Submitted by JÚLIO HEBER SILVA (julioheber@yahoo.com.br) on 2016-12-02T18:19:39Z No. of bitstreams: 2 Dissertação - Fausto da Silva Moraes - 2016.pdf: 3698287 bytes, checksum: cbcf2894f028c54e60d90033a04bdc5e (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Jaqueline Silva (jtas29@gmail.com) on 2016-12-13T15:32:00Z (GMT) No. of bitstreams: 2 Dissertação - Fausto da Silva Moraes - 2016.pdf: 3698287 bytes, checksum: cbcf2894f028c54e60d90033a04bdc5e (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2016-12-13T15:32:00Z (GMT). No. of bitstreams: 2 Dissertação - Fausto da Silva Moraes - 2016.pdf: 3698287 bytes, checksum: cbcf2894f028c54e60d90033a04bdc5e (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2016-11-09 / For the future of cellular networks is estimated a significant increase in the number of connected devices and the bulk data traffic, especially video content. That poses as a challenge for the next generation network. With the rise of new communications paradigm, D2D communication emerge as potential approach for cellular network data offloading, especially when paired with caching solutions within the network. Also, the impact of viral videos could be mitigated by proactively caching the contents being shared on Online Social Networks (OSNs). This work presents a new approach to proactive content cache for D2D-enabled networks, that is aware of users social interaction on OSN’s. Our proposal consists in the combination of user mobile and social informations to find the best located device to cache a content being shared online. Results obtained through simulation show that the proposed approach can improve offload rate, reduce energy consumption and provide a faster content access when compared with other work in literature. / As estimativas de um crescimento significativo do número de dispositivos conectados às redes sem fio e móveis, e do aumento do volume de dados trafegados, em especial o tráfego de conteúdo em vídeo, representam um desafio para a próxima geração de redes celulares. Com o surgimento de novos paradigmas de redes sem fio, a comunicação D2D (Dispositivo-a-Dispositivo) figura como uma abordagem promissora para permitir o descarregamento de dados da rede celular, principalmente quando aliada a técnicas de cache de conteúdo nos dispositivos dos usuários. Além disso, as informações de interações dos usuários em redes sociais online poderiam ser empregadas para suavizar o impacto dos vídeos virais sendo compartilhados entre os usuários, através de uma solução de cache proativa. Nesta dissertação, apresentamos uma proposta de cache proativo de conteúdo ciente às interações sociais online para redes com comunicação D2D. Em nossa proposta, determinamos de forma probabilística qual o melhor dispositivo para armazenar um conteúdo que esteja sendo compartilhado entre os usuários com base nas informações de contatos entre os usuários na rede D2D. Os resultados coletados com o simulador ns-3 mostram que a proposta apresentada pode melhorar o descarregamento de dados da rede celular, reduzir os gastos de energia dos dispositivos dos usuários, e fornecer tempos de transmissão de conteúdo menores, em comparação com outro trabalho da literatura,

Les réseaux fixes et mobiles font face à une croissance dramatique du trafic de données, qui est principalement due à la distribution de contenus vidéo. Les opérateurs Télécoms envisagent donc de décentraliser la distribution de contenus dans les futures architectures convergées fixe-mobile (FMC). Cette décentralisation, conjointement au déploiement d'un cœur de réseau mobile distribué, sera un élément majeur des futurs réseaux 5G. L'approche SIPTO définie par 3GPP permet déjà le délestage sur le réseau fixe du trafic mobile, et pourra donc être utilisée en 5G. SIPTO s'appuie sur la distribution des passerelles de données (PGW) qui permet ainsi de décharger le cœur du réseau mobile actuel. Cependant, dans certains cas de mobilité des usagers, SIPTO ne supporte pas la continuité de session, quand il est nécessaire de changer de PGW, donc de modifier l'adresse IP du terminal. Cette thèse commence par quantifier le gain apporté par le délestage du trafic mobile en termes de capacité requise pour différentes portions du réseau. Un état de l'art des différentes solutions de délestage du trafic de données mobiles est fourni, démontrant qu'aucune des solutions existantes ne supporte la continuité de service pour les sessions de longue durée. C'est pourquoi, cette thèse propose des solutions pour supporter une mobilité transparente ; ces solutions s'appuient à la fois sur SIPTO et sur le protocole MultiPath TCP (MPTCP). Les protocoles du 3GPP sont inchangés car la continuité est maintenue par les extrémités. Enfin, ces solutions sont appliquées aux différentes implémentations d'architectures FMC envisagées à ce jour. / Fixed and mobile networks are currently experiencing a dramatic growth in terms of data traffic, mainly driven by video content distribution. Telecoms operators are thus considering de-centralizing content distribution architecture for future Fixed and Mobile Converged (FMC) network architectures. This decentralization, together with a distributed mobile EPC, would be used for future 5G networks. Mobile data offloading, in particular SIPTO approaches, already represent a good implementation model for 5G network as it allows the use of distributed IP edges to offload Selected IP traffic off the currently centralized mobile core network. However, in some cases, SIPTO does not support session continuity during users' mobility. This is due to the fact that user's mobility may imply packet gateway (PGW) relocation and thus a modification of the UE's IP address.This PhD thesis first quantifies the gain, in terms of bandwidth demands on various network portions, brought by the generalized use of mobile traffic offloading. A state of art of existing mobile data offloading solutions is presented, showing that none of the existing solutions solve the problem of session continuity for long-lived sessions. This is why, in the context of future FMC mobile network architectures, the PhD thesis proposes solutions to provide seamless mobility for users relying on SIPTO with the help of Multipath TCP (MPTCP). 3GPP standards are not modified, as session continuity is ensured by end-points. Lastly, the proposed solutions are mapped on different architecture options considered for future FMC networks.

Nowadays more and more data is generated at the edge of the network, and people are beginning to consider decentralizing computing tasks to the edge of the network. The network architecture of edge computing is different from the traditional network architecture. Its distributed configuration can make up for some shortcomings of traditional networks, such as data congestion, increased delay, and limited capacity. With the continuous development of 5G technology, satellite communication networks are also facing many new business challenges. By using idle computing power and storage space on satellites and integrating edge computing technology into satellite communication networks, it will greatly improve satellite communication service quality, and enhance satellite task processing capabilities, thereby improving the satellite edge computing system performance. The primary problem that limits the computing performance of satellite edge networks is how to obtain a more effective dynamic service offloading strategy. To study this problem, this thesis monitors the status information satellite nodes in different periods, such as service load and distance to the ground, uses the Markov decision process to model the dynamic offloading problem of the satellite edge computing system, and finally obtains the service offloading strategies. The deployment plan is based on deep reinforcement learning algorithms. We mainly study the performance of the Deep Q-Network (DQN) algorithm and two improved DQN algorithms Double DQN (DDQN) and Dueling DQN (DuDQN) in different service request types and different system scenarios. Compared with existing service deployment algorithms, deep reinforcement learning algorithms take into account the long-term service quality of the system and form more reasonable offloading strategies. / Med den snabba utvecklingen av mobil kommunikationsteknik genereras mer och mer data i utkanten av nätverket, och människor börjar överväga att decentralisera datoruppgifter till kanten av nätverket. Och byggde ett komplett mobilt edge computing -arkitektursystem. Edge -dators nätverksarkitektur skiljer sig från den traditionella nätverksarkitekturen. Dess distribuerade konfiguration kan kompensera för eventuella brister i traditionella nätverk, såsom överbelastning av data, ökad fördröjning och begränsad kapacitet. Med den ständiga utvecklingen av 5G -teknik står satellitkommunikationsnät också inför många nya affärsutmaningar. Genom att använda inaktiv datorkraft och lagringsutrymme på satelliter och integrera edge computing -teknik i satellitkommunikationsnät kommer det att förkorta servicetiden för traditionella mobila satelliter kraftigt, förbättra satellitkommunikationstjänstkvaliteten och förbättra satellituppgiftsbehandlingsförmågan och därigenom förbättra satelliten edge computing systemprestanda. Det primära problemet som begränsar datorprestanda för satellitkantnät är hur man får en mer effektiv dynamisk tjänstavlastningsstrategi. Detta papper övervakar servicebelastningen av satellitnoder i olika perioder, markpositionsinformation och annan statusinformation använder Markov - beslutsprocessen för att modellera den dynamiska distributionen av satellitkantstjänster och får slutligen en uppsättning tjänstedynamik baserad på modell och design . Distributionsplanen är baserad på en djupt förbättrad algoritm för dynamisk distribution av tjänster. Det här dokumentet studerar huvudsakligen prestandan för DQN -algoritmen och två förbättrade DQN - algoritmer Double DQN och Dueling DQN i olika serviceförfrågningstyper och olika systemscenarier. Jämfört med befintliga algoritmer för serviceutplacering är prestandan för algoritmer för djupförstärkning något bättre.

The growing number of players in Massively Multiplayer Online games puts a heavy load on the network infrastructure and the general-purpose CPU of the game servers. A game server’s network stack processing needs equal treatment to the game-related processing ability. It is a fact that the networkcommunication tasks on the CPU reach the same order of magnitude as the game-related tasks, and the computing capability of the CPU can be a factor that limits the maximum number of players. Therefore, CPU offloading is becoming vital. FPGAs play an essential role in dedicated computation and network communication due to their superiority in flexibility and computation-oriented efficiency. Thus, an FPGA can be a good hardware platform to implement a network stack to replace the CPU in processing the network computations. However, most commercial and open-source network stack IPs support only one or few connections. This thesis project explores a network server on FPGA, implemented in RTL, that can handle multiple connections, specialized in the TCP protocol. The design in this project adds a cached memory hierarchy that provides a filter against port numbers of multiple connections from the same application and an Application Layer Controller, based on an open-source Ethernet, to increase the number of TCP connections further. A proof of concept was built, and its performance was tested. As a result, the TCP server on the FPGA was designed to handle a maximum of 40 configurable connections, but only 25 connections could be maintained during operation due to operational latency constraints. This FPGA server solution provides a latency of 1 ms in LAN. The babbling idiot and out-of-order packet transfer tests from clientswere also performed to guarantee robustness. During testing, poor performance in Packet Loss and Packet Error Handling was noted. In the future, this issue needs to be addressed. In addition, further investigations of methods for expanding the cache need to be done to allow handling more clients. / Det växande antalet spelare i Massively Multiplayer Online-spel belastar nätverksinfrastrukturen och spelservrarnas CPU:er. En spelservers förmåga att bearbeta nätverksstacken måste behandlas lika med den spelrelaterade bearbetningsförmågan. Det är ett faktum att nätverkskommunikationsuppgifterna på processorn når samma storleksordning som de spelrelaterade uppgifterna, och processorns beräkningsförmåga kan vara en faktor som begränsar det maximala antalet spelare. Därför blir avlastning av CPU-viktig. FPGA:er spelar en viktig roll i dedikerad beräkning och nätverkskommunikation på grund av dess överlägsenhet vad gäller flexibilitet och beräkningsorienterad effektivitet. Således kan en FPGA vara en bra hårdvaruplattform för att implementera en nätverksstack, för att ersätta CPU:n vid bearbetning av nätverksberäkningsarna. Men, de flesta kommersiella och öppna källkodsnätverksstack- IP:er stöder dock bara en eller ett fåtal anslutningar. Detta examensarbete utforskar en nätverksserver på FPGA, implementerad mha RTL, som kan hantera flera anslutningar, specialiserad på TCP-protokollet. Designen i detta projekt lägger till en cachad minneshierarki som ger ett filter mot portnummer för flera anslutningar från samma applikation och en Application Layer Controller, baserad på öppen källkod för Ethernet, för att öka antalet TCP-anslutningar ytterligare. Ett proof of concept byggdes och dess prestanda testades. Som ett resultat designades TCP-servern på FPGA:n att kunna hantera maximalt 40 konfigurerbara anslutningar, men endast 25 anslutningar kunde bibehållas under drift på grund av driftsfördröjningar. Denna FPGA-serverlösning ger en latens på 1 ms i LAN. Tester inkluderande den babblande idioten och out-of-order paketöverföring från klienter utfördes också för att garantera robusthet. Under testningen noterades dålig prestanda i paketförlust och paketfelshantering. I framtiden måste denna fråga åtgärdas. Dessutom behöver ytterligare undersökningar av metoder för att utöka cachen göras för att kunna hantera fler klienter.

In Human-Robot Collaboration scenarios, safety must be ensured by a risk management process that requires the execution of computationally expensive perception models (e.g., based on computer vision) in real-time. However, robots usually have constrained hardware resources that hinder timely responses, resulting in unsafe operations. Although Multi-access Edge Computing allows robots to offload complex tasks to servers on the network edge to meet real-time requirements, this might not always be possible due to dynamic changes in the network that can cause congestion or failures. This work proposes a safety-based task offloading strategy to address this problem. The goal is to intelligently use edge resources to reduce delays in the risk management process and consequently enhance safety. More specifically, depending on safety and network metrics, a Reinforcement Learning (RL) solution is implemented to decide whether a less accurate model should run locally on the robot or a more complex one should run remotely on the network edge. A third possibility is to reuse the previous output through verification of temporal coherence. Experiments are performed in a simulated warehouse scenario where humans and robots have close interactions. Results show that the proposed RL solution outperforms the baselines in several aspects. First, the edge is used only when the network performance is good, reducing the number of failures (up to 47%). Second, the latency is also adapted to the safety requirements (risk X latency reduced up to 48%), avoiding unnecessary network congestion in safe situations and letting other robots in hazardous situations use the edge. Overall, the latency of the risk management process is largely reduced (up to 68%), and this positively affects safety (time in safe zone increased up to 3:1%). / I scenarier med människa-robotkollaboration måste säkerheten säkerställas via en riskhanteringsprocess. Denna process kräver exekvering av beräkningstunga uppfattningsmodeller (t.ex. datorseende) i realtid. Robotar har vanligtvis begränsade hårdvaruresurser vilket förhindrar att respons uppnås i tid, vilket resulterar i osäkra operationer. Även om Multi-access Edge Computing tillåter robotar att avlasta komplexa uppgifter till servrar på edge, för att möta realtidskraven, så är detta inte alltid möjligt på grund av dynamiska förändringar i nätverket som kan skapa överbelastning eller fel. Detta arbete föreslår en säkerhetsbaserad uppgiftsavlastningsstrategi för att hantera detta problem. Målet är att intelligent använda edge-resurser för att minska förseningar i riskhanteringsprocessen och följaktligen öka säkerheten. Mer specifikt, beroende på säkerhet och nätverksmätvärden, implementeras en Reinforcement Learning (RL) lösning för att avgöra om en modell med mindre noggrannhet ska köras lokalt eller om en mer komplex ska köras avlägset på edge. En tredje möjlighet är att återanvända sista utmatningen genom verifiering av tidsmässig koherens. Experimenten utförs i ett simulerat varuhusscenario där människor och robotar har nära interaktioner. Resultaten visar att den föreslagna RL-lösningen överträffar baslinjerna i flera aspekter. För det första används edge bara när nätverkets prestanda är bra, vilket reducerar antal fel (upp till 47%). För det andra anpassas latensen också till säkerhetskraven (risk X latens reducering upp till 48%), undviker onödig överbelastning i nätverket i säkra situationer och låter andra robotar i farliga situationer använda edge. I det stora hela reduceras latensen av riskhanterings processen kraftigt (upp till 68%) och påverkar på ett positivt sätt säkerheten (tiden i säkerhetszonen ökas upp till 4%).

Les technologies de l'Internet des objets (IdO) et de l'externalisation des services informatiques ont conduit à l'émergence de nouvelles menace sur la vie privée des utilisateurs. Cependant, l'implémentation des moyens de sécurité traditionnels sur les équipements de l'IdO constitue un premier défi lié aux limites de capacités. D'un autre côté, la délégation du traitement et du stockage des données, nous pose le problème de confiance envers les fournisseurs de service.Dans ce contexte, nous avons proposé une solution de chiffrement qui assure une protection de données centrée sur leurs propriétaires et adaptée à l'environnement contraignant des objets connectés. Ce modèle se base sur le chiffrement par attributs avec externalisation sécurisée et la technologie de la Blockchain. Ensuite, en réponse à la problématique de la confiance et de la sélection du service, nous avons exploré les possibilités offertes par les outils de l'intelligence artificielle. Pour ce faire, nous avons proposé un modèle de filtrage collaboratif basé sur les cartes de Kohonen avec une solution pour détecter les utilisateurs non fiable. / Internet of Things (IoT) and IT service outsourcing technologies have led to the emergence of new threats to users' privacy. However, the implementation of traditional security measures on IoT equipment is a first challenge due to capacity limitations. On the other hand, the offloading of data processing and storage poses the problem of trust in service providers.In this context, we have proposed an encryption solution that provides owner-centric data protection adapted to the constraining environment of IoT. This model is based on attribute-based encryption with secure offloading capability and Blockchain technology. Then, in response to the issue of trust and service selection, we explored the possibilities offered by artificial intelligence tools. To do this, we proposed a collaborative filtering model based on Kohonen maps and efficient solution to detect the untrusted users.

Um Systeme der fünften Generation zellularer Kommunikationsnetze (5G) zu ermöglichen, sind Energie effiziente Architekturen erforderlich, die eine zuverlässige Serviceplattform für die Bereitstellung von 5G-Diensten und darüber hinaus bieten können. Device Enhanced Edge Computing ist eine Ableitung des Multi-Access Edge Computing (MEC), das Rechen- und Speicherressourcen direkt auf den Endgeräten bereitstellt. Die Bedeutung dieses Konzepts wird durch die steigenden Anforderungen von rechenintensiven Anwendungen mit extrem niedriger Latenzzeit belegt, die den MEC-Server allein und den drahtlosen Kanal überfordern. Diese Dissertation stellt ein Berechnungs-Auslagerungsframework mit Berücksichtigung von Energie, Mobilität und Anreizen in einem gerätegestützten MEC-System mit mehreren Benutzern und mehreren Aufgaben vor, das die gegenseitige Abhängigkeit der Aufgaben sowie die Latenzanforderungen der Anwendungen berücksichtigt. / To enable fifth generation cellular communication network (5G) systems, energy efficient architectures are required that can provide a reliable service platform for the delivery of 5G services and beyond. Device Enhanced Edge Computing is a derivative of Multi-Access Edge Computing (MEC), which provides computing and storage resources directly on the end devices. The importance of this concept is evidenced by the increasing demands of ultra-low latency computationally intensive applications that overwhelm the MEC server alone and the wireless channel. This dissertation presents a computational offloading framework considering energy, mobility and incentives in a multi-user, multi-task device-based MEC system that takes into account task interdependence and application latency requirements.

Introduction: Instead of replacing existing analogue water meters with Internet of Things (IoT) connected substitutes, an alternative would be to attach an IoT connected module to the analogue water meter that optically reads the meter value using Optical Character Recognition (OCR). Such a module would need to be battery-powered given that access to the electrical grid is typically limited near water meters. Research has shown that offloading the OCR process can reduce the power dissipation from the battery, and that this dissipation can be reduced even further by reducing the amount of data that is transmitted.  Purpose: For the sake of minimising energy consumption in the proposed solution, the purpose of the study is to find out to what extent it is possible to reduce an input image’s file size by means of resolution, colour depth, and compression before the Google Cloud Vision OCR engine no longer returns feasible results.   Method and implementation: 250 images of analogue water meter values were processed by the Google Vision Cloud OCR through 38 000 different combinations of resolution, colour depth, and upscaling.  Results: The highest rate of successful OCR readings with a minimal file size were found among images within a range of resolutions between 133 x 22 to 163 x 27 pixels and colour depths between 1- and 2-bits/pixel.  Conclusion: The study shows that there is a potential for minimising data sizes, and thereby energy consumption, by offloading the OCR process by means of transmitting images of minimal file size.

Note-taking is ubiquitous. Whether we write down a grocery list, type our intentions for the day, or record a voice note for a friend, we often use notes to externalize our thoughts. We “delegate” this information to “in the world” extensions of our cognition, thus lightening our cognitive loads. The current thesis investigates the effects of note-taking as a form of cognitive offloading in an applied research setting. The note-taking literature is fragmented regarding practical note-taking recommendations for educators. The current thesis advances our knowledge and understanding of the crossover between cognitive offloading, note-taking, and individual differences. The laboratory research presented in this thesis uses novel materials that mimic the classroom environment, with future goals of translating this research into the actual classroom. Chapter 2 explores note-taking from a cognitive offloading perspective and demonstrates how differences in note-taking quantity affect recall. Chapter 3 showcases how differences in learning between note-taking modalities are seen sporadically and only when they intersect with the type of test. Chapter 4 investigates the importance of individual differences (e.g., working memory capacity) when exploring cognitive offloading and note-taking, and demonstrates how surface findings are not generalizable once we investigate underlying individual differences. While our research started as a way to understand how we offload information via note-taking and its effects on learning, we hope our findings and general discussion encourage the reader to explore the generalizability of applied cognitive research. Note-taking is a complex process, and our future work aims to investigate how learners differ and how we might disseminate research in education to be inclusive and diverse. / Thesis / Doctor of Philosophy (PhD) / Imagine a world without technology or external resources to express your thoughts: no “to-do” lists, calendar reminders, sketches, or notes. Since our memory capacities are limited, we often maintain a record of information “in the world”, as opposed to only using our limited memory stores. This form of mental delegation is known as “cognitive offloading”. One way we offload information for future access is through note-taking. In an educational setting, note-taking influences our ability to learn and review information. How we take notes, and their effects on learning, have been debated in the literature. This thesis explores the impact of cognitive offloading and note-taking on learning, and demonstrates the importance of exploring individual differences (e.g., memory capacity and note-taking preference) in applied educational research. Throughout this thesis, we prompt our audience to frame note-taking and cognitive research takeaways through an inclusive educational lens.

No description available.