A grande popularidade dos dispositivos móveis que provêm acesso ubíquo à Internet de banda larga, através de redes de rádio, e o volume de tráfego gerado por estes dispositivos estão aumentando a cada ano. Além disso, vem ampliando consideravelmente a frequência com que usuários de dispositivos móveis estão usando serviços baseados na Web. Alguns destes usuários podem estar acessando serviços que precisam de transmissão contínua como, por exemplo, vídeos interativos, outros podem estar apenas lendo e-mails, o que não exige um fluxo contínuo de dados. Mais do que isso, usuários com altos níveis de sinal podem atingir melhores taxas de transferência do que os com níveis menores. Portanto, encontrar a melhor relação entre os usuários que estão acessando serviços sensíveis ao atraso e aqueles que maximizam a taxa de transferência, e ainda ser justo na transmissão, é um relevante desafio para o escalonamento dos recursos de uma rede sem fio. Embora as pesquisas de escalonamento de recursos em redes sem fio tenham evoluído neste sentido, o recente aumento do volume de tráfego mencionado pode levar a uma sobrecarga no sistema, comprometendo o escalonamento. A fim de enfrentar estes desafios, o Orthogonal Frequency Division Multiple Access (OFDMA), tecnologia fundamental para o acesso múltiplo em redes de quarta geração, tem sido considerado também para ser utilizado na próxima geração de rádios móveis. Para implementar um serviço efetivo aos usuários, requisitos, tais como, altas taxas de transferência, tolerância baixa ao atraso, minimização da perda de pacotes e maximização da justiça no escalonamento, devem somar-se à característica, de alta densidade de usuários, que surgiu após o advento da popularização dos dispositivos móveis. Portanto, novas estratégias de escalonamento devem ser idealizadas. Nesta dissertação, deu-se um passo além na proposição de um escalonador para as redes móveis de próxima geração, que busca melhorar a relação entre taxa de transferência e atraso, consequentemente, levando a maiores índices de justiça no escalonamento resultante. O escalonador foi especialmente desenvolvido para lidar com altas densidades de usuários, inerentes às redes modernas, e as redes LTE foram utilizadas como caso de estudo. Desta forma, um novo escalonador ótimo que considera provisão dos requisitos acima mencionados, é modelado. Além disso, uma nova heurística parametrizável, baseada na qualidade do canal do usuário, no atraso permitido por cada serviço e na justiça do escalonamento é proposta, a fim de lidar com cenários sobrecarregados. Resultados demonstram que a abordagem de escalonamento proposta leva a uma taxa de transferência apenas 7,5% menor que os valores ótimos, com 25% a menos de perda de pacotes em cenários sobrecarregados. O modelo também garante que o escalonamento resultante seja pelo menos 0,91 na escala do índice de justiça de Jain. Finalmente, os resultados mostram uma melhor relação entre a eficiência espectral e as métricas de QoS. / The huge popularity of mobile devices that provides a ubiquitous Internet broadband access via radio networks and the volume of traffic generated by these devices in the base stations are increasing every year. Furthermore, the frequency which, mobile users are using web-based services, is increasing, requiring high transfer rates such as transmission of interactive videos. These factors have become the main challenges for the scheduling of radio resources. In order to meet these challenges, the Orthogonal Frequency Division Multiple Access (OFDMA), a key technology for multiple access in fourth generation networks, has also been considered for use in next-generation mobile radios. To implement an effective service to users, requirements such as high transfer rates, lower delay tolerance, minimum packet loss and maximum scheduling fairness, should be added to the requirements that emerged after the advent of the popularity of mobile devices. Therefore, new scheduling strategies should be projected. Despite efforts to solve the downlink (DL) scheduling problem on wireless networks, we are not aware of previous attempts that have addressed the above requirements in a single strategy. In this thesis, we took a step further in this direction and still considering the high densities in small cells inherent in modern networks. In additional, we address the radio DL resource scheduling problem for multiple users using LTE networks as a case study. A new optimal scheduler is modeled regarding Quality of Service (QoS) provisioning. In addition, a parameterized heuristic based on user channel quality and service delay is proposed to reach scheduling solutions for overbooked scenarios. Results demonstrate that the proposed scheduling approaches led to a throughput of 7.5% lower than the optimal ones and 25% lower packet losses in overloaded scenarios. Our model also ensures that the resultant scheduling is at least as fair as 0.91 in Jain fairness index. Additionally, the obtained results show a reasonable trade-off between spectral efficiency and QoS metrics.

Applications with timing requirements, such as multimedia and live multi-user interaction, are becoming more prevalent in wide area networks. The desire to provide more predictable performance for such applications in packet switched wide area networks is evident in the channel management provided by Asynchronous Transfer Mode (ATM) networks and in the extensions to the Internet protocols proposed by the Internet Engineering Task Force (IETF) working groups on integrated and differentiated service. The ability to provide guarantees on the performance of traffic flows, such as packet delay and loss characteristics, relies on an accurate model of the traffic arrival and service at each node in the network. This thesis surveys the work in bounding packet delay based on various proposed queuing disciplines and proposes a method for more accurately defining the traffic arrival and worst case backlog experienced by packets. The methods are applied to the first in first out (FIFO) queuing discipline to define equations for determining the worst case backlog and queuing delay in multihop networks. Simulation results show a significant improvement in the accuracy of the delay bounds over existing bounds published in the literature. An improvement of two orders of magnitude can be realised for a ten hop path and the improvement increases exponentially with the length of the path for variable rate network traffic. The equations derived in the thesis also take into consideration the effect of jitter on delay, thereby removing the requirement for rate controllers or traffic shaping within the network. In addition to providing more accurate delay bounds, the problem of providing fault tolerance to channels with guaranteed quality of service (QoS) is also explored. This thesis introduces a method for interleaving resource requirements of backup channels to reduce the overall resource reservations that are required to provide guaranteed fault recovery with the same QoS as the original failed channel. An algorithm for selecting recovery paths that can meet a channel's QoS requirements during recovery is also introduced. / Thesis (Ph.D.)--Computer Science, 2003.

Avec les progrès réalisés dans les réseaux sans fil et mobiles, chercheurs et industriels ont déjà commencé le développement des futurs réseaux mobiles de 4ème génération (4G). Dans cette génération, l'équipement terminal, de l'utilisateur recherchera en permanence le meilleur réseau d'accès en fonction des besoins de l'utilisateur. Ce dernier aura plusieurs technologies d'accès sans fil à sa disposition et souhaitera être connecté au mieux, n'importe où, n'importe quand et avec n'importe quel réseau d'accès. Pour cela, les différentes technologies sans fil qui seront présentes dans la 4g (UMTS, WLAN, WMAN, etc), doivent coexister. Notre travail dans cet axe de recherche consiste à proposer une structure hiérarchique des réseaux sans fil et des protocoles de handovers entre les différents systemes. Le changement de réseau doit être rapide et sans perte de données afin de garantir la transparence vis-à-vis des utilisateurs mobiles. Les réseaux ad hoc sont considérés comme généralisation ultime des réseaux sans fil car ils limitent au maximum le rôle de l'infrastructure fixe. Ces dernières années, on observe des progrès significatifs pour étudier la QoS dans les réseaux ad hoc mobiles. Cette thèse présente une vision critique des travaux intéressants dans le domaine de la QoS dans les réseaux ad hoc. Les points forts et faibles de chaque proposition sont exploités pour réaliser un modèle complet de QoS. Ce modèle possède un protocole de routage avec QoS qui trouve la meilleure route pour chaque type de flux selon la demande. Il limite l'apparition des congestions et la surutilisation de la bande passante par un mécanisme de contro^le d'admission et de réservation.

With the advent of portable devices (e.g., smartphones) and the advances in wireless networking technologies (e.g., WLAN, GPRS, UMTS), the vision of ubiquitous computing is becoming a reality. It aims to facilitate user tasks through the seamless utilization of heterogeneous computing and communication capabilities (represented as services) available in the environment. Service discovery, which is necessary for achieving the above goal, must be aware of the service's non-functional properties due to the challenges posed by ubiquitous computing, such as device portability and mobility. This thesis proposes an overall solution that supports QoS-aware service discovery in ubiquitous computing environments. Our contribution lies in substantiating QoS awareness in the following three aspects. Firstly, during the process of discovering services, the expiring wireless links resulting from device mobility are identified and avoided since they cause service failures and thus hamper service reliability. Secondly, as mu tiple services can be discovered, a comprehensive utility function is proposed to evaluate services in terms of their various non-functional properties, meanwhile taking into account the service user's preferences among them, for the purpose of selecting the best one. Thirdly, to avoid untrustworthy services, a distributed reputation mechanism is proposed to facilitate the evaluation of the service host's trustworthiness. The above three proposed solutions are extensively evaluated respectively, based on analysis and simulation. They are further incorporated into a middleware that supports QoS aware Web service discovery in ubiquitous computing environments. A prototype implementing the middleware is deployed and evaluated. The results show that the overhead introduced by QoS awareness seems reasonable.

<p>Many efforts have been carried out to provide transmission reliability in the history of communication systems. As the demand for real-time applications increased, providing a reliable communication in a timely manner for such applications is strongly desired. Considering timing constraints makes the issue of achieving reliability more difficult. This thesis concentrates on providing reliability for real-time communication in embedded networks by achieving a timing analysis and using the ARQ concept. What is carried out in this thesis is providing retransmission in a real-time manner for embedded networks according to application request. The thesis work focuses on one packet retransmission over a point to point link, but the concept is rich and can be extended to cover application request in real-time embedded networks. Two methods have been fulfilled, and a simulation has been done on the timing analysis focusing on the performance in accepting real-time traffic in the form of separate channels for each application request. The protocol combines ARQ and a scheduling algorithm as a base to support retransmission for hard real-time applications in embedded networks.</p>

Currently, a viable solution to provide multicast provision over a multiprotocol label switch with traffic engineering (MPLS-TE) domain is unavailable because of the missing link able to couple multicast traffic distribution with an MPLS-TE enabled network. This is due to the limited or less research investigation that was done in this area. Most of the investigation methods tackle the problem individually such as deploying internet protocol (IP) multicast in a plain network or MPLS domain but without considering a combination of both technologies that is aware of differentiated services requirements. This thesis presents an alternative solution for the multicast differentiated services provision problem over a MPLS-TE enabled network. The approach is exposed and analyzed through a practical solution that was developed within a network simulation environment. The research presented in this thesis orchestrates the already available technologies offered by the multicast protocols suite and differentiated services (DiffServ) aware MPLS-TE that allows applying separately the constraint-based routing and admission control to different classes of services. The novelty and solution of this thesis relies on using MPLS constraint-based routing concepts (e.g.traffic trunks) in order to solve TE issues revealed during multicast traffic distribution.

The QoS Steiner Tree Problem asks for the most cost efficient way to multicast multimedia to a heterogeneous collection of users with different data consumption rates. We assume that the cost of using a link is not constant but rather depends on the maximum bandwidth routed through the link. Formally, given a graph with costs on the edges, a source node and a set of terminal nodes, each one with a bandwidth requirement, the goal is to find a Steiner tree containing the source, and the cheapest assignment of bandwidth to each of its edges so that each source-to-terminal path in the tree has bandwidth at least as large as the bandwidth required by the terminal. Our main contributions are: (1) New flow-based integer linear program formulation for the problem; (2) First implementation of 4.311 primal-dual constant factor approximation algorithm; (3) an extensive experimental study of the new heuristics and of several previously proposed algorithms.

Currently, a viable solution to provide multicast provision over a multiprotocol label switch with traffic engineering (MPLS-TE) domain is unavailable because of the missing link able to couple multicast traffic distribution with an MPLS-TE enabled network. This is due to the limited or less research investigation that was done in this area. Most of the investigation methods tackle the problem individually such as deploying internet protocol (IP) multicast in a plain network or MPLS domain but without considering a combination of both technologies that is aware of differentiated services requirements. This thesis presents an alternative solution for the multicast differentiated services provision problem over a MPLS-TE enabled network. The approach is exposed and analyzed through a practical solution that was developed within a network simulation environment. The research presented in this thesis orchestrates the already available technologies offered by the multicast protocols suite and differentiated services (DiffServ) aware MPLS-TE that allows applying separately the constraint-based routing and admission control to different classes of services. The novelty and solution of this thesis relies on using MPLS constraint-based routing concepts (e.g.traffic trunks) in order to solve TE issues revealed during multicast traffic distribution.

Existing QoS-aware Web Services discovery architectures tend to focus solely on fulfilling the requirements of either the client or the provider. However, the interests of the provider and client are not equivalent. The provider’s goal is to maximize the profit and consume the least amount of resources. On the other hand, the client’s selection is determined by their own requirements which do not always reflect the real resource overheads. This research aims to provide a novel mobile Web Services discovery and selection method based on utility functions to balance the requirements for clients and providers. In the mobile environment, it is critical to conserve resource consumption in addition to fulfilling user requirements, as resources such as wireless network bandwidth and mobile device power are precious. The proposed service selection strategy enables service providers to balance the cost/performance ratios and utilize the network bandwidth more effectively, while the clients can still attain the functional and quality levels specified in the service request.

Existing QoS-aware Web Services discovery architectures tend to focus solely on fulfilling the requirements of either the client or the provider. However, the interests of the provider and client are not equivalent. The provider’s goal is to maximize the profit and consume the least amount of resources. On the other hand, the client’s selection is determined by their own requirements which do not always reflect the real resource overheads. This research aims to provide a novel mobile Web Services discovery and selection method based on utility functions to balance the requirements for clients and providers. In the mobile environment, it is critical to conserve resource consumption in addition to fulfilling user requirements, as resources such as wireless network bandwidth and mobile device power are precious. The proposed service selection strategy enables service providers to balance the cost/performance ratios and utilize the network bandwidth more effectively, while the clients can still attain the functional and quality levels specified in the service request.