Search

Cloud computing has known an impressive success over the last few years and is continuously emerging personal and profesional life thanks to its elasticity, pricing model, etc. This innovative technology has attracted both industrial and research communities and becomes omnipresent.In order to take benefit from the Cloud expansion, providers require effecient management strategies to properly supply their physical capabilities such as network resources. Besides, Cloud providers have to respect the Service Level Agreements (SLA) and avoid any outage impact in order to guarantee the Cloud-based service continuity. In this thesis we tackle the problem of reliable virtual network embedding within the Cloud backbone by considering the impact of physical equipments' outages. Our main focus is to improve the provider's turnover by i) maximizing the acceptance rate of the incoming virtual networks issued from clients' requestand ii) minimizing the penalties induced by service disruption due to the physical failures. This optimization problem is NP-hard with multi-objective and non-linear formalization. To cope with this complexity and since reaching the optimal solution is computationally intractable, we propose different strategies that aim to respect the aformentionned objectives. First, we propose a preventive approach named PR-VNE that urges the use of reliable network resources in order to avoid the physical failures impact.PR-VNE strongly relies on the Artificial Bee Colony metaheuristic to reach an optimized solution. It should be highlighted that PR-VNE does not adopt a recovering mechanism to deal with the network outages. Second, we devise a new reactive approach named CG-VNE that does not consider any backup resources but re-embed the impacted virtual resources once an outage occurs in the underlying network. As well as this reactive mechanism, CG-VNE adopts the same preventive strategy like PR-VNE by avoiding the unreliable resources. It should be noted that CG-VNE is devised basing on a Game Theory framework by defining a collaborative mapping game. Finally, we deal with the survivable batch mapping problem that considers the embedding of a virtual networks set instead of one client request. We introduce a new reliable batch embedding strategy named BR-VNE that relies on Monte-Carlo Tree Search algorithm.BR-VNE delegates the embedding of one virtual network request to any online algorithm and focuses to find the best mapping sequence order. The performance evaluation of our algorithms leads to efficient results., Cloud computing a connu un succès impressionnant au cours des dernières années. Cette expansion en permanence se manifeste clairement non seulement par l'omniprésence du Cloud dans les environnements académiques et industriels mais aussi par la hausse impressionnante du chiffre d'affaire des services Cloud. Ce succès important réalisé en peu de temps est achevé grâce à la caractéristique d'élasticité du Cloud. Afin de bien tirer profit de cette nouvelle architecture software et hardware, les fournisseurs de service Cloud requièrent des stratégies adéquates pour une gestion efficace des équipements physiques continuellement allouées à plusieurs clients simultanément. En effet, le fournisseur Cloud doit respecter le Service Level Agreement (SLA) et assurer la continuité de service afin de minimiser le coût des pénalités. Dans cette thèse, nous abordons la problématique de l'instanciation (i.e., mapping) des réseaux virtuels au sein du réseau coeur du Cloud tout en considérant la fiabilité des équipements physiques (i.e., routeurs et liens).Notre objectif principal est de maximiser le chiffre d'affaires du fournisseur Cloud par le biais de i) la maximisation du taux des réseaux virtuels acceptés dans le réseau coeur du Cloud etii) la minimisation des pénalités induites par les interruptions de service en raison de pannes des équipements du réseau. Il a été démontré que ce type de problème est NP-dur donc aucune solution optimale ne peut être calculée à large échelle. Premièrement, on a proposé un algorithme nommé PR-VNE qui favorise l'utilisation des équipements les plus fiables, lors de l'instanciation des réseaux virtuels, afin de minimiser l'impact néfaste des pannes. Cette proposition exploite les avantages de la métaheuristique de colonie d'abeilles pour assurer une qualité optimisée de mapping de la solution en termes d'optimalité. Il est à souligner que PR-VNE ne réserve aucune ressource de secours dans le cas où panne se présente. Deuxièmement, on a défini un algorithme qui adopte non seulement la stratégie préventive mais aussi un mécanisme curatif pour remédier aux pannes imprévisibles. Cet algorithme nommé CG-VNE est conçu par une modélisation basée sur la théorie des jeux. CG-VNE reconfigure (re-instancie) les ressources virtuelles impactées par les pannes dans d'autres équipements réseau sans pour autant réserver des ressources de secours. Finalement, on a considéré la vision macroscopique en définissant un algorithme qui traite un lot de requêtes tout en considérant la fiabilité. Cet algorithme nommé BR-VNE se base principalement sur la stratégie Monte-Carlo Tree Search pour trouver la meilleure séquence de mapping. L'évaluation des performances de nos propositions prouve leurs efficacités par rapport aux méthodes concurrentes de la littérature en terme de : i) taux de réseaux virtuels acceptés, ii) taux de requêtes impactées par les pannes et iii) revenu du fournisseur de service Cloud.

International audience

International audience; Vehicular ad hoc networks (VANETs) are witnessing in recent years a rapid development for road transmissions and are considered as one of the most important types of next generation networks, in which drivers can have access anywhere and anytime to information. However, vehicles have to deal with many challenges such as the links failures due to their frequent mobility as well as limited degrees of freedom in their mobility patterns. In this paper, we propose a new quality of service multicast and multipath routing protocol for VANETs, based on the paradigm of bee's communication, called multicast quality of service swarm bee routing for VANETs (MQBV). The MQBV finds and maintains robust routes between the source node and all multicast group members. Therefore, the average end-to-end delay and the normalized overhead load should be reduced, while at the same time increasing the average bandwidth and the packet delivery ratio. Extensive simulation results were obtained using ns-2 simulator in a realistic VANET settings and demonstrated the efficiency of the proposed protocol.

International audience; Sensor placement is a fundamental issue in wireless sensor networks (WSNs). The sensor–positions can be predetermined to guarantee the quality of surveillance provided by the WSN. However, in remote or hostile sensor field, randomised sensor placement often becomes the only option. In this paper, we survey existing random node placement strategies. We categorise random placement strategies into simple and compound. An empirical study has been carried out yielding a detailed analysis of random deployment intrinsic properties, such as coverage, connectivity, fault–tolerance, and network lifespan.We also investigate the performance of a hybridisation of the simple diffusion model that places a large number of nodes around the sink and the constant diffusion that provides high coverage and connectivity rates. We show that such hybridisation ensures better performance. The obtained results give helpful design guidelines in using random deployment strategies.

International audience; Increasing interests in Vehicular Ad hoc NETworks (VANETs) over the last decade have led to huge investments in technologies and research to improve road safety by providing timely and accurate information to drivers and authorities. To achieve the timely dissemination of messages, various routing protocols for VANETs have been recently proposed. We present a Hybrid Bee swarm Routing (HyBR) protocol for VANETs. HyBR is based on the continuous learning paradigm in order to take into account the dynamic environmental changes in real-time which constitute a key property of VANETs. The protocol combines the features of topology routing with those of geographic routing. HyBR is a unicast and a multipath routing protocol (aimed at both urban and rural scenarios) which guarantees road safety services by transmitting packets with minimum delays and high packet delivery. To demonstrate the effectiveness and the performance of HyBR, we conducted a performance evaluation based on several metrics such as end-to-end delay, packet delivery ratio, and normalized overhead load. We obtained better performance results with HyBR in contrast to results obtained from traditional routing algorithms such as Ad hoc On-Demand Distance Vector (AODV) topology-based routing protocol and Greedy Perimeter Stateless Routing (GPSR) geography-based protocol.

International audience

International audience; One of the fundamental services provided by a wireless sensor network (WSN) is the monitoring of a specified region of interest (RoI). Considering the fact that emergence of holes in the RoI is unavoidable due to the inner nature of WSNs, random deployment, environmental factors, and external attacks, assuring that the RoI is completely and continuously covered is very important. This paper seeks to address the problem of hole detection and healing in mobile WSNs. We discuss the main drawbacks of existing solutions and we identify four key elements that are critical for ensuring effective coverage in mobile WSNs: 1) determining the boundary of the RoI, 2) detecting coverage holes and estimating their characteristics, 3) determining the best target locations to relocate mobile nodes to repair holes, and 4) dispatching mobile nodes to the target locations while minimizing the moving and messaging cost. We propose a lightweight and comprehensive solution, called holes detection and healing (HEAL), that addresses all of the aforementioned aspects. The computation complexity of HEAL is O(v2) , where v is the average number of 1-hop neighbors. HEAL is a distributed and localized algorithm that operates in two distinct phases. The first identifies the boundary nodes and discovers holes using a lightweight localized protocol over the Gabriel graph of the network. The second treats the hole healing, with novel concept, hole healing area. We propose a distributed virtual forces-based local healing approach where only the nodes located at an appropriate distance from the hole will be involved in the healing process. Through extensive simulations we show that HEAL deals with holes of various forms and sizes, and provides a cost-effective and an accurate solution for hole detection and healing.

International audience; A vehicular ad hoc network (VANET) is a subclass of mobile ad hoc networks, considered as one of the most important approach of intelligent transportation systems (ITS). It allows inter-vehicle communication in which their movement is restricted by a VANET mobility model and supported by some roadside base stations as fixed infrastructures. Multicasting provides different traffic information to a limited number of vehicle drivers by a parallel transmission. However, it represents a very important challenge in the application of vehicular ad hoc networks especially, in the case of the network scalability. In the applications of this sensitive field, it is very essential to transmit correct data anywhere and at any time. Consequently, the VANET routing protocols should be adapted appropriately and meet effectively the quality of service (QoS) requirements in an optimized multicast routing. In this paper, we propose a novel bee colony optimization algorithm called bees life algorithm (BLA) applied to solve the quality of service multicast routing problem (QoS-MRP) for vehicular ad hoc networks as NP-Complete problem with multiple constraints. It is considered as swarm-based algorithm which imitates closely the life of the colony. It follows the two important behaviors in the nature of bees which are the reproduction and the food foraging. BLA is applied to solve QoS-MRP with four objectives which are cost, delay, jitter, and bandwidth. It is also submitted to three constraints which are maximum allowed delay, maximum allowed jitter and minimum requested bandwidth. In order to evaluate the performance and the effectiveness of this realized proposal using C++ and integrated at the routing protocol level, a simulation study has been performed using the network simulator (NS2) based on a mobility model of VANET. The comparisons of the experimental results show that the proposed algorithm outperformed in an efficient way genetic algorithm (GA), bees algorithm (BA) and marriage in honey bees optimization (MBO) algorithm as state-of-the-art conventional metaheuristics applied to QoS-MRP problem with the same simulation parameters.

Today, maintaining the best quality of experience (QoE) as the subjective perception of end-users is one of the most important requirements of both wireless and wired networks. To achieve this goal, we frequently use network functions such as admission control, resource management, routing and traffic control. We focus on the design and evaluation of a QoE-aware routing mechanism. We present an end-to-end model where Smartphone users connect to content servers through a wired network across a wireless access network. We propose an adaptive and evolutionary system with a QoE-aware routing protocol, namely QoE Q-learning-based adaptive routing. Our experimental results demonstrate significant performance improvements in mean opinion score for various network loads, convergence time during the initialisation phase and control overheads over other traditional approaches.Copyright © 2013 John Wiley & Sons, Ltd.

International audience; The micro grid is a small intelligent electricity network (power grid) that uses computer technology to optimize production, distribution, and consumption and can operate independently by managing "intelligently" its own expenses and production capacity. The main objective is to ensure the best user perception without any human monitoring based on effective needs and better use. To do so, we need a control mechanism to drive the microgrid. In this article, we propose a new easily deployable framework called SMILAY based on two key ideas: a hierarchical architecture and a service-specific overlay network (SSON). SMILAY works by deploying SMIBOXES over all the buildings of the microgrid and interconnecting them through an overlay network. A multicriteria selection mechanism is used to design a small set of SMIBOXES in charge of the information flow management. Our simulation results show that the proposed approach provides fast convergence time performance of the information flow compared to other related approaches using minimum network capacity.

International audience; The major problem with Empirical Mode Decomposition (EMD) algorithm is its lack of a theoretical framework. So, it is difficult to characterize and evaluate this approach. In this paper, we propose in the two dimensional case, the use of an alternative implementation to the algorithmic definition of the so-called sifting process used in the original Huangs Empirical Mode Decomposition method. This approach, especially based on Partial Differential Equations (PDE) was presented by O. Niang and al. in previous works, in 2005 and 2007 and lays on a nonlinear diffusion-based filtering process to solve the meanenvelope estimation problem. In 1D case, the efficiency of the PDE-based method, compared to the original EMD algorithmic version, is also illustrated in recent paper [1]. Recently, several bidimensional extensions for EMD method were proposed. Despite some efforts, 2D versions for EMD appear poorly performing and are very time consuming. So in this work, an extension to 2D space of the PDE-based approach is extensively described. This approach has been applied in case of both signal and image decomposition. Obtained results confirm the usefulness of the new PDE-based sifting process for decomposition of various kinds of data. Some results have been provided in the case of image decomposition. The effectiveness of the approach encourages its usage in a number of signal and image applications such as denoising, detrending, or texture analysis.

Despite the disparity in the objectives of sensor applications, the main task of sensor networks is to sense and collect data from a target domain, process the collected data, and transmit the information back to the base station. Achieving this task efficiently requires the development of an energy-efficient routing protocol to set up paths between sensor nodes and the base station. The path selection must be such that the lifetime of the network is maximized. The definition of this metric is determined by the type of the provided service. Using several definitions of lifetime and a common evaluation framework, the lifetime of some representative flat and hierarchical sensor network routing protocols has been analyzed, namely: DIRECT, FLOODING, GOSSIPING, LEACH, and HEED. Extensive simulations have been carried out yielding a detailed analysis of the spatial-temporal distribution of lifetimes. Through this study, a new technique aimed at providing a good spatial-temporal distribution of lifetime is proposed. The resulting EHEED protocol is compared to others. Experimental results show that EHEED can extend remarkably the network lifetime and can be very effective for long-lived sensor network.

We propose in this letter a bio-inspired Quality of Service (QoS) routing algorithm that is based on the trial/error paradigm combined with a continuous adaptive function to optimize three QoS criteria: static cumulative cost path and dynamic end-to-end delay and residual bandwidth. Our proposed approach uses a model that combines a stochastic planned pre-navigation for the exploration phase and a deterministic approach for the backward phase. We adopt a unified framework of online learning to develop a cost function. We evaluated the performance of our QoS-routing algorithm and the simulation results demonstrate substantial performance improvements for networks with dynamically changing traffic.

International audience; In this article, we introduce a Quality of Service (QoS) routing algorithm based on dynamic state-dependent policies. The proposed algorithm uses a bio-inspired approach based on trial/error paradigm to optimize two QoS different criteria: Energy and end-to-end delay. Our proposal, called EDEAR “Energy and Delay Efficient Adaptive Routing”, uses a paradigm based on routes’ exploration. In this phase, we collect information in terms of energy and delay by using continuous learning parameters on the network and update routing table maintained at each node of the network. This phase of exploration has been optimized by proposing a new algorithm based on multipoint relay for energy consumption, thus reducing the overhead generated by the packets exploration. Numerical results obtained with NS simulator for different levels of traffic’s load and mobility show that EDEAR gives better performances compared to traditional approaches.

International audience; We propose a novel approach to treat the problem of void avoidance in sensor networks. In our approach, the choice of the forwarding region around a discovered void is oriented by the locations of (i) the sender node, (ii) the void center and (iii) the destination node. Three void-avoidance schemes are described which use the same void-discovery algorithm. Based on an oriented face routing, the first scheme is used by the boundary nodes to prevent packets from dropping. The second scheme is used by sender nodes located n-hops away from the void boundary to launch a preventive rerouting. The third scheme tends to protect packets from both the concave zones and the boundaries nodes of voids. Proposed schemes are simple to implement. They save network resources and could be easily associated with existing geographic routing protocols. Simulation results show the efficiency of the proposed approach.

International audience; Reactive routing protocols for mobile ad hoc networks reduce the routing cost in high mobility environments where link failures are frequent.However, route discovery in these protocols is typically performed via network-wide flooding, which consumes a Substantial amount of bandwidthand causes a significant latency to data packets. To improve the dynamic source routing (DSR) protocol and overcome these limitations, we proposetwo optimization techniques, viz. generalized salvaging mechanism and cache maintenance using a distributed topology discovery mechanismthrough mobile ant-agents. We show, by simulations, that our contributions improve the DSR performance, and particularly in large scale networks with high mobility and heavy load that cause frequent link failures.

Today, the Internet has become the most important communication infrastructure of our society. It enables worldwide users (individual, groups, and organizations) to access and exchange remote information scattered all over the world. Currently, due to the growing needs in telecommunications (Video on Demand, Video-Conference, VoIP, etc.) and the diversity of transported flows, the Internet still does not meet the requirements of integrated service networks that carry multimedia data traffic with a high quality of service (QoS). The main drivers of this evolution are the continuous growth of the bandwidth requests, the promise of cost improvements, and finally the possibility of increasing profits by offering new services. First, the Internet does not support resource reservation which is primordial to guarantee an end-to-end QoS (bounded delay, bounded delay jitter, and/or bounded loss ratio). Second, flow packets may be subjected to unpredictable delays and thus may arrive at their destination after the expiration time, which is undesirable for continuous real-time media. In this context, to optimize the financial investment of their networks, operators must use the same underlying Internet infrastructure to transport all types of flows including those associated with strict QoS requirements. Therefore, we need to develop high quality control mechanisms that can support efficient end-to-end QoS requirements. It is clear that the integration and support of these QoS parameters increases the complexity of current QoS routing algorithms and models. Many QoS challenges still exist in emerging hybrid networks often consisting of a highly heterogeneous mix of several different types of networks (wireless, broadcast, mobile, fixed, etc.) especially during the routing process which is central to the delivery of optimal performances in hybrid network environments. Providing a good QoS in heterogeneous networks for irregular traffic flows remains a significant challenge. A general problem of large-scale, distributed systems is the everincreasing complexity of their operations. This complexity is mainly driven by heterogeneity. The wide variety of underlying technologies deployed within a network and their different, if not proprietary, operational paradigms makes it hard for network operators to solve. Moreover, network operations are typically handled by one or more human operators. Manual control is time-consuming, expensive, and error-prone. Nevertheless, both technologies and needs continue to develop and grow. The risk is thus that complexity and cost become limiting factors in the evolution of networks in the future and for the enriched services these networks are expected to deliver. We have witnessed significant interests in recent years in the area of QoS routing algorithms that can support Ann. Telecommun. (2008) 63:543–544 DOI 10.1007/s12243-008-0066-x

International audience; In the context of modern high-speed internet network, routing is often complicated by the notion of guaranteed Quality of Service (QoS), which can either be related to time, packet loss or bandwidth requirements: constraints related to various types of QoS make some routing unacceptable. Due to emerging real-time and multimedia applications, efficient routing of information packets in dynamically changing communication network requires that as the load levels, traffic patterns and topology of the network change, the routing policy also adapts. We focused in this paper on QoS based routing by developing a neuro-dynamic programming to construct dynamic state-dependent routing policies. We propose an approach based on adaptive algorithm for packet routing using reinforcement learning called N best optimal path Q-routing algorithm (NOQRA) which optimizes two criteria: cumulative cost path (or hop count if each link cost=1) and end-to-end delay. A load balancing policy depending on a dynamical traffic path probability distribution function is also defined and embodied in NOQRA to characterize the distribution of the traffic over the N best paths. Numerical results obtained with OPNET simulator for different packet interarrival times statistical distributions with different levels of traffic’s load show that NOQRA gives better results compared to standard optimal path routing algorithms.

International audience