fine-granular computing resource scaler in a cloud-native environment. The scaler uses AI/ML models to predict microservice performances, with an XAI module conducting root-cause analysis for service degradation. Afterwards, the proposed framework scales only the needed resources (i.e., CPU or memory) to overcome the service degradation. Finally, in the last contribution, an architecture of CECC Application Orchestrator is proposed, leveraging applications and infrastructures profiling for efficient management. These profiles represent current and future applications' requirements, guiding decision-making processes (placement, resources scaling. migration) to minimize carbon footprint and deployment costs

La disponibilité des contenus partagés en ligne devient un élément essentiel pour toute la chaîne de distribution de vidéos. Pour fournir des contenus aux utilisateurs avec une excellente disponibilité et répondre à leurs exigences toujours croissantes, les opérateurs de content delivery networks (CDNs) doivent assurer une haute qualité de services, définie par des métriques comme le taux de transfert ou la latence inclus dans les contrats de Service Level Agreement (SLA). La réplication adaptative se présente comme un mécanisme de stockage très prometteur pour at- teindre cet objectif. Par contre, une question importante reste encore ouverte: comment assurer la mise en place de ces SLAs, tout en évitant le gaspillage de ressources? Le sujet de la thèse porte précisément sur l'étude et l'évaluation de systèmes de réplication de données pour la nouvelle génération de CDNs hybrides, dont une partie des ressources de réseaux et de stockage proviennent de l'équipement des utilisateurs. Pour cela, nous proposons (i) une architecture de gestion de ressources des utilisateurs nommée Caju, et (ii) trois nouveaux systèmes de réplication adaptatifs, AREN, Hermes, et WiseReplica. Des simulations précises avec Caju montrent que nos systèmes de réplication adaptatifs sont très performants et peuvent être facilement étendus à d'autres types d'architecture. Comme perspectives, nous comptons réaliser le développement et l'évaluation d'un prototype proof-of-concept sur PlanetLab.

L'Internet des objets (IoT), ne comprenant à l'origine que quelques dispositifs de détection simple, atteint aujourd’hui 34 milliards d’objets connectés d'ici fin 2020. Ces objets ne peuvent plus être définis comme de simples capteurs de surveillance. Les capacités de l'IoT ont été améliorées ces dernières années tandis-que que les capacités de calcul et de stockage de masse sont devenus des marchandises. Aux débuts de l'IoT, le traitement et le stockage étaient généralement effectués dans le cloud. Les nouvelles architectures IoT sont capables d'exécuter des tâches complexes directement sur l'appareil, permettant ainsi le concept d'un continuum de calcul étendu. Les scénarios critiques et temps réel, comme par exemple la détection de véhicules autonomes, la surveillance de zone ou le sauvetage en cas de catastrophe, nécessitent que l’ensemble des acteurs impliqués soient coordonnés et collaborent sans interaction humaine vers un objectif commun, partageant des données et des ressources, même dans les zones couvertes par des réseaux intermittents. Cela pose de nouveaux problèmes dans les systèmes distribués, la gestion des ressources, l'orchestration des appareils et le traitement des données. Ce travail propose un nouveau cadre de communication et d'orchestration, à savoir le C-Continuum, conçu dans des architectures IoT hétérogènes à travers plusieurs scénarios d'application. Ce travail se concentre pour gérer les ressources sur deux macro-scénarios clés de durabilité : (a) la détection et la sensibilisation à l'environnement, et (b) le soutien à la mobilité électrique. Dans le premier cas, un mécanisme de mesure de la qualité de l'air sur une longue période avec différentes applications à l'échelle mondiale (3 continents et 4 pays) est introduit. Le système a été développé en interne depuis la conception du capteur jusqu'aux opérations de mist-computing effectuées par les nœuds. Dans le deuxième scénario une technique pour transmettre de grandes quantités de données, entre un véhicule en mouvement et un centre de contrôle est proposé. Ces données sont de haute granularité temporelle relatives et permettent conjointement d'allouer des tâches sur demande dans le continuum de calcul
As Internet of Things (IoT), originally comprising of only a few simple sensing devices, reaches 34 billion units by the end of 2020, they cannot be defined as merely monitoring sensors anymore. IoT capabilities have been improved in recent years as relatively large internal computation and storage capacity are becoming a commodity. In the early days of IoT, processing and storage were typically performed in cloud. New IoT architectures are able to perform complex tasks directly on-device, thus enabling the concept of an extended computational continuum. Real-time critical scenarios e.g. autonomous vehicles sensing, area surveying or disaster rescue and recovery require all the actors involved to be coordinated and collaborate without human interaction to a common goal, sharing data and resources, even in intermittent networks covered areas. This poses new problems in distributed systems, resource management, device orchestration,as well as data processing. This work proposes a new orchestration and communication framework, namely CContinuum, designed to manage resources in heterogeneous IoT architectures across multiple application scenarios. This work focuses on two key sustainability macroscenarios: (a) environmental sensing and awareness, and (b) electric mobility support. In the first case a mechanism to measure air quality over a long period of time for different applications at global scale (3 continents 4 countries) is introduced. The system has been developed in-house from the sensor design to the mist-computing operations performed by the nodes. In the second scenario, a technique to transmit large amounts of fine-time granularity battery data from a moving vehicle to a control center is proposed jointly with the ability of allocating tasks on demand within the computing continuum

The Internet of Things is a recent computing paradigm, de- fined by networks of highly connected things – sensors, actuators and smart objects – communicating across networks of homes, buildings, vehicles, and even people. The Internet of Things brings with it a host of new problems, from managing security on constrained devices to processing never before seen amounts of data. While cloud computing might be able to keep up with current data processing and computational demands, it is unclear whether it can be extended to the requirements brought forth by Internet of Things. Fog computing provides an architectural solution to address some of these problems by providing a layer of intermediary nodes within what is called an edge network, separating the local object networks and the Cloud. These edge nodes provide interoperability, real-time interaction, routing, and, if necessary, computational delegation to the Cloud. This paper attempts to evaluate Go, a distributed systems language developed by Google, in the context of requirements set forth by Fog computing. Similar methodologies of previous literature are simulated and benchmarked against in order to assess the viability of Go in the edge nodes of Fog computing architecture.

Cette thèse est dédiée à l'étude des systèmes de communications pour application aux services vidéo fournis par les réseaux radio mobiles. Ce travail met l'accent sur les systèmes point à multipoints et propose plusieurs améliorations : Dans un premier temps, on définit un système qui combine le décodage robuste aux retransmissions ARQ, et ce de telle façon à réduire le nombre de retransmissions tout en gardant le même niveau de qualité. Contrairement aux systèmes actuels (avec ou sans retransmissions), ce systèmes offre également la possibilité de choisir le compromis débit/qualité via un paramètre du système. Par la suite, on considère les sytèmes de transmission d'une vidéo scalable vers plusieurs terminaux. Des extensions des systèmes Go-Back-N (GBN) Automatic Repeat reQuest (ARQ) et Selective Repeat (SR) ARQ sont étudiées et comparées à un nouveau système. On montre que ce dernier limite les besoins de bufferisation au niveau du terminal récepteur tout en ayant des performances optimales (en termes de quantité de données transmises avec succès sur une période donnée). Finalement, on montre que même sous une contrainte débit on peut utiliser les retransmissions dans les communications point à multipoints à condition de ne pas dépasser une certaine limite sur le nombre d'utilisateurs. Si l'utilisation des retransmissions ARQ est introduite dans les sytèmes de Multicast/Broadcast 3GPP et/ou WiMAX, le système pourra garantir une qualité nominale à un certain nombre d'utilisateurs, ce qui n'est pas le cas des systèmes de Multicast/Broadcast actuels
In this thesis, video communication systems are studied for application to video services provided over wireless mobile networks. This work emphasizes on point-to-multipoint communications and proposes many enhancements to the current systems : First, a scheme combining robust decoding with retransmissions is defined so that the number of retransmissions is reduced and the quality of the received video can be controlled. As opposed to current retransmissionless and retransmission-based schemes, this scheme also offers the possibility to trade throughput for quality and vice versa. Then, the transmission of a two-level scalable video sequence towards several clients is considered. Schemes using the basic Go-back-N (GBN) and Selective Repeat (SR) Automatic Repeat reQuest (ARQ) techniques are studied. A new scheme is also proposed and studied. The new scheme reduces the buffering requirement at the receiver end while keeping the performance optimal (in terms of the amount of data successfully transmitted within a given period of time). The different schemes were shown to be applicable to 2G, 3G and WiMAX systems. Finally, we prove that retransmissions can be used in point-to-multipoint communications up to a given limit on the number of receivers (contrary to the current wireless systems where ARQ is only used sin point-to-point communications). If retransmissions are introduced in the current Multicast/Broadcast services (supported by the 3GPP and mobile WiMAX), the system will guarantee a certain amount of receivers to have the nominal quality whereas the current Multicast/Broadcast services do not garantee any receiver of the nominal quality

Cette thèse est dédiée à l'étude des systèmes de communications pour application aux services vidéo fournis par les réseaux radio mobiles. Ce travail met l'accent sur les systèmes point à multipoints et propose plusieurs améliorations : Dans un premier temps, on définit un système qui combine le décodage robuste aux retransmissions ARQ, et ce de telle façon à réduire le nombre de retransmissions tout en gardant le même niveau de qualité. Contrairement aux systèmes actuels (avec ou sans retransmissions), ce systèmes offre également la possibilité de choisir le compromis débit/qualité via un paramètre du système. Par la suite, on considère les sytèmes de transmission d'une vidéo scalable vers plusieurs terminaux. Des extensions des systèmes Go-Back-N (GBN) Automatic Repeat reQuest (ARQ) et Selective Repeat (SR) ARQ sont étudiées et comparées à un nouveau système. On montre que ce dernier limite les besoins de bufferisation au niveau du terminal récepteur tout en ayant des performances optimales (en termes de quantité de données transmises avec succès sur une période donnée). Finalement, on montre que même sous une contrainte débit on peut utiliser les retransmissions dans les communications point à multipoints à condition de ne pas dépasser une certaine limite sur le nombre d'utilisateurs. Si l'utilisation des retransmissions ARQ est introduite dans les sytèmes de Multicast/Broadcast 3GPP et/ou WiMAX, le système pourra garantir une qualité nominale à un certain nombre d'utilisateurs, ce qui n'est pas le cas des systèmes de Multicast/Broadcast actuels.

Le paradigme de MEC (Mobile Edge Computing) consiste à mettre les ressources de calcul et de stockage aux « extrémités » du réseau à proximité des utilisateurs finaux. Le terme « edge » désigne n’importe quel type de station de base de réseau. Les motivations pour l’adoption de ce nouveau concept sont principalement la réduction de la charge au cœur du réseau et la diminution de la latence grâce à la proximité des ressources et ainsi améliorer l’expérience utilisateur. Les serveurs MEC sont de bons candidats pour héberger les applications mobiles et diffuser le contenu Web. La mise en cache à l’extrémité du réseau, ou Edge Caching en anglais, est l’une des technologies les plus émergentes connues comme solution de récupération de contenu au bord du réseau. Elle est aussi considérée comme une technologie permettant la mise en place du concept MEC puisqu’elle présente une opportunité intéressante pour implémenter les services de mise en cache. En particulier, les serveurs MEC sont implémentés directement au niveau des stations de base, ce qui permet la mise en cache à l’extrémité du réseau et assure un déploiement à proximité des utilisateurs finaux. Cependant, l’intégration des serveurs MEC dans les stations de base complexifie le problème de la consommation de l’énergie, particulièrement dans un tel environnement qui est dynamique et sujet à des changements au fil du temps. Par ailleurs, la demande des utilisateurs des appareils mobiles est en constante augmentation ainsi que leur expectation d’une expérience meilleure. Sachant que le cache est d’une taille limitée, il est donc nécessaire et crucial que les mécanismes de mise en cache soient en mesure de faire face à cette situation et de proposer des solutions valables et satisfaisants à long terme. La plupart des études existantes se sont focalisées sur l’allocation de cache, la popularité du contenu ou encore la manière de concevoir le cache. Dans cette thèse, nous présentons une nouvelle stratégie de mise en cache écoénergétique basée sur la logique floue (Fuzzy logic). Notre proposition prend en compte les quatre caractéristiques d’un environnement mobile et introduit une implémentation matérielle en utilisant les FPGA (Field-Programmable Gate Array) pour réduire les besoins globaux en énergie. L’adoption d’une stratégie de mise en cache adéquate sur les serveurs MEC ouvre la possibilité d’utiliser des techniques d’intelligence artificielle (IA) et d’apprentissage automatique (Machine Learning) aux extrémités des réseaux mobiles. L’exploitation des informations de contexte des utilisateurs permet de concevoir une mise en cache intelligente sensible au contexte. La reconnaissance du contexte permet au cache de connaître son environnement, tandis que l’intelligence lui permet de prendre les bonnes décisions en sélectionnant le contenu approprié à mettre en cache afin d’optimiser les performances du caching. Inspiré par le succès de l’apprentissage par renforcement utilisant des agents pour traiter des problèmes de prise de décision, nous avons étendu notre système de mise en cache basé sur la logique floue à un modèle d’apprentissage par renforcement modifié. Le cadre proposé vise à maximiser le taux de réussite du cache (hit rate) et nécessite une prise de conscience multiple sure les conditions de web et l’utilisateur final. La méthode d’apprentissage par renforcement modifiée diffère des autres algorithmes par le taux d’apprentissage qui utilise la méthode du gradient stochastique décent (stochastic gradient decent) en plus de tirer parti de l’apprentissage en utilisant la décision de mise en cache optimale obtenue à partir des règles de la logique floue
Mobile edge computing (MEC) concept proposes to bring the computing and storage resources in close proximity to the end user by placing these resources at the network edge. The motivation is to alleviate the mobile core and to reduce latency for mobile users due to their close proximity to the edge. MEC servers are candidates to host mobile applications and serve web contents. Edge caching is one of the most emerging technologies recognized as a content retrieval solution in the edge of the network. It has been also considered as enabling technology of mobile edge computing that presents an interesting opportunity to perform caching services. Particularly, the MEC servers are implemented directly at the base stations which enable edge caching and ensure deployment in close-proximity to the mobile users. However, the integration of servers in mobile edge computing environment (base stations) complicates the energy saving issue because the power consumed by mobile edge computing servers is costly especially when the load changes dynamically over time. Furthermore, users with mobile devices arise their demands, introducing the challenge of handling such mobile content requests beside the limited caching size. Thus, it is necessary and crucial for caching mechanisms to consider context-aware factors, meanwhile most existing studies focus on cache allocation, content popularity and cache design. In this thesis, we present a novel energy-efficient fuzzy caching strategy for edge devices that takes into consideration four influencing features of mobile environment, while introducing a hardware implementation using Field-Programmable Gate Array (FPGA) to cut the overall energy requirements. Performing an adequate caching strategy on MEC servers opens the possibility of employing artificial intelligence (AI) techniques and machine learning at mobile network edges. Exploiting users context information intelligently makes it possible to design an intelligent context-aware mobile edge caching. Context awareness enables the cache to be aware of its environment, while intelligence enables each cache to make the right decisions of selecting appropriate contents to be cached so that to maximize the caching performance. Inspired by the success of reinforcement learning (RL) that uses agents to deal with decision making problems, we extended our fuzzy-caching system into a modified reinforcement learning model. The proposed framework aims to maximize the cache hit rate and requires a multi awareness. The modified RL differs from other RL algorithms in the learning rate that uses the method of stochastic gradient decent beside taking advantage of learning using the optimal caching decision obtained from fuzzy rules

This thesis starts by considering the problem of data dissemination, investigating several network schemes. Besides, to ensure the consistency of the data collected, a distributed consensus sensing application is designed. Then, a mobile Edge computing system is modeled. This paradigm provides computational capabilities at the edge of the network and is able to fulfill the requirements of the Internet of Mobile Things. The model is used to derive the minimum number of processors to be allocated to obtain a given requests dropping probability. Finally, mobile Edge computing and Cloud computing systems are compared. Two analytical models are developed and validated, considering the total service time as a key metric. The comparison gives some insight on how these systems should be designed to handle a given load.

L’Information-Centric Networking si è affermato come una delle tecnologie più promettenti dell'Internet del Futuro, in grado di abilitare la fornitura di servizi scalabili ed efficienti nelle architetture mobili. Per servire i consumatori mobili, la maggior parte dei contributi scientifici estendono le primitive di comunicazione dell’Information-Centric Networking tramite metodi basati sul pull, secondo il quale il consumatore mobile invia richieste ogni volta che raggiunge un nuovo punto di accesso alla rete. Questo approccio genera due importanti problemi. Innanzitutto, le richieste consegnate prima dell'handover genereranno percorsi obsoleti con informazioni di instradamento errate nei router di rete. Di conseguenza, alcuni nuovi contenuti verranno consegnati anche nei punti di accesso precedenti, sprecando banda ed energia. In secondo luogo, durante gli handover, i consumatori mobili potrebbero perdere alcuni contenuti rilasciati in tempo reale. Per risolvere questi problemi, questo lavoro concepisce una nuova architettura protocollare che sfrutta opportunamente le funzionalità dei paradigmi Information-Centric Networking, Multi-access Edge Computing e Software Defined Networking. L'impatto dell'architettura protocollare ideata sull'overhead di comunicazione, sulla memoria dei router e sul consumo di energia è modellato e valutato analiticamente in scenari con diverse condizioni di topologia, mobilità, natura dell’applicazione e numero di consumatori mobili. I test condotti dimostrano l'efficacia dell'architettura protocollare proposta, attraverso simulazioni al computer, in tutti gli scenari considerati.
Information-Centric Networking emerged as one of the most promising technologies of the Future Internet and a powerful enabling technology for the provisioning of scalable and efficient services in mobile architectures. To serve mobile consumers, most of scientific contributions extend the information-centric communication primitives by means of a pull-based methodology, according to which the mobile consumer issues pending requests every time it reaches a new network attachment point. This approach generates two important shortcomings. First, the requests delivered before the handover will generate stale paths with wrong forwarding information in their network routers. As a consequence, some new contents will be delivered also to previous locations, thus wasting bandwidth and energy. Second, during handovers, mobile consumers may miss some contents released in real-time. In order to solve these issues, this work conceives a novel protocol architecture that properly customizes the functionalities of Information-Centric Networking, Multi-access Edge Computing, and Software Defined Networking paradigms. The impact of the devised protocol architecture on the communication overhead, router memory, and energy consumption is analytically formulated and evaluated in scenarios with different topology, mobility, application settings, and number of mobile consumers. The conducted tests demonstrate the effectiveness of the proposed protocol architecture, through computer simulations, in all the considered scenarios.

Tese de doutoramento em cotutela, na área de Ciências e Tecnologias da Informação, apresentada à Faculdade de Ciências e Tecnologia da Universidade de Coimbra e à University of Bern.
With the recent advances in mobile technology, such as the boom in the usage of smartphones and mobile networks, content demand of mobile users has increased significantly. This exponential increase exposed several limitations on the current mechanisms for content delivery. Namely, the current paradigm for requesting content focuses on resources and not on content as it would be ideal to improve content delivery. Despite the existing efforts to overcome this limitation that greatly affects overall performance and efficiency, there are still open challenges that need to be addressed. The first challenge is how to explore new technologies together with a new content request paradigm, aiming at having deeper integration with existing networks and availability of compute, storage and network resources whenever and wherever they are necessary to handle different amounts of user loads. The second challenge is dealing with the integration of content delivery mechanisms with mobile networks and all its particularities, such as constrained architectures and demanding processing requirements. The third challenge is the usage of multiple radio technologies in a transparent and coordinated manner to improve overall efficiency and performance of mobile networks. The fourth challenge is the extension of content caching to the edge of mobile networks with efficient usage of storage resources and reduced latency for content delivery. The fifth and last challenge is the proper distribution of content among edge caches ensuring adaptability to the mobility of users. The key contributions of this thesis aim at addressing those challenges, by providing an integrated architecture with a set of strategies, mechanisms and algorithms that tackle the identified problems in detail and span across multiple knowledge domains. A first contribution concerns a cloud-based system for content delivery, which is easily deployable in new locations, integrates with other services and adapts itself to different user loads. Later on, the integration of this system with mobile networks is depicted to bring the enhancements of that system directly to content delivery in mobile networks, and mechanisms to ensure that it is feasible and follows current standards and specifications are described. Afterwards, and considering that multiple network technologies can be used simultaneously, strategies are proposed to efficiently handle load balancing and offloading of content delivery between different radio technologies in mobile networks, ensuring complete transparency for end users and efficient usage of available resources. Next, and because the previous contributions make caching at the edge of mobile networks a reality, caching strategies for the edge of mobile networks are highlighted, focusing on maximizing performance in terms of latency reduction while optimizing storage usage. Finally, content distribution strategies for edge caches based on users' movement and their interests are presented. These aim at improving edge caching by trying to guarantee that content is cached where it yields the greatest benefits for nearby users. Results gathered from the evaluation of the contributions of this thesis demonstrate that they bring major benefits for content delivery and that performance is greatly enhanced. These improvements are very important from multiple perspectives, as all the involved stakeholders, from mobile users to content providers and mobile operators, may benefit at different levels such as quality of experience, satisfaction and costs.
Com os recentes avanços na tecnologia móvel, tais como o aumento estrondoso da utilização de smartphones e redes móveis, a procura de conteúdos por parte de utilizadores móveis tem aumentado significativamente. Este aumento exponencial expôs várias limitações nos mecanismos actuais para a entrega de conteúdos. Nomeadamente, o actual paradigma para solicitar conteúdos foca-se em recursos e não nos conteúdos como seria ideal para melhorar a entrega de conteúdos. Apesar dos esforços já existentes para ultrapassar esta limitação que afecta de forma substancial e genérica a performance e eficiência, ainda existem desafios em aberto para serem endereçados. O primeiro desafio é como explorar novas tecnologias em conjunto com um novo paradigma para pedido de conteúdos, com o objectivo de ter maior integração com as redes existentes e a disponibilidade de recursos de computação, armazenamento e rede quando e onde forem necessários, para assim acomodar diferentes quantidades de carga de utilizadores. O segundo desafio prende-se com a integração de mecanismos para entrega de conteúdos nas redes móveis e todas as particularidades que isso acarreta, nomeadamente arquitecturas limitadas e requisitos de processamento exigentes. O terceiro desafio é a utilização de múltiplas tecnologias de rádio de forma transparente e coordenada para aumentar de forma global a eficiência e performance das redes móveis. O quarto desafio é a extensão de caches de conteúdos até à orla das redes móveis com vista a uma utilização eficiente de recursos de armazenamento e latências reduzidas na entrega de conteúdos. O quinto e último desafio é a distribuição adequada de conteúdos entre caches na orla das redes móveis, assegurando adaptabilidade à mobilidade dos utilizadores. As contribuições chave desta tese têm como objectivo endereçar os desafios referidos anteriormente, providenciando uma arquitectura integrada com um conjunto de estratégias, mecanismos e algoritmos que tratam dos problemas identificados em detalhe e abrangem múltiplos domínios do conhecimento. Uma primeira contribuição diz respeito a um sistema baseado na nuvem para entrega de conteúdos, o qual é facilmente implementável em novas localizações, integra com outros serviços e adapta-se a si próprio a diferentes cargas de utilizadores. De seguida, é apresentada a integração deste sistema com redes móveis e são descritos mecanismos para garantir a sua praticabilidade e adequação às normas e padrões existentes, para assim trazer as suas melhorias directamente à entrega de conteúdos em redes móveis. Depois, e considerando que múltiplas tecnologias de rede podem ser utilizadas em simultâneo, são apresentadas estratégias para efectuar balanceamento e transferência de carga entre múltiplas tecnologias de rádio em redes móveis, assegurando total transparência para os utilizadores finais e uma utilização eficiente dos recursos existentes. Mais tarde, e porque as contribuições anteriores tornam a existência de caches na orla das redes móveis uma realidade, estratégias para caching na orla das redes móveis são realçadas, focando-se em maximizar a performance em termos de redução de latência e na optimização da utilização de recursos de armazenamento. Finalmente, estratégias para a distribuição de conteúdos em caches na orla das redes móveis baseadas em movimento e interesses dos utilizadores são apresentadas. Estas visam a melhoria das caches na orla das redes móveis ao tentar garantir que os conteúdos são colocados em caches onde irão trazer os maiores benefícios possíveis para utilizadores na imediação. Os resultados obtidos da avaliação das contribuições desta tese demonstram que as mesmas trazem benefícios relevantes para a entrega dos conteúdos juntamente com um aumento significativo da performance. Estas melhorias são muito importantes de múltiplas perspectivas, dado que todas as partes interessadas, desde utilizadores móveis até produtores de conteúdos e operadores móveis, podem beneficiar a diferentes níveis tais como qualidade da experiência, satisfação e custos.
Universidade de Berna, Suíça
OneSource, Consultoria Informática, Lda.

Centralized Technical Management (CTM) Systems allow the control and monitoring of various devices installed on the factory floor or in other environments, such as offices, buildings, or hospitals. These devices can capture data about the medium (sensors) or interact with the same medium (actuators). Temperature sensors, presence detectors, or energy meters can be used to control HVAC systems, lighting, or even trigger an alarm in an emergency. With the increasing Internet of Things (IoT) and Industrial Internet of Things (IIoT) applications, it becomes necessary to use effective CTM systems. With these systems, a technician can manage multiple devices in a centralized unit without having to be in direct contact with the device and possibly without having to travel to the building where the instruments are. There are several setbacks when designing a GTC system. One of them is that the devices to be controlled use different protocols to communicate with each other. With this in mind, within the scope of this work, we developed a framework for developing CTM systems. This framework allows all devices connected to the central unit to be controlled through a graphical interface in the same way. That is, the system creates a layer that abstracts the communication protocol used by the various devices. In this work, tools such as EdgeX Foundry, InfluxDB, Telegraf, and Grafana were used to implement a framework. The functioning of the framework was validated using commercial devices (KNX and Modbus protocols) and a device developed from scratch (MQTT protocol). Additionally, a mechanism for prioritizing messages considered critical, which uses an IP protocol, was implemented. This mechanism allows a specific bandwidth to be reserved for the desired protocol. To this end, Software-Defined Network and OpenFlow principles were used to implement a mechanism that prioritizes MQTT packages. Two setups were compared to test the implemented system, with and without the message prioritization mechanism.
Sistemas de Gestão Técnica Centralizada (GTC) permitem o controlo e monitorização de diversos dispositivos instalados em chão de fábrica ou em outros ambientes, como por exemplo, escritórios, prédios ou hospitais. Estes dispositivos podem capturar dados sobre o meio onde estão instalados (sensores) ou até interagir com o mesmo meio (atuadores). Sensores de temperatura, detetores de presença ou medidores de energia podem ser utilizados para controlar sistemas de AVAC, iluminação, ou até disparar um alarme em caso de emergência. Com o aumento de aplicações baseadas em Internet of Things (IoT) e Industrial Internet of Things (IIoT), torna-se necessária a utilização de sistemas de GTC eficazes. Com estes sistemas, um técnico pode gerir múltiplos dispositivos numa unidade centralizada, sem ter de estar em contacto direto com o dispositivo e, possivelmente, sem ter de se deslocar ao edifício onde os instrumentos se encontram. Existem várias contrariedades quando se projeta um sistema de GTC, sendo uma o facto de os dispositivos que se pretende controlar utilizarem diferentes protocolos para comunicarem entre si. Com isto em mente, no âmbito deste trabalho foi desenvolvida uma framework para o desenvolvimento de sistemas de GTC. Esta framework permite que todos os dispositivos ligados à unidade central sejam controlados através de uma interface gráfica, de igual forma. Ou seja, o sistema cria uma camada que abstrai o protocolo de comunicação utilizado pelos diversos dispositivos. Neste trabalho foram utilizadas ferramentas como EdgeX Foundry, InfluxDB, Telegraf e Grafana para implementar a framework. O funcionamento da framework foi validado utilizando dispositivos comerciais (protocolo KNX e Modbus) e um dispositivo desenvolvido de raiz (protocolo MQTT). Adicionalmente, foi implementado um mecanismo de priorização de mensagens consideradas críticas, que utilizem um protocolo IP. Este mecanismo permite que uma determinada largura de banda seja reservada para o protocolo desejado. Para tal, foram utilizados princípios de Software-Defined Networking e OpenFlow para implementar um mecanismo que prioriza os pacotes MQTT. Para testar o sistema implementado foram comparados dois setups, com e sem o mecanismo de priorização de mensagens.
Mestrado em Engenharia Eletrónica e Telecomunicações