Over-the-air federated learning (OTA-FL) is a communication-efficient paradigm that leverages the superposition property of wireless channels to aggregate client updates simultaneously, significantly reducing uplink latency and bandwidth usage. While OTA-FL offers advantages in scalability and speed, it poses challenges in energy efficiency and delay management. This paper proposes a multi-attribute client selection framework that addresses these challenges through a multi-objective optimization approach. We analytically model selection attributes: energy efficiency, communication delay, loss, and fairness, and formulate three optimization problems to capture different trade-offs. To solve them, we employ the Multi-Objective Grey Wolf Optimizer (MOGWO), a nature-inspired metaheuristic algorithm that effectively balances exploration and exploitation. Experiments on MNIST, Fashion MNIST, and CIFAR-10 demonstrate that our approach outperforms baseline and loss-aware methods, achieving up to 13% energy savings while improving model accuracy, fairness, and reliability.

Programmable data planes facilitate near line-rate network traffic classification by maintaining per-flow statistics directly within the data plane memory. However, this capability introduces a significant vulnerability: spoofing-based attacks can inject substantial volumes of dummy flows, swiftly depleting limited memory resources and overwriting legitimate flow records. Consequently, this degrades the classification performance and disrupts the reliable traffic analysis. In this paper, we address the challenge of legitimate flow overwrites caused by spoofed flow explosions and propose a resilient in-network detection framework that safeguards data-plane resources without compromising inference accuracy. Our approach employs a data-driven method to identify spoofing behavior based on traffic characteristics and dynamically filters spoof-like sources using early-stage flow differentiation parameters. The control plane learns threshold values for these parameters from observed data and deploys them in the data plane to validate suspicious flows before allocating memory, thereby limiting unnecessary register usage by attack-like traffic. We prototype our solution in a P4-enabled software switch and demonstrate its effectiveness in mitigating memory exhaustion attacks, preserving classification reliability, and maintaining in-network inference performance under high-volume spoofing attacks.

Ensuring high Quality of Experience (QoE) is crucial for seamless wireless access in immersive metaverse environments, yet this goal faces challenges due to constrained communication and computing resources. Coordinating x-applications (xAPPs) in open radio access network (O-RAN) effectively is essential for optimizing these limited resources to meet high QoE demands, thus enhancing the user experience. Federated reinforcement learning (FRL) further supports this by facilitating collaboration among distributed agents, boosting training efficiency. This paper proposes a QoE utility model based on xAPPs design for the metaverse, which incorporates energy efficiency, video quality, and communication and computation delays to optimize power consumption and bandwidth while minimizing delay. Tailored for FRL-based resource allocation within a sliced O-RAN architecture for the metaverse, the QoE utility model manages key network functions-such as power control and physical resource block allocation-via multi-agent xAPPs, while improving training efficiency. Numerical results demonstrate that the proposed algorithm significantly enhances QoE and reduces delay compared to benchmark algorithms.

The rapid growth of the Internet of Things (IoT) has triggered the need for scalable and energy-efficient communication solutions. While LoRaWAN is widely used for long-range wireless access, its Aloha-based MAC protocol struggles with high collision rates in dense networks. Existing solutions such as irregular repetition slotted ALOHA (IRSA) and contention resolution diversity slotted ALOHA (CRDSA) have improved network performance by using packet repetitions and successive interference cancellation. However, they do not fully leverage the unique properties of LoRaWAN Spreading Factors (SFs). To address this gap, we propose a new approach called SF-IRSA, where IoT devices transmit replicas using different SFs, enabling the decoder to apply an SF-IRSA-SIC process that leverages both temporal and spatial dimensions for efficient packet decoding. Our theoretical analysis and simulations show that SF-IRSA outperforms IRSA and CRDSA in terms of throughput and reliability. Specifically, using up to two SFs results in a 16.2% increase in the asymptotic throughput compared to standard IRSA. When extending to three SFs, the throughput gain reaches 116.9%, with a maximum of $\mathbf{2 2 4. 5 2 \%}$ while using $\mathbf{6}$ SFs.

Federated learning (FL) faces significant challenges due to the non-independent and identically distributed (non-IID) data and the heterogeneous nature of clients’ characteristics. Clustered federated learning (CFL) addresses these issues by grouping similar clients and creating cluster-specific models. However, CFL introduces additional challenges, such as determining the optimal clustering criteria and managing the dynamic nature of client availability and data distribution. This paper proposes a novel CFL approach that integrates a full spectrum of relevant factors where clients are clustered based on data distribution, device type, and geographical location. Each group selects a subset of clients based on the information’s age, the client’s motivation, and the availability of resources to participate in the learning process. Unlike previous approaches that focus on a limited set of criteria, our method considers a holistic view of client attributes to improve clustering performance. The experimental results demonstrate the efficiency and effectiveness of the proposed method, highlighting significant improvements in communication efficiency and model quality. Furthermore, our approach adapts dynamically to changes in client availability, ensuring robust learning over time. By optimizing client selection and leveraging cluster-specific characteristics, the proposed approach enhances the scalability, robustness, and overall performance of FL systems.

Federated Learning (FL) presents a promising paradigm for decentralized model training; however, its real-world adoption is hindered by several critical challenges, including non-independent and identically distributed (non-IID) data across clients, heterogeneous computational capabilities, and significant communication overhead. To address these issues, this paper introduces a novel multi-attribute client clustering and selection framework for FL. The proposed approach groups clients according to data distribution, device capabilities, geographic location, and model update behavior. Within each cluster, an adaptive client selection mechanism leverages dynamic attributes such as residual energy, data freshness, and client participation motivation to identify the most suitable participants. Experimental evaluations on standard FL benchmark datasets demonstrate that the proposed framework achieves faster convergence, higher global model accuracy, and improved energy efficiency compared to state-of-the-art approaches.

Innovative holographic applications such as holographic concerts have recently emerged. They are expected to play an important role in the Metaverse. Hybrid Network Function Virtualization (NFV) / IN-Network Computing (INC) network infrastructures are needed to provision them as recently shown in the literature. INC is an emerging technology that aims to distribute the computational workload across the network by placing computational tasks on programmable devices (e.g., routers or switches). However, the integration of INC in existing infrastructures does face significant management and orchestration challenges. Although the ETSI Management and Orchestration (MANO) architectural framework designed for 5G facilitates application provisioning in networks that are NFV enabled, it lacks support for networks that are INC enabled. Therefore it is necessary to have a new MANO architecture in order to provision applications which have both NFV and INC components. This paper proposes a hybrid NFV-INC MANO architecture for provisioning holographic applications in hybrid NFV/INC environment. The proposed architecture is an extension of the ETSI NFV MANO. It will certainlyplay an important role in 6G since many applications foreseen for 6G will have the same stringent requirements as holographic applications. It is evaluated through a proof of concept prototype. The following tools were used for the prototype: Open Source MANO (OSM) and Mininet emulator.

The integration of terrestrial and non-terrestrial networks (TN-NTN) with open radio access network (O-RAN) technology presents a significant advancement for facilitating scalable and immersive Metaverse services within 6G networks. Seamless virtual experiences necessitate highly reliable, low-latency communication, effective resource management, and adaptive decision-making to satisfy the varied and rigorous requirements of Metaverse applications, including gaming, healthcare, and autonomous systems. The inherent heterogeneity, dynamic nature, and substantial resource requirements of TN-NTN present significant challenges for effective resource allocation and optimizing quality of experience (QoE). Then, we formulate a multi-objective optimization problem for joint resource allocation and spectrum sharing in O-RAN-enabled TN-NTN Metaverse environments. This problem is inherently NP-hard due to the intricate coupling between continuous action spaces and discrete decision variables. Solving such a complex problem using traditional optimization approaches is complex. To overcome this, we transform the problem into a decentralized partially observable Markov decision process (Dec-POMDP) and address it using a hierarchical multi-agent deep reinforcement learning (MADRL) approach. This study presents a hierarchical multi-agent proximal policy optimization (MAPPO) framework, a new MADRL solution for dynamic resource allocation and spectrum sharing in O-RAN-enabled TN-NTN Metaverse environments. MAPPO facilitates collaborative learning among intelligent agents to optimize resource management strategies in a decentralized manner, considering essential metrics, including energy consumption, latency, and meta-distance. The proposed framework enhances resource utilization efficiency, minimizes latency, and improves the QoE for Metaverse users through the seamless allocation and management of resources. Comprehensive simulations show that MAPPO outperforms baseline methods, such as conventional reinforcement learning and centralized optimization approaches, achieving better energy efficiency, lower latency, and improved QoE. This demonstrates its effectiveness in adapting to dynamic 6G-enabled Metaverse requirements, enabling intelligent and scalable TN-NTN networks.

The increasing demand for ultra-low latency and high bandwidth in emerging applications, such as virtual reality gaming and telesurgery, is challenging current network infrastructures. In-Network Computing (INC) has emerged as a promising solution to these challenges by optimizing network performance, reducing congestion, and minimizing both latency and bandwidth usage. As 6G networks strive to deliver unprecedented connectivity and ultra-low latency, INC is poised to become an integral part of future network architectures. However, a significant gap exists in the literature concerning a comprehensive model for describing INC components, which is crucial for effective INC provisioning. To address this gap, we propose the In-Network Computing Ontology (INCO), a domain-independent, ontology-based model designed to describe and discover INC components in a centralized repository. Our model covers both the functional and non-functional specifications of INC components. Furthermore, we introduce a semantic matchmaking algorithm that uses the INCO model to automatically discover and select the most relevant INC components from the repository based on user requests. Experimental simulations validate our approach, demonstrating the effectiveness of the semantic matchmaking algorithm, particularly regarding response time and consistency.

Amidst the emergence of immersive applications, such as, the metaverse, Virtual Reality (VR), Augmented Reality (AR), and Holography, it is clear that substantial enhancements to our existing internet infrastructure are imperative. Fulfilling the stringent Quality of Service (QoS) requirements—which include ultra-low latency, high bandwidth, and optimal frame refresh rates—hinges on the seamless integration of Communication, Caching, and Computing, collectively referred to as the "3C". These elements must be interwoven within the network fabric. Our study presents a novel approach to optimize these 3C resources within a network framework that incorporates Edge and Cloud computing, and In-Network Computing (INC). We propose a resource allocation solution tailored to networks enabled by INC. We aim to efficiently manage the distribution of Service Function Chains and the storage of relevant data for immersive applications. Given the inherent complexity of the tackled problem, we propose two solutions: a Particle Swarm Optimization (PSO)-based meta-heuristic and a simpler, yet effective, greedy heuristic. Our comprehensive simulations, grounded in realistic VR scenarios, validate the effectiveness of the proposed solution, which not only enhances resource efficiency and reduces operational costs but also guarantees high refresh rates and maintains a Motion-To-Photon latency under 22 ms.

Emerging technologies that encompass a multitude of tiny wearable devices are vulnerable to cyberattacks that can turn them into bots for launching Distributed Denial of Service (DDoS) attacks. In-network Machine Learning (ML) has emerged as a prominent solution for detecting and responding to such attacks in the shortest possible time to avoid disrupting user experience. However, the dynamic nature of attack traffic patterns necessitates continuous adaptation of conventional one-size-fits-all ML models. The manual process of identifying novel malicious traffic patterns and updating the ML model from the control plane to the network data plane is time-consuming and labor-intensive. This study aims to automate the identification of unseen malicious traffic patterns and update the ML model in programmable networks using a data-driven approach. Specifically, we determine drift detection thresholds from the baseline performance of historical (i.e., training) data and consider any deviation as anomalies in unseen (i.e., testing) data. These thresholds are continuously updated by considering changes in the data distribution and in-network inference results. We utilize an intrusion detection dataset (CIC-IDS2017) to illustrate the impact of emerging attacks on model performance degradation and the efficacy of our proposed data-driven method in mitigating these attacks. Our approach has proven effective in safeguarding against evolving DDoS attacks.

This paper proposes two unmanned aerial vehicles (UAV) trajectory planning solutions taking into consideration mission deadline, energy consumption, and communication constraints. The problem is mathematically formulated as a mult-iobjective optimization problem. The NP-hardness complexity is demonstrated then two heuristic solutions are proposed. Specifically, we first present a near-optimal UAV trajectory planning algorithm that reduces the number of stop/hovering points. Then, we propose an artificial bee colony-based algorithm with enhanced candidate selection. Extensive simulation results show that both our schemes outperform the well-known Successive Convex Approximation (SCA) technique, under low-moderate traffic demand. They perform as well as SCA under high demand.

Cloud and edge computing have proposed different functionalities to enable multiple applications requiring different communication, computing, and caching (3C) resources. The upcoming futuristic applications (e.g., metaverse, holographic, and haptic communication) impose further stringent requirements (e.g., ultra-low latency, ultra-high reliability) on the infrastructure. These requirements call for a paradigm shift in the infrastructure architecture where all resource components and owners collaborate from the cloud up to the edge, creating a cloud-to-edge continuum of integrated resources. Furthermore, we argue that artificial intelligence (AI) and collaborative-based decisions are promising techniques to efficiently manage the highly complex architecture that jointly leverages 3C in the continuum. This article presents a comprehensive survey of existing research, including AI and collaborative-based studies, targeting the effective and seamless provision of 3C resources and services in the cloud-to-edge continuum. Through an extensive analysis of driving use cases, the synergy between these three main services is scrutinized to highlight its crucial role in the next-generation network infrastructures (NGNI). Finally, a discussion on the opportunities and challenges brought by integrating 3C in NGNI from different perspectives, including architectural design as well as the regulatory and business aspects, are presented.

Grant-free access, in which each Internet-of-Things (IoT) device delivers its packets through a randomly selected resource without spending time on handshaking procedures, is a promising solution for supporting the massive connectivity required for IoT systems. In this paper, we explore grant-free access with multi-packet reception capabilities, with an emphasis on ultra-low-end IoT applications with small data sizes, sporadic activity, and energy usage constraints. We propose a power allocation scheme aimed at maximizing throughput while minimizing power consumption by considering the traffic and energy constraints of IoT devices. Our approach employs a stochastic geometry framework and mean-field game theory to model and analyze the mutual interference among active IoT devices. Additionally, we utilize a Markov chain model to capture and track the queue length of IoT devices, enabling the derivation of the transmission success probability at steady-state. The simulation results illustrate the optimal power allocation strategy and evaluate the proposed approach’s performance in terms of packet transmission success probability and average delay.

The emergence of a new class of enhanced multimedia services has pushed network operators to support innovative network services while meeting end-to-end Quality of Service requirements and maintaining profitability. Recent works have hailed Network Function Virtualization (NFV) as a cost-effective enabling technology for novel service delivery in 5G networks. Over time, the association between these NFV-based services and multi-domain networks has grown. As the market competition posed by third-party operators such as virtual operators and service providers has intensified, profitability has become a crucial factor in resource allocation issues. In this paper, we formulate the problem of multi-domain network slicing in dynamic market environments and investigate the effects of these variables on service placement. Due to the NP-hardness of the problem, we employ a node ranking-based algorithm to determine optimal slice entry points in order to maximize profits while meeting end-user QoS requirements. In terms of slice placement acceptance rate and profit growth, numerical results demonstrate the superior performance of our proposed solutions.

The Internet of Things has become highly popular in the healthcare sector due to its benefits for patients, doctors and health authorities. In particular, the resulting sub-fields, such as medical IoT and mobile health (mHealth), have become essential in pervasive healthcare monitoring and preventing the spread of viruses. However, the massive amount of data generated by millions of IoT devices challenges the current infrastructure of cloud and edge computing, leading to high response times and unreliable results. Security and privacy concerns make these technologies particularly vulnerable to attacks. In this paper, we propose an mHealth solution based on in-network computing paradigm to provide privacy-aware, time-constrained and reliable results for IoT applications in a mobile health context. Our simulation and analytical results show that our solution satisfies mHealth applications requirements and outperforms conventional solution based on edge computing.

In-network traffic classification is a new paradigm in developing accurate and early-stage traffic classification solutions. However, despite having good accuracy, the one-fit machine learning model becomes outdated as the traffic pattern changes over time. This changing traffic pattern leads to misclassification, i.e., incorrect mapping of traffic flows to the Quality of Service (QoS) classes, resulting in a service quality violation and the imposition of a penalty. This paper proposes a profit-aware adaptive traffic classification approach in the data plane. We particularly design an economic model to measure the impact of per-class misclassification rate on the infrastructure provider's profit and use an adaptive method to handle misclassification directly inside a programmable data plane. The evaluation result shows that optimal path allocation for various traffic classes determines the targeted revenue, while improving classifier accuracy reduces penalty and maintains the maximum profit.

The fifth-generation (5G) mobile standard has been designed to support new use cases such as ultra-reliable and low-latency communication (URLLC). The future 6G is envisioned to support extreme URLLC with higher QoS requirements (e.g., remote surgery, autonomous driving, etc.). URLLC applications need higher QoS that require more power allocations. Consequently, QoS variance will increases, which is intolerable for URLLC. An important amount of energy can be saved through a power control scheme. In this work, we are interested in energy-aware self-organizing networks that provide satisfactory performance for URLLC. We propose a predictive QoS paradigm to enhance satisfaction and reduce power consumption under URLLC’s constraints. A predictive QoS is an intelligent paradigm that allows Mobile/IoT-device to adjust power allocation to the minimum required to achieve the target QoS. First, we model power control as a satisfactory game, where IoT-devices aim to meet their target demands instead of maximizing them. Next, we introduce a distributed satisfactory learning scheme, called Robust Banach-Picard (RBP), to allow devices to self-adjust their power allocation to maintain reliability and latency within the tolerated range of the URLLC application. The algorithm implements deep learning and a derivative concept of federated learning to account for channel variability in power control. Extensive simulations exhibit the advantages and drawbacks of the proposed scheme for URLLC applications. Results show that RBP can maintain instantaneous reliability and latency within the tolerated request at the minimum energy costs. Consequently, RBP can be safe to use for URLLC use cases compared to conventional Banach-Picard iterates.

In natural disasters and unforeseen incidents, such as floods, earthquakes and hurricanes, the traditional communication infrastructure may become unavailable to support the emergency tele-operations. Under such circumstances, deploying unmanned aerial vehicles (UAVs) as small flying base stations is seen as a promising solution to provide real-time data communication between physicians and remote robots in both up-link and down-link directions with strict transmission requirements. This paper studies the joint optimization problem of resource allocation and UAVs positioning in UAV-assisted wireless networks with the goal of minimizing the number of deployed UAVs. Since the formulated problem is a non-convex mixed-integer non-linear programming problem, efficient heuristic and genetic solutions are proposed. Simulation results show that the proposed heuristic algorithm approaches the genetic one with an important reduction in computational complexity.

Fog computing (FC) emerged as a new paradigm enabling the deployment of new Internet of Things (IoT) applications. Fog infrastructure is composed of heterogeneous nodes characterized by a complex distribution, mobility, and sporadic resource availability. Hence, resource coordination for continuous Quality-of-Service (QoS) satisfaction becomes challenging, and accurate resource tracking is needed for flawless servicing. In this context, we investigate and propose online resource allocation solutions. The main objective is to maximize the number of satisfied users within a predefined latency requirement. Hence, we model the FC environment as a Markov Decision Process, and then, we formulate the optimization problem. Due to the problem’s NP-hardness, we leverage the reinforcement learning (RL) tool to develop resource allocation schemes. First, a centralized method where a smart fog controller possesses a global awareness of the FC environment is proposed. Next, a more practical and collaborative solution is presented, where each RL-enabled agent manages a group of fog nodes and their resources in order to satisfy computing requests. Based on real-world mobility data sets, simulation results illustrate the high efficiency of the proposed solutions with a preference for the collaborative approach. The superiority of our proposed solutions over state-of-the-art methods is also illustrated.

With the growing demand for openness, scalability, and granularity, mobile network function virtualization (NFV) has emerged as a key enabler for the most of mobile network operators. NFV decouples network functions from hardware devices. This decoupling allows network services, called Virtualized Network Functions (VNFs), to be hosted on commodity hardware which simplifies and enhances service deployment and management for providers, improves flexibility, and leads to efficient and scalable resource usage, and lower costs. The proper placement of VNFs in the hosting infrastructures is one of the main technical challenges. This placement significantly influences the network’s performance, reliability, and operating costs. The VNF placement is NP-Hard. Therefore, there is a need for placement methods that can cope with the complexity of the problem and find appropriate solutions in a reasonable duration. The primary purpose of this study is to provide a taxonomy of optimization techniques used to tackle the VNF placement problems. We classify the studied papers based on performance metrics, methods, algorithms, and environment. Virtualization is not limited to simply replacing physical machines with virtual machines or VNFs, but may also include micro-services, containers, and cloud-native systems. In this context, the second part of our article focuses on the placement of Containers Network Functions (CNFs) in edge/fog computing. Many issues have been considered as traffic congestion, resource utilization, energy consumption, performance degradation, etc. For each matter, various solutions are proposed through different surveys and research papers in which each one addresses the placement problem in a specific manner by suggesting single objective or multi-objective methods based on different types of algorithms such as heuristic, meta-heuristic, and machine learning algorithms.

Network Function Virtualization (NFV) is a network service deployment technology that reduces capital and operational costs while yielding flexibility and scalability for service operators. As such, an ordered chain of Virtual Network Functions (VNFs), known as a VNF Forwarding Graph (VNF-FG), should be composed and embedded into the underlying substrate network. In the literature, the composition and embedding stages of VNF-FGs are usually targeted separately, which may result in undesired solutions. In this paper, we propose our joint VNF-FG composition and embedding solution, which considers the variations of service demands while also accounting for dynamic network conditions. Specifically, our proposed solution relies on deep reinforcement learning empowered by two components for estimating dynamic parameters: network resource utilization and service demand analyzers. Moreover, to efficiently explore the problem’s large discrete action space, we utilize a specialized branching Q-network and enhance it with an action filtering mechanism. We evaluated our proposed method against joint and disjoint composition and embedding heuristics as well as versus other deep learning-based methods. Our results show that the proposed method can achieve up to a 95% improvement of embedding cost compared to our benchmarks.

This paper investigates the use of cell-free unmanned aerial vehicles (UAVs)-assisted wireless networks and optimizes the number of deployed UAVs under quality of service and coverage constraints. The formulated problem tackles the user-UAVs association, UAVs placement, channel assignment and transmit power allocation while considering both access and backhaul networks. Since the problem is a non-convex and non-linear mixed-integer programming, low-complexity efficient greedy-based algorithmic solutions are proposed. The first one finds the UAVs' best positions and allocates resource whereas the second one guarantees the problem feasibility (i.e., all users can be satisfied) by removing the worst users. For comparison purposes, a meta-heuristic solution based on the Particle Swarm Optimization technique is proposed. Simulation results illustrate the efficiency of the proposed algorithms in terms of the number of deployed UAVs in cell-free wireless networks.

Microservices are a promising technology for future networks, and many research efforts have been devoted to optimally placing microservices in cloud data centers. However, microservices deployment in edge and in-network devices is more expensive than the cloud. Additionally, several works do not consider the main requirements of microservice architecture, such as service registry, failure detection, and each microservice's specific database. This paper investigates the problem of placing components (i.e. microservices and their corresponding databases) while considering physical nodes' failure and the distance to service registries. We propose a Components-aware Microservices Placement for In-Network Computing Cloud-Edge Continuum (CaMP-INC). We formulate an Integer Linear Programming (ILP) problem with the objective of cost minimization. Due to the problem's$\mathcal{NP}$-hardness, we propose a heuristic solution. Numerical results demonstrate that our proposed solution CaMP-INC reduces the total cost by 15.8% on average and has a superior performance in terms of latency minimization compared to benchmarks.

In-network traffic classification is a class of in-network computing that brings significant benefits to the network, i.e., the first line of defence, classification at line rate and fast reaction time. However, it is still challenging to accurately and efficiently classify Internet traffic at an early stage due to a clear trade-off between flow identification time and classification accuracy - both are competing objectives. To this end,$w$e introduce a framework that focuses on deploying an accurate network traffic classifier inside a programmable data plane that can classify the traffic at maximal speed while considering the underlying constraints of the device. Notably,$w$e move from statistical feature-based traffic analysis and argue that traffic flow can be classified using a single feature called sequential packet size information as input. We evaluate our approach by identifying different types of IoT traffic inside a programmable data plane. Our findings demonstrate that accurate and early-stage network traffic classification is achievable with minor use of networking device resources.

The unprecedented increase of heterogeneous devices connected to the Internet, along with tight requirements of future networks, including 5G and beyond, poses new design challenges to network infrastructures. Collaborative computing, caching and communication paradigm together with artificial intelligence have the potential to enable the Next-Generation Networking Infrastructure (NGNI) that is needed to fulfill the stringent requirements of emerging applications. In this paper, we propose the SCORING project vision for reshaping the current network infrastructure towards an NGNI acting as a truly distributed, collaborative, and pervasive system that enables the execution of application-specific tasks and the storage of the related data contents in the Cloud-Edge-Mist continuum with high QoS/QoE guarantees.

Smal1 cell densification is recognized as one of the most significant characteristics in the fifth-generation of communication systems (5G) and beyond. A substantial capacity boost can be achieved at a low cost by supplementing macro networks with numerous small cells to create ultra-dense heterogeneous networks, which can serve as the foundation for the next generation of services. In this paper, we investigate a model that accounts for the location and channel quality of an enhanced Mobile Broadband (eMBB) user as well as the locations, density, and energy levels of a large number of Internet of Things (IoT) devices. More specifically, the eMBB user is randomly distributed in the coverage area of the MBS, and given its channel gain, it adjusts its transmit power to achieve an acceptable Quality of Service (QoS). In contrast, the IoT devices are gathered around SBS and regulate their transmission power in accordance with their energy budget to minimize energy-efficient utility function. Due to the coupling, the Stackelberg-Nash differential game is initially used to model the power allocation problem, with the eMBB user playing the role of the leader and the IoT devices playing the role of the followers. Then, we use the mean-field approximation to construct a hierarchical mean-field game from which we can recover a set of equations that may be solved iteratively to provide the optimal power allocation strategies. Simulation results illustrate the optimal power allocation strategies and show the effectiveness of the proposed approach.

The narrowband Internet of Things (NB-IoT) is a cellular technology introduced by the third-generation partnership project (3GPP) to provide connectivity to a large number of low-cost Internet of Things (IoT) devices with strict energy consumption limitations. However, in an ultradense small cell network employing NB-IoT technology, intercell interference can be a problem, raising serious concerns regarding the performance of NB-IoT, particularly in uplink transmission. Thus, a power allocation method must be established to analyze uplink performance, control and predict intercell interference, and avoid excessive energy waste during transmission. Unfortunately, standard power allocation techniques become inappropriate as their computational complexity grows in an ultradense environment. Furthermore, the performance of NB-IoT is strongly dependent on the traffic generated by IoT devices. In order to tackle these challenges, we provide a consistent and distributed uplink power allocation solution under spatiotemporal fluctuation incorporating NB-IoT features, such as the number of repetitions and the data rate, as well as the IoT device’s energy budget, packet size, and traffic intensity, by leveraging stochastic geometry analysis and mean-field game (MFG) theory. The effectiveness of our approach is illustrated via extensive numerical analysis, and many insightful discussions are presented.

As the World is still facing the COVID-19 pandemic, several researchers and industry players have proposed technological solutions to help fight the pandemic and pave the way for post-pandemic era precautions. In this matter, the potential benefits of remote health monitoring have been brought back to the spotlight. Indeed, with current advances in wireless communications, core network virtualization, and computing architectures as enablers, consistently guaranteeing the stringent quality-of-service (QoS) requirements of remote health monitoring, e.g., ultra-low latency, may be achievable. Notably, the fog computing (FC) paradigm has been advocated as a potential solution for remote health monitoring. However, the unreliability of fog nodes in FC networks is a critical aspect often overlooked despite its significant impact on vital latency requirements. This paper proposes a reliable fog-based remote health monitoring framework operating under uncertain fog computing conditions. Specifically, we formulate the problem of assigning tasks of remote sensors attached to patients to their adequate applications deployed in fog nodes aiming to maximize the number of satisfied tasks with respect to the fog nodes’ availability and communication latency constraints. Due to the problem’s NP-hardness, we leverage a differential evolution-based algorithm enhanced by reinforcement learning to deploy applications in fog nodes. Numerical results demonstrate the superior reliability performance of our proposed solution, in terms of the average success ratio of tasks, compared to benchmarks. Specifically, our simulations show up to 60 % performance improvement compared to benchmarks in specific scenarios. Moreover, by investigating the impact of several key parameters, we identify a design trade-off between the number of fog nodes and the latter’s intrinsic failure rates.

In this paper, we propose a collaborative smart solution for online traffic offloading among device-to-device (D2D) users underlying a cellular network. Specifically, we investigate the distributed pairing problem between requesting users and caching devices in their vicinity. Given that this problem is NP-hard, we propose a novel multi-agent reinforcement learning approach based on QMIX algorithm, where each requesting user is an agent capable of deciding to which cache device to pair, while respecting the quality-of-service of cellular users. Through simulations, we show the efficiency of the proposed algorithm in achieving D2D pairing. Finally, the impact of several parameters, such as the size of the network, size of files library, and communication requirements, is investigated.

Nowadays, unmanned aerial vehicles (UAVs) are being used in several novel applications, especially in the telecommunication domain. However, ensuring UAV communication and networking for the purpose of a specific application is still challenging. Indeed, due to the mobility of a UAV in a vast area, permanent connectivity over the backhaul is very sporadic and might be lost. In this article, we consider an aerial mesh network where each UAV can serve as a flying base station to boost terrestrial base station in case of damaged infrastructure case for example, or/and provide connectivity for uncovered or poorly covered nodes, and behaves as a relay to establish communication between two components owing to a lack of reliable direct communication link between them. We then detail a case study where a UAV-fleet is used to collect data from the ground Internet-of-Things (IoT) devices and forward it to the cloud for further processing passing by a remote gateway. We aim here to build a queueing framework, including network layer, MAC layer, and physical layer, and investigate both uplink and downlink communication links. Next, we derive some closed forms allowing us to predict the network performance in terms of traffic intensity at every UAV of the aerial mesh network, end-to-end (E2E) throughput, and E2E delay of ongoing streams. Next, we conduct extensive simulations to illustrate the benefit of our framework. Results discussion and numerous insights on parameter setting, target quality of service, and design consideration are also drawn.

The IPv6 routing protocol for low power and lossy networks (RPL) has some shortcomings, such as high packet loss rate and low network lifetime when used in Internet-of-Things (IoT) environments under heavy traffic. To overcome the RPL limitations, the current research tends to focus on a new paradigm of routing, referred to as Anycast Routing, in which a source node targets a set of destinations rather than a single one. In this article, we present a protocol that exploits the anycast perspective in the routing process since an important aspect of most IoT environments is to forward packets to a gateway, no matter which. Besides, we interconnect various instances of RPL to achieve better routing performance by offering the possibility of cooperation among various simultaneous instances of RPL within the network. Finally, to reach higher performance, we use a rank computation and parent selection mechanism that is different from those of the RPL. The evaluation results, which are obtained through simulation with the Cooja simulator, show that EM-RPL outperforms RPL in reducing the environmental footprint of the network, while it extends the network lifetime, decreases packet loss ratio, and better controls interpacket intervals and parent change overhead.

With the wide adoption of large-scale internet services and big data, the cloud has become the ideal environment to satisfy the ever-growing storage demand. In this context, data replication has been touted as the ultimate solution to improve data availability and reduce access time. However, replica management systems usually need to migrate and create a large number of data replicas over time between and within data centers, incurring a large overhead in terms of network load and availability. In this paper, we propose CRANE, an effiCient Replica migrAtion scheme for distributed cloud Storage systEms. CRANE complements any replica placement algorithm by efficiently managing replica creation in geo-distributed infrastructures in order to (1) minimize the time needed to copy the data to the new replica location, (2) avoid network congestion, and (3) ensure the minimum desired availability for the data. Through simulation and experimental results, we show that CRANE provides a sub-optimal solution for the replica migration problem with lower computational complexity than its integer linear program formulation. We also show that, compared to OpenStack Swift, CRANE is able to reduce by up to 60 percent the replica creation and migration time and by up to 50 percent the inter-data center network traffic while ensuring the minimum required data availability.

Remote robotic surgery, one of the most interesting 5G-enabled Tactile Internet applications, requires an ultra-low latency of 1 ms and high reliability of 99.999%. Communication disruptions such as packet loss and delay in remote robotic surgery can prevent messages between the surgeon and patient from arriving within the required deadline. In this paper, we advocate for scalable Gaussian process regression (GPR) to predict the contents of delayed and/or lost messages. Specifically, two kernel versions of the sequential randomized low-rank and sparse matrix factorization method ($\ell _{1}$-SRLSMF and SRLSMF) are proposed to scale GPR and address the issue of delayed and/or lost data in the training dataset. Given that the standard eigen decomposition for online GPR covariance update is cost-prohibitive, we employ incremental eigen decomposition in$\ell _{1}$-SRLSMF and SRLSMF GPR methods. Simulations were conducted to evaluate the performance of our proposed$\ell _{1}$-SRLSMF and SRLSMF GPR methods to compensate for the detrimental impacts of excessive delay and packet loss associated with 5G-enabled Tactile Internet remote robotic surgery. The results demonstrate that our proposed framework can outperform state-of-the-art approaches in terms of haptic data generalization performance. Finally, we assess the proposed framework’s ability to meet the Tactile Internet requirement for remote robotic surgery and discuss future research directions.

With the explosive demands for data, content delivery networks are facing ever-increasing challenges to meet end-users' quality-of-experience requirements, especially in terms of delay. Content can be migrated from surrogate servers to local caches closer to end-users to address delay challenges. Unfortunately, these local caches have limited capacities, and when they are fully occupied, it may sometimes be necessary to remove their lower-priority content to accommodate higher-priority content. At other times, it may be necessary to return previously removed content to local caches. Downloading this content from surrogate servers is costly from the perspective of network usage, and potentially detrimental to the end-user QoE in terms of delay. In this paper, we consider an edge content delivery network with vehicular nodes and propose a content migration strategy in which local caches offload their contents to neighboring edge caches whenever feasible, instead of removing their contents when they are fully occupied. This process ensures that more contents remain in the vicinity of end-users. However, selecting which contents to migrate and to which neighboring cache to migrate is a complicated problem. This paper proposes a deep reinforcement learning approach to minimize the cost. Our simulation scenarios realized up to a 70% reduction of content access delay cost compared to conventional strategies with and without content migration.

Robots are being used more and more in surgery due to the many benefits they bring (e.g. reduction of patient discomfort, precision, reliability). Remote robotic surgery is now expected to become a reality due to the emergence of 5G. Needle insertion is a crucial element of many robotic surgical procedures such as biopsies, injections, neurosurgery, and brachytherapy cancer treatment. During needle insertion in remote robotic surgery, there is still no guarantee that the surgeon will obtain the haptic feedback from the patient side within the stringent deadlines, even in 5G settings. This paper proposes a framework for learning by imitation as a way to predict the messages that will eventually fail to reach their destination within the required deadlines. By leveraging expert demonstrations, the Hidden Markov Model is used to encapsulate a set of expert force/torque profiles and corresponding parameters during the off-line training process. A Gaussian mixture regression is then used to reproduce a generalized version of the force/torque profile and corresponding parameters during the prediction. Simulations are conducted to evaluate the performance of the proposed method. They show that our proposed framework is able to execute predictions in much less than the 1ms end-to-end latency requirement of remote robotic surgery.

Using Information Centric Networks (ICNs) instead of IP networks will improve Quality of Experience (QoE) by enabling efficient and scalable Content Delivery Networks (CDNs). However, Information Centric based CDNs still face many challenges. The rate adaptation algorithms used in IP based CDNs can for instance lead to inaccurate estimations of Round-Trip Times (RTTs) in ICN settings. Furthermore, the network storage feature of ICN can introduce major oscillations in adaptive streaming. In this paper, we use the ETSI Multi Access Edge Computing (MEC) to tackle these issues. An overall system view (which includes the MEC server) and a novel rate adaptation algorithm are proposed. The proposed rate adaptation algorithm is validated with simulations and the results show that it significantly improves users' QoE due to higher accuracy and stability in rate estimation.

In this paper, we investigate the uplink power allocation problem in a large scale environment for user devices with multi-homing capabilities. We introduce an analytical model for multi-homing ultra-dense heterogeneous networks, which takes into account spatial randomness and user device diversity. Coupling stochastic geometry analysis and mean-field approximation, we formulate the problem as a mean-field optimal control with two populations. Then, the optimality conditions are derived using Lagrangian dual formulation to obtain the mean-field equilibrium. Finally, by using a finite difference method, we illustrate the optimal transmit power for both uni-homed and dual-homed devices.

The advent of a new breed of enhanced multimedia services has put network operators into a position where they must support innovative services while ensuring both end-to-end Quality of Service requirements and profitability. Recently, Network Function Virtualization (NFV) has been touted as a cost-effective underlying technology in 5G networks to efficiently provision novel services. These NFV-based services have been increasingly associated with multi-domain networks. However, several orchestration issues, linked to cross-domain interactions and emphasized by the heterogeneity of underlying technologies and administrative authorities, present an important challenge. In this paper, we tackle the cross-domain interaction issue by proposing an intelligent and profitable auction-based approach to allow inter-domains resource allocation.

Slices were introduced in 5G to enable the co-existence of applications with different requirements on a single infrastructure. Slices may be delay-constrained for mission-critical applications such as Tactile Internet applications. When delay-constrained slices are implemented as collections of virtual network function (VNF) chains, a key challenge is to place the VNFs and route the traffic through the chains to meet a strict delay constraint. Parallel VNF processing has been proposed as a promising approach. However, this approach increases the number of physical nodes in the chains, and thus decreases the reliability, which is also critical for Tactile Internet applications. Furthermore, the cost depends upon the specific VNF placement and traffic routing, as nodes and links are heterogeneous. This article tackles the issues of reliability and cost when embedding delay-constrained slices. We model the problem as an optimization problem that minimizes reliability degradation and cost while ensuring the strict delay constraint when a parallel VNF processing approach is used. Due to the complexity of the formulated problem, we also propose a Tabu search-based algorithm to find sub-optimal solutions. The results indicate that our proposed algorithm can significantly improve cost and reliability while meeting a strict delay constraint.

In recent years, fog computing has increasingly become popular with the advent of Internet of Things (IoT) applications characterized by strict Quality of Service (QoS) requirements. To deploy applications, applications are typically decomposed into services then embedded with fog nodes. However, an overlooked aspect in container placement strategies is the heterogeneous inter-container network communication technologies and their impact on application performances in fog networks. We propose and evaluate in this paper, a near optimal genetic algorithm based container placement strategy that takes into account Remote Direct Memory Access as well host and overlay mode for inter-container communication to ensure application response time requirements.

IoT sensor networks are applied in several areas, including research, security, and monitoring. All these applications are based on data collection. A practical solution is a use of unmanned aerial vehicles (UAVs) as aerial relays to ensure connectivity between IoT-devices and the destination node. Most applications have been based on using a single UAV to transmit information, while our work focuses on forming a network of multiple UAVs. In this paper, a novel deployment of UAVs moving along non-concentric rings is analyzed. First, they collect packets generated by IoT-devices and then transmit them throughout a UAV-to-UAV forwarding schema to the destination. We derive a closed formula for the end-to-end throughput of the proposed gathering network architecture.

5G networks are required to provide ultra reliable low latency communications while dealing with the permanent growth of data traffic. In Heterogeneous Networks (Hetnets) assisted with Device-to-Device (D2D) communications, traffic can be offloaded to small base stations or to devices in order to improve the transmission delays even with small caching. In this paper, we aim at reducing the average content delivery delay by optimizing the caching placement strategy in the context of D2D-assisted Hetnets. First, we analytically derive an upper bound on the average content delivery delay. Then, we formulate the problem of minimizing this upper bound through caching placement. The optimal solution is obtained for a single file, then used to propose a low-complex heuristic solution for multiple files. Numerical results illustrate the efficiency of our solution compared to other strategies.

Streaming service is continuously growing, which makes it the killer application of current 4G networks, as it demands more resources from mobile networks. The cellular network receives a large number of requests, most times for the same content, consuming the spectrum, energy, in addition to monetary costs inefficiently. In order to optimize spectrum utilization and reduce the induced costs, we propose a noncooperative game framework allowing to understand the user's behaviors. We observed a volunteer Dilemma-like situation when a mobile user could stream the requested video to its neighbors over a D2D link. Afterward, we provide a full description of both pure and mixed Nash equilibria (NE). Furthermore, to ensure convergence to NE points, we use linear reward-inaction and Gibbs Boltzmann learning algorithms. Finally, we show how our scheme could be exploited through extensive numerical simulation. Our framework capture the user's selfish behavior and provides a solution regarding setting and parameters allowing to reach high performance in terms of spectrum utilization, energy efficiency, and overall cost.

Fog computing has emerged as a promising technology that can bring cloud applications closer to the devices at the network edge. The fog infrastructure contains mainly distributed and heterogeneous fog devices such as in the context of the Internet of Things. Unlike traditional data centers, those devices are characterized by sporadic resources availability, mobility, and increased flexibility. However, resource allocation mechanisms proposed currently for fog computing still lack the support of dynamic behavior. In this article, we propose novel resource management algorithms capable of flexible service provisioning in a dynamic fog computing environment. Specifically, the joint problem of container placement and task provisioning is formulated with integer linear programming. Due to its NP-hardness, we propose a low-complex particle-swarm-optimization-based metaheuristic and a greedy heuristic. Our solutions aim to optimize the number of served end-users with a predefined delay-threshold while considering dynamic fog nodes behavior/mobility and resources availability of fog nodes. Using real-world mobility data sets and different resources' availability models, conducted simulations demonstrate that the PSO-based algorithm achieves near-optimal results. Whereas, the greedy algorithm realizes only 10%-30% less success ratio than the optimal solution with negligible execution time.

Multimedia conferencing is the conversational exchange of multimedia content between multiple parties. It has a wide range of applications (e.g., massively multiplayer online games (MMOGs) and distance learning). Media handling services (e.g., video mixing, transcoding, and compressing) are critical to multimedia conferencing. However, efficient resource usage and scalability still remain important challenges. Unfortunately, the cloud-based approaches proposed so far have several deficiencies in terms of efficiency in resource usage and scaling, while meeting quality of service (QoS) requirements. This paper proposes a solution which optimizes resource allocation and scales in terms of the number of participants while guaranteeing QoS. Moreover, our solution composes different media handling services to support the participants' demands. We formulate the resource allocation problem mathematically as an integer linear programming (ILP) problem and design a heuristic for it. We evaluate our proposed solution for different numbers of participants and different participants' geographical distributions. Simulation results show that our resource allocation mechanism can compose the media handling services and allocate the required resources in an optimal manner while honoring the QoS in terms of end-to-end delay.

Orchestration and management of cloud computing entities necessitate measuring and analysis of real-time monitored performance metrics. However, decision making in current management platforms are addressed separately in different cloud stack layers. These isolated active management decisions may degrade the total performance of the cloud system. Since, cloud computing platforms lack an integrated analytics and management capability, in this paper, we propose an integrated platform to detect and predict situations where corrective actions are required. First, a Dynamic Bayesian Network (DBN) is trained and updated by collected data to calculate the causal dependencies among various entities in different cloud service layers. The correlation values are then fed into a Long Short-Term Memory (LSTM) neural network to predict the future states. States that violate the Service Level Agreement(SLA) of cloud services are learned with training data, and if the forecasted states threaten the SLA of cloud services, associated events are generated to trigger management actions. Next, management actions are assigned a different set of events using a reinforcement learning approach. A set of experiments based on collected data from a real cloud service environment is conducted to validate the proposed approach. Experimental results indicate that the proposed method outperforms the current management solutions and improves web request response time by up to 7% and decreases SLA violation by 79% in the context of web application auto-scaling.

Device-to-Device (D2D) communication has been proposed as a promising technique to improve resource utilization in fifth generation (5G) cellular networks by offloading the traffic from backhaul to local direct links. The throughput for delivery of multimedia content can be greatly enhanced by D2D communications, where direct links between pairs of user devices are set up without involving a central base station. In this paper, we propose the scheduling algorithms for effectively sharing the multimedia content using D2D communication. The proposed algorithms ensure the liveness for live video streaming while reducing the stall events for on-demand video streaming. Through extensive simulation, the proposed approaches are shown to provide sizeable gains compared with existing solution.

User Generated Videos (UGV) are the dominating content stored in scattered caches to meet end-user Content Delivery Networks (CDN) requests with quality of service. End-User behaviour leads to a highly variable UGV popularity. This aspect can be exploited to efficiently utilize the limited storage of the caches, and improve the hit ratio of UGVs. In this paper, we propose a new architecture for Data Analytics in Cloud-based CDN to derive UGVs popularity online. This architecture uses RESTful web services to gather CDN logs, store them through generic collections in a NoSQL database, and calculate related popular UGVs in a real time fashion. It uses a dynamic model training and prediction services to provide each CDN with related popular videos to be cached based on the latest trained model. The proposed architecture is implemented with k-means clustering prediction model and the obtained results are 99.8% accurate.

Ethernet has been proposed for the 5G fronthaul to transport the Common Public Radio Interface (CPRI) traffic between the radio equipment (RE) and the radio equipment control (REC). The advantages of adopting an Ethernet transport are threefold 1) the low cost of equipment, 2) the use of a shared infrastructure with statistical multiplexing, as well as 3) the ease of operations, administration and maintenance (OAM). In this paper, we introduce distributed timeslot scheduler for CPRI over Ethernet (DTSCoE) as a scheduling algorithm for IEEE 802.1Qbv to support CPRI traffic. DTSCoE is built upon the stream reservation protocol (SRP) IEEE 802.1Qcc to propagate timeslot information across the datapath without any centralized coordination. The simulation results demonstrate that DTSCoE reduces one-way delay to minimum and reduces the jitter to zero which satisfies the CPRI requirements.

Multimedia conferencing is used extensively in a wide range of applications, such as online games and distance learning. These applications need to efficiently scale the conference size as the number of participants fluctuates. Cloud is a technology that addresses the scalability issue. However, the proposed cloud-based solutions have several shortcomings in considering the future demand of applications while meeting both Quality of Service (QoS) requirements and efficiency in resource usage. In this paper, we propose an Adaptive and Dynamic Scaling mechanism (ADS) for multimedia conferencing services in the cloud. This mechanism enables scalable and elastic resource allocation with respect to the number of participants. ADS produces a cost efficient scaling schedule while considering the QoS requirements and the future demand of the conferencing service. We formulate the problem using Integer Linear Programming (ILP) and design a heuristic for it. Simulation results show that ADS mechanism elastically scales conferencing services. Moreover, the ADS heuristic is shown to outperform a greedy algorithm from a resource-efficiency perspective.

Offloading the network, minimizing the power consumption as well as reducing interference are important issues in wireless networks. These requirements mandates that future cellular networks need to use Device-to-Device communication as a key enabler. To harness this solution, we propose a two-device system that combines cellular and Device-to-Device (D2D) communication in an uplink communication. We model this system as a quitting game where devices choose simultaneously either to continue or to quit transmitting over the cellular network. The devices will strategically choose whether to compete or to cooperate through mobile relaying. We first calculate the throughput and the outage probability in a fading channel, then we find the Sub-game Perfect Equilibrium of this game by determining the pure and mixed Nash equilibrium of each subgame. Results show that the outage probability depends on the transmission power and the distance separating a device from its serving BS. The quitting decision of devices depends on the fraction of throughput they would get after quitting, on the quitting frame and on the quitting regret.

Recently, UAV has provided a significant role to support the wireless network, thanks to the several advantages that can offer in comparison with the terrestrial base station. In this paper, we aim to deal with data collection and charging depletion ground IoT devices through UAV station, which is used as a flying base station. To extend the network lifetime, we present a novel use of UAV with energy harvesting module. Thus, the UAV can be used as an energy source to serve depleted IoT devices. On one hand, the UAV charges the depletion ground IoT devices starting initially with those with battery level under a certain threshold. On the other hand, the UAV station collects data from IoT devices that have sufficient energy to transmit their packets, and in the same phase, the UAV exploits the RF signals transmitted by IoT devices to extract and harvest energy. Furthermore, and as the UAV station has a limited coverage time due to its known energy constraint, we investigate in this work the trade- off between both times that the UAV reserves to devices in order to recharge them and to collect data. Numerical results evaluate different metrics of performances in which we examine the added value of UAV with energy harvesting module.

Traditional Content Delivery Networks (CDNs) built with traditional Internet technology are less and less able to cope with today's tremendous growth of content. Information Centric Networks (ICN), a proposed future Internet technology, may aid in remedying the situation. Unlike the current Internet, it decouples information from its sources and provides in- network storage. We expect traditional CDN and ICN-based CDN to co-exist in the foreseeable future, especially as it is now known that it might be possible to evolve traditional CDNs to gain the benefits promised by ICN. 5G providers must therefore aim to offer core network slices on which both ICN-based CDNs and traditional CDNs can be built. These slices could of course also be offered to providers of other applications with requirements similar to those of content delivery. This paper tackles the problem of slicing 5G for content delivery over ICN- based CDNs and traditional CDNs. Only virtualized Evolved Packet Core (EPC)-based 5G is considered. The problem is defined as a resource allocation problem which aims at minimizing the cost of slice assignment, while meeting QoS requirements. An Integer linear programming (ILP) formulation is provided and evaluated in a small-scale scenario.

Information Centric Network emerges a paradigm shift from host centric to information centric communication model. ICN deployment faces a significant challenge in poorly developed telecommunications infrastructures where end-to-end paths are not guaranteed. In this context, delay-tolerant networks (DTN) were introduced as an initiative that efficiently handles network interruptions. This paper proposes a new incentive approach for content caching, by exploiting ICN and DTN advantages while avoiding their shortcomings. The proposed caching solution intends to achieve a logical trade-off between the dissemination rate and the energy efficiency in such environments. To this end, we propose a reputation-based content caching mechanism where DTN stations (relays) attempt to deliver content stored in ICN stations. Our approach is designed to ensure an efficient equilibrium between the overall delivery probability and the energy consumption. Furthermore, the network performance is exhibited through a numerical investigation. We conclude that the proposed mechanism can significantly enhance the caching and the energy efficiency by achieving an optimum dissemination rate.

Device-to-device (D2D) communications combined with Heterogeneous networks (Hetnets) has attracted growing interest. Indeed, Hetnets deploy small-cells within macro-cells in order to offload traffic and improve the overall network coverage and capacity. Whereas, D2D promotes the use of communications between users for content delivery without going through the small or macro bases stations. Hence, it reduces communication delays and improves the spectral efficiency. In this context, we aim in this paper at reducing the average transmission delay, defined as the average sum delays of contents transmission to satisfy users' requests in a macro-cell, by jointly optimizing caching placement and channel resource allocation, in cache-enabled Hetnet with D2D assistance. At first, a lower-bound expression of the average transmission delay is derived. Then, the optimization problem is formulated. Afterwards, we propose a sub-optimal random search algorithm and a low-complexity greedy algorithm that solve the problem. Finally, numerical results illustrate the performances of the proposed algorithms.

Device-to-Device (D2D) communication has gained interest as a promising technology for next generation wireless networks. D2D communication promotes the use of point-to-point communications between users without going through the base stations. In this paper, we aim at maximizing the sum rate of a D2D network, under the assumption of realistic time-varying channels and D2D interference. Specifically, we formulate channels as Finite-State Markov Channels (FSMC). With realistic FSMC, the complexity of the problem is high. Consequently, we propose the use of a centralized Deep Reinforcement Learning (DRL) transmission scheme for D2D communications, where transmission decisions are taken by one agent that has a global knowledge of the D2D network. We compare the DRL-based scheme with other transmission schemes. The results show that it outperforms other approaches in terms of achieved sum rate.

Value-added services (e.g., overlaid video advertisements) have become an integral part of today's content delivery networks (CDNs). To offer cost-efficient, scalable, and more agile provisioning of new value-added services in CDNs, network functions virtualization paradigm may be leveraged to allow implementation of fine-grained services as a chain of virtual network functions (VNFs) to be placed in CDN. The manner in which these chains are placed is critical as it both affects the quality of service (QoS) and provider cost. The problem is however, very challenging due to the specifics of the chains (e.g., one of their end-points is not known prior to the placement). We formulate it as an integer linear program and propose a cost efficient proactive VNF placement and chaining algorithm. The objective is to find the optimal number of VNFs along with their locations in such a manner that the cost is minimized while QoS is met. Apart from cost minimization, the support for large-scale CDNs with a large number of servers and end-users is an important feature of the proposed algorithm. Through simulations, the algorithm's behavior for small-scale to large-scale CDN networks is analyzed.

Due to the recent surge in end-users demands, value-added video services (e.g. in-stream video advertisements) need to be provisioned in a cost-efficient and agile manner in Content Delivery Networks (CDNs). Network Function Virtualization (NFV) is an emerging technology that aims to reduce costs and bring agility by decoupling network functions from the underlying hardware. It is often used in combination with Software Defined Network (SDN), a technology to decouple control and data planes. This paper proposes an NFV and SDN-based architecture for a cost-efficient and agile provisioning of value-added video services in CDNs. In the proposed architecture, the application-level middleboxes that enable value-added video services (e.g. mixer, compressor) are provisioned as Virtual Network Functions (VNFs) and chained using application-level SDN switches. HTTP technology is used as the pillar of the implementation architecture. We have built a prototype and deployed it in an OPNFV test lab and in SAVI, a Canadian distributed test bed for future Internet applications. The performance is also evaluated.

Content placement (CP) algorithms are an integral component of Cloud-based Content Delivery Networks (CCDNs) that select a subset of content from the myriad catalogue, to be placed on surrogates to meet end-user requests with quality of service (QoS). It is challenging to conjure popularity of content, due to size of the catalogue, the heavy and long tail nature of the popularity distribution function and complexity arising from Online Social Networking (OSN) relationships. Therefore, we leverage hierarchical organization of surrogates to push and store content that is preemptively and strategically chosen, such that popular and correlated content always remains within QoS distance of each other. We design an Integer Linear Programming (ILP) model and solve it optimally and nearoptimally using CPLEX and Particle Swarm Optimization-based heuristic, respectively. We compare our model with state-of- the-art CP algorithm to show the benefits of popularity and correlation aware CP.

Data center applications require strict characteristics regarding packet loss, fairness, head-of-line blocking, latency, and low processing overhead. Motivated by the emergence of IEEE Data Center Bridging, we explore the possibility of controlling congestion in Ethernet layer while guaranteeing those requirements. We propose HetFlow (Heterogeneous Flow) as a delay-based congestion control mechanism that controls & prevents congestion while achieving minimum queue length, minimum network latency, and high link utilization. HetFlow guarantees fairness between flows with different packet sizes and different round-trip times (RTTs). The results obtained through prototype and simulations show that HetFlow succeeded in preventing congestion and achieving low queue length, high link utilization, and fairness between flows.

In the era of Big data, cloud storage services have become the option of choice to store and share data thanks to their cost-effectiveness and seemingly limitless capacity. The increasing success of these services is driving cloud providers to further improve their storage management systems in order to offer more stringent guarantees on data availability and access time. However, despite recent efforts towards this goal, existing solutions have largely overlooked the heterogeneity of the workloads and the underlying storage components in terms of failure rates, capacity and I/O speed. To fill this gap, we present in this paper a heterogeneity-aware data management scheme (dubbed Heron) based on a genetic algorithm that takes into consideration disk heterogeneity to satisfy SLA requirements in terms of access time and availability and minimizes costs in terms of data migration, storage and energy consumption. Through realistic simulations, we show that Heron significantly improves data availability and access time and ensures minimal storage costs and data migration overhead compared to heterogeneity-oblivious solutions.

Content Delivery Networks (CDN) and their globally dispersed caches host a myriad of User Generated Videos (UGV) to meet end-user requests with quality of service. To efficiently utilize the limited storage of the caches, it is imperative to improve the hit ratio of UGVs. In contrast to the traditional static content, UGV popularity is highly dynamic and dependent on end-user behavior. Therefore, we devise a novel popularity prediction model for UGV, using a hybrid regression model. Our hybrid regression model dynamically adapts the popularity of UGV that is built from a historical training dataset. We reduce error in predicting popularity by up to 14%, when compared to pure offline and online approaches, with a small increase in the execution time and memory overhead. Our novel popularity prediction model accounts for end- user behavior by considering the end-user video watch time and the number of shares for the UGVs. To improve cache performance in CDN, we employ a cache replacement strategy that leverages our popularity prediction model to efficiently evict the less popular UGVs for more popular content. We compare our novel cache replacement strategy with the traditional and state-of-the-art cache replacement strategies and show an increase in the average hit ratio of up to 74% and 7%, respectively, for UGVs with shortterm popularity.

Multimedia conferencing is the real-time exchange of multimedia content between multiple parties. It is the basis of a wide range of applications (e.g., multimedia multiplayer game). Cloud-based provisioning of the conferencing services on which these applications rely will bring benefits, such as easy service provisioning and elastic scalability. However, it remains a big challenge. This paper proposes a PaaS for conferencing service provisioning. The proposed PaaS is based on a business model from the state of the art. It relies on conferencing IaaSs that, instead of VMs, offer conferencing substrates (e.g., dial-in signaling, video mixer and audio mixer). The PaaS enables composition of new conferences from substrates on the fly. This has been prototyped in this paper and, in order to evaluate it, a conferencing IaaS is also implemented. Performance measurements are also made.

Abstract This paper considers uplink channel access in a zone of closed‐access small‐cells that is deployed in a macrocell service area. All small‐cell user equipments (SUEs) have access to a common orthogonal set of channels, leading to intercell interference. In addition, each channel forms a separate collision domain in each cell, thus can be successfully used only by one SUE of that cell. This paper proposes two non‐cooperative Bayesian games, G1 and G2, that are played among the SUEs. G1 assumes the availability of channel state information at the transmitters, while G2 assumes the availability of only the distribution of the channel state information. Each SUE can choose to transmit over one of the channels or not to transmit. The emphasis of the paper is on the set of symmetric threshold strategies where the Nash equilibrium is fully determined by a single parameter. The existence and uniqueness of pure Bayesian–Nash symmetric equilibrium of G1 in threshold strategies and mixed Bayesian–Nash symmetric equilibrium of G2 in uniformly distributed threshold strategies are proven. Numerical results corroborate the theoretical findings and benchmark against another decentralized scheme. Copyright © 2015 John Wiley & Sons, Ltd.

Multimedia conferencing is the conversational exchange of multimedia content between multiple parties. It has a wide range of applications (e.g. Massively Multiplayer Online Games (MMOGs) and distance learning). Many multimedia conferencing applications use video extensively, thus video mixing in conferencing settings is of critical importance. Cloud computing is a technology that can solve the scalability issue in multimedia conferencing, while bringing other benefits, such as, elasticity, efficient use of resources, rapid development, and introduction of new applications. However, proposed cloud-based multimedia conferencing approaches so far have several deficiencies when it comes to efficient resource usage while meeting Quality of Service (QoS) requirements. We propose a solution to optimize resource allocation for cloud-based video mixing service in multimedia conferencing applications, which can support scalability in terms of number of users, while guaranteeing QoS. We formulate the resource allocation problem mathematically as an Integer Linear Programming (ILP) problem and design a heuristic for it. Simulation results show that our resource allocation model can support more participants compared to the state-of-the-art, while honoring QoS, with respect to end-to-end delay.

Content Placement (CP) problem in Cloud based Content Delivery Networks (CCDNs) leverage resource elasticity to build cost effective CDNs that guarantee QoS. In this paper, we present our novel CP model, which optimally places content on surrogates in the cloud, to achieve (a) minimum cost of leasing storage and bandwidth resources for data coming into and going out of the cloud zones and regions, (b) guarantee Service Level Agreement (SLA), and (c) minimize degree of QoS violations. The CP problem is NP Hard, hence we design a unique push based heuristic, called Weighted Social Network Analysis (W SNA) for CCDN providers. W-SNA is based on Betweeness Centrality (BC) from SNA and prioritizes surrogates based on their relationship to the other vertices in the network graph. To achieve our unique objectives, we further prioritize surrogates based on weights derived from storage cost and content requests. We compare our heuristic to current state of the art Greedy Site (GS) and purely Social Network Analysis (SNA) heuristics, which are relevant to our work. We show that W-SNA outperforms GS and SNA in minimizing cost and QoS. Moreover, W-SNA guarantees SLA but also minimizes the degree of QoS violations. To the best of our knowledge, this is the first model and heuristic of its kind, which is timely and gives a fundamental pre allocation scheme for future online and dynamic resource provision for CCDNs.

In this paper, the user association problem under quality of service (QoS) requirements in a heterogeneous and small cells network (HetSNet) is considered. We have shown in a previous work that this problem is NP-hard and thus cannot be solved optimally in polynomial time unless P = NP. Therefore, new suboptimal algorithms are needed in order to solve it efficiently. Even though, it is very hard to implement the suboptimal algorithm in a centralized fashion because it needs a high amount of information exchange between the base stations and the users and it suffers from a huge computational complexity. Thus, in this paper, we model the problem of user association in HetSNets as a non-cooperative game and we propose a completely distributed algorithm inspired by the theory of learning to solve it. Specifically, we propose a modified win-stay-lose-shift learning model in order to converge to a near optimal user association. We evaluate by simulations the performance of the proposed algorithm and and we show that it is close to the performance of the computationally complex optimal centralized algorithm which assumes complete channel information knowledge.

We consider a point-to-point wireless link with automatic repeat request (ARQ) based packet transmission where both the transmitter and receiver nodes are energy harvesting (EHNs). Transmitter EHN has access to low-grade channel state information (CSI) as it is implicitly obtained from ARQ feedback. Furthermore, signal processing tasks such as sampling and decoding at the receiver EHN can be interrupted if there is insufficient energy in the battery that cause loss of packet and wastage of harvested energy both at the transmitter and receiver EHNs. We propose selective sampling (SS) scheme where only part of the transmitted packet is sampled and stored depending on the receiver nodes stored energy. SS information (SSI) is then fed back to the transmitter. Packet decoding is not performed until full packet is constructed. Hence, we modify the conventional ARQ messages, i.e., ACK/NAK by adding few more bits to carry additional SSI as well. Another objective is to find the optimal power allocation policy to adapt to the low-grade CSI and SSI available at the transmitter such that harvested energy can be utilized efficiently especially at the receiver. Furthermore, using a decision-theoretic framework, we propose greedy power allocation scheme to evaluate the performance of the proposed retransmission scheme. In numerical examples, we illustrate that our proposed scheme has lower average packet transmission time and packet drop probability (PDP) compared to the equal power allocation and greedy power allocation with conventional retransmission scheme.

Switch fabric in routers requires very tight characteristics in term of packet loss, fairness in bandwidth allocation, no head-of-line blocking and low latency. Such attributes are traditionally resolved using specialized and expensive switch devices. Motivated by the emergence of IEEE Data Center Bridging, we explore the possibility of using commodity Ethernet switches to achieve scalable, flexible, and more cost efficient solutions, while still guaranteeing the switch characteristics. In this context, we propose Ethernet Congestion Control & Prevention (ECCP), a novel concept to control and prevent congestion in switch fabrics. ECCP consists of (1) a method to estimate the available bandwidth along a given network path using a train of probes and (2) a rate control algorithm to adjust the sending rate of traffic along this path based on the estimated bandwidth. To prove ECCP and evaluate its characteristics, we present a first prototype based on the OMNEST simulator and conduct extensive experiments. Our analysis confirms that ECCP is a viable solution to (1) avoid congestion within the fabric, thus minimizing path latency and avoiding packet loss, (2) guarantee fair share of the link capacity between flows, and (3) avoid head of line blocking.

Finding an effective and simple unified control plane (UCP) for IP/Dense Wavelength Division Multiplexing (DWDM) multi-layer optical networks is very important for network providers. Generalized Multi-Protocol Label Switching (GMPLS) has been in development for decades to control optical transport networks. However, GMPLS-based UCP for IP/DWDM multi-layer networks is extremely complex to be deployed in a real operational products because still there are a lot of non-capable GMPLS equipments. DRAGON (Dynamic Resource Allocation via GMPLS Optical Networks) [1] is a software that solves this issue making these equipments capable for working in a GMPLS network. On the other hand, OpenFlow (OF), one of the most widely used SDN (Software Defined Networking) implementations, can be used as a unified control plane for packet and circuit switched networks [2]. In this paper, we propose and experimentally evaluate two solutions using OpenFlow to control both packet and optical networks (OpenFlow Messages Mapping and OpenFlow Extension). These two solutions are compared with GMPLS-based UCP. The experimental results show that the OpenFlow Extension solution outperforms the OpenFlow Messages Mapping and GMPLS solutions.

Minimum-collisions assignment (MCA), in a wireless network, is the distribution of a finite resource set, such that the number of neighbor cells which receive common elements is minimized. In classical operator deployed networks, resources are assigned centrally. Heterogeneous networks contain user deployed cells, therefore centralized assignment is problematic. MCA includes orthogonal frequency bands, time slots, and physical cell identity (PCI) allocation. MCA is NP-complete, therefore a potential-game-theoretic model is proposed as a distributed solution. The players of the game are the cells, actions are the set of PCIs and the cost of a cell is the number of neighbor cells in collision. The price of anarchy and price of stability are derived. Moreover the paper adapts a randomized-distributed-synchronous-update algorithm, for the case, when the number of PCIs is higher than the maximum degree of the neighbor relations graph. It is proven that the algorithm converges to a optimal pure strategy Nash equilibrium in finite time and it is robust to node addition. Simulation results demonstrate that the algorithm is sub-linear in the size of the input graph, thus outperforms best response dynamics.

This paper addresses the problem of competition vs. cooperation in the downlink, between base stations (BSs), of a multiple input multiple output (MIMO) interference, heterogeneous wireless network (HetNet). This research presents a scenario where a macrocell base station (MBS) and a cochannel femtocell base station (FBS) each simultaneously serving their own user equipment (UE), has to choose to act as individual systems or to cooperate in coordinated multipoint transmission (CoMP). The paper employes both the theories of non-cooperative and cooperative games in a unified procedure to analyze the decision making process. The BSs of the competing system are assumed to operate at the maximum expected sum rate (MESR) correlated equilibrium (CE), which is compared against the value of CoMP to establish the stability of the coalition. It is proven that there exists a threshold geographical separation, dth, between the macrocell user equipment (MUE) and FBS, under which the region of coordination is non-empty. Theoretical results are verified through simulations.

Cognitive radio networks are emerging kind of wireless networks with cognitive radio nodes able to have dynamic spectrum access in order to make use more efficiently of the spectrum. The routing problem in such networks is quite complex due to the dynamic nature of the spectral environment where the availability of frequency bands for cognitive radio nodes is opportunistic. In this paper, we propose a robust and efficient routing solution in terms of throughput. Our proposition is a reactive routing protocol (named cognitive temporary ordering routing algorithm) CO-TORA based on the classic TORA protocol proposed for non cognitive wireless ad-hoc networks. We also implement CO-TORA in largely used NS-2 simulator. The utilization of CO-TORA protocol brings to the system many performance improvements that are evaluated by simulations and shown by comparing it with classical TORA.

Femtocell and WiFi play crucial roles in sustaining the continued growth in mobile traffic. Deploying Femtocells in WiFi hotspots would allow the access providers to provide more capacity for users and improve their quality of experience during mobility. Hence, the co-existence of Femtocell and WiFi carries critical importance for improving the total performance of the users and meeting the promised quality of service (QoS) satisfaction of Femtocell end users. In this paper, we propose and develop a framework allowing to make use of unlicensed band and to increase the total throughput of Femtocells while offloading the traffic of Femtocell users to unlicensed bands in case of severe interference with Macrocell. The channel access of both Femtocell and WiFi networks are analytically modeled and numerically verified. Moreover, the effects of WiFi channel access parameters on the performance of WiFi and Femtocell networks are investigated. Numerical evaluation of our proposed scheme show that by adequately tuning and giving priority, the throughput of small cells and utilization of unlicensed spectrum have been improved.

Reducing the energy consumed in cloud computing is becoming one of the most challenging research directions due to the overwhelming growth of services that are hosted and delivered by cloud computing. Indeed, the energy consumed by data transport represents a significant percentage according to the overall consumption of the cloud. Hence, by exploiting network and router virtualization technologies, we firstly propose a Green Cloud Architecture (GCA), where we can either shut down, or make in sleeping mode virtual routers; or migrate virtual routers towards another physical router according to energy-awareness. Secondly, we evaluate our green cloud architecture by proposing an energy-aware resource allocation algorithm. The mapping algorithm is evaluated through simulations and our green architecture significantly reduces the power consumption during data transport by up to 41%.

This paper considers the allocation of frequency and time resources in a heterogeneous network, in a self-organizing manner. The general problem is to assign a resource set, so as to minimize the number of pairs of adjacent base stations that obtain the same resource. This can be modeled by Minimum-Collisions Coloring (MCC) on an undirected graph, where the colors are the resources, the vertices are the wireless nodes and the edges represent interference relations between nodes. The MCC decision problem is NP-complete. This paper develops a game-theoretic model for the MCC problem. The players of this game are a set of colored agents, which in practice could be software robots. The game is proven to possess multiple pure-strategy Nash Equilibria (NEs). Then a swapping mechanism is developed to improve the NE performance and the resulting coloring is shown to be pairwise-Nash stable. Further refinement is proposed by making use of an external referee. All theoretical results are corroborated through simulations.

Femtocell and WiFi are often presented as opposing technologies. The truth is that both of them play a crucial role in sustaining the continues growth in mobile traffic. In many cases both technologies will eventually be employed in a single box with access via an intelligent mobile device that will automatically select the best option. Deploying Femtocells in WiFi hotspots would let access providers add 3G capacity for users who do not have WiFi on their device and improve their quality of experience during mobility. Partitioning of the spectrum resources carries critical importance for maximizing the total capacity and quality of service (QoS) satisfaction of end users. This paper proposes a fair and QoS-based unlicensed spectrum splitting strategy between WiFi and Femtocell networks. Numerical results show that spectrum splitting under total capacity maximization constraint allows for unfair spectrum allocation, while a more equitable spectrum splitting can be accomplished by taking into account the fairness and QoS constraints.

In this paper we consider the problem of user assignment and power allocation in a small cell environment which is one of the most important problems in present wireless cellular network research. We consider a two-tier cellular network where randomly dispersed overlay femtocell base stations (FBSs) coexist with a macrocell. Our objective is to maximize the total number of users served by the FBSs while satisfying their signal to noise and interference (SINR) requirements. This problem is known to be NP-Hard and hence there is no known optimal solution to solve it in polynomial time. First we formulate the problem of maximization of allocated users under SINR constraints with constant transmit power as an integer programming problem. We provide two heuristic polynomial time algorithms. Then we propose a third algorithm for joint power and user allocation. We evaluate the complexity of the proposed algorithms and furthermore compare the results against the brute force optimal solution and a basic random user assignment through simulations. The results demonstrate the performance and the efficiency of the proposed algorithms. We see in the simulation that the best proposed heuristic for maximizing the number of assigned users is only 3% less than the optimal while reducing the power consumption below that of the optimal user assignment algorithm.

The emergence of new kinds of applications and technologies (e.g., data-intensive applications, server virtualization) has led to a better utilization of the network resources. However, it has also led to more bandwidth consumption and more congestion especially inside data center networks. Thus, researchers are focusing again on TCP and Active Queue Management (AQM) mechanisms in order to better control congestion and to cope with application requirements in terms of end-to-end delay [1], [2], [3]. Recently, we proposed a new AQM mechanism (called α_SNFAQM) that uses traffic prediction to accurately detect future congestion and to proactively act upon it [4]. In this paper, we develop an analytical model to assess the effect of α_SNFAQM on TCP. The study proves that this AQM is efficient enough to stabilize queue size in routers/switches, and thereby allowing to control end-to-end packet delay. These results have been also validated by simulations for a topology with multiple bottleneck links. They show that α_SNFAQM outperforms other AQM schemes like RED, PAQM and APACE in stabilizing instantaneous queue length, while keeping a high utilization of the links and the same packet loss rate.

Burst contention is a well-known challenging problem in optical burst switching (OBS) networks. Contention resolution approaches are always reactive and attempt to minimize the BLR based on local information available at the core node. On the other hand, a proactive approach that avoids burst losses before they occur is desirable. To reduce the probability of burst contention, a more robust routing algorithm than the shortest path is needed. This paper proposes a new routing mechanism for JET-based OBS networks, called graphical probabilistic routing model (GPRM) that selects less utilized links, on a hop-by-hop basis by using a Bayesian network. We assume no wavelength conversion and no buffering to be available at the core nodes of the OBS network. We simulate the proposed approach under dynamic load to demonstrate that it reduces the BLR burst loss ratio compared to static approaches by using Network Simulator 2 (ns-2) on NSFnet network topology and with realistic traffic matrix. Simulation results clearly show that the proposed approach outperforms static approaches in terms of BLR.

Optical Burst Switching (OBS) has been developed as an efficient switching technique for the next generation optical Internet. A critical issue for OBS networks is the burst loss which could occur due to contention and/or insufficient offset time. Burst Loss Ratio (BLR) is used as the main performance parameter in bufferless OBS networks. This paper proposes a new TCP statistics based method to predict the BLR without using any feedback information from the network. The idea is to estimate the BLR based on the TCP statistics available at the edge node. Our proposed BLR prediction method is then integrated into the closed loop feedback control model to control the BLR inside the network. Our simulation results clearly show that our proposed method improves the efficiency of the closed loop feedback control model while avoiding the use of any feedback information from the network.

Optical Burst Switching (OBS) has been developed as an efficient switching technique for the next generation optical Internet. A critical issue in OBS is the burst loss which could occur due to contention and/or insufficient offset time. Burst Loss Ratio (BLR) is used as the main performance parameter in OBS networks. In this paper, we investigate the assignment of the offset time and its effect on the measured end to end (E2E) delay. A novel feedback control technique is proposed to adapt the offset time based on the network condition in terms of BLR. Simulations show that the feedback control is able to adjust the offset automatically and dynamically. Hence, it reduces both the BLR due to insufficient offset and the E2E delay.

This paper proposes a novel scheme that uses feedback control approaches to support quality of service (QoS) in optical burst switching (OBS) networks. This work provides service level objectives in terms of burst loss ratio (BLR) for each class of bursts. The BLR is the ratio between the lost bursts to the sent bursts. Using feedback control approaches computes accurate burstification rate for each class of bursts. Burstification rates are computed at each burst manager controller for each class based on the previous measured value of the burst loss rate and the desired burst loss rate. Simulation results on NSFNET network have showed that the feedback control scheme guarantees a BLR to hover around the desired value for each class of bursts.

This paper considers the use of closed loop feedback control theoretic techniques to improve the performance of optical burst switching (OBS) networks. In OBS networks, the burst loss ratio (BLR) is the ratio between the lost bursts to the sent bursts. The BLR is used as a performance metric. The desired burst loss ratio depends on the application using the network. Burstification rate is the rate of injecting bursts into the OBS network. In this paper, a novel technique to control the burst loss ratio in OBS networks is proposed. The technique is based on classical control theory approaches to tune the burstification rate in order to achieve a desired burst loss ratio to satisfy the application requirements. Extensive simulations on the NSFNET topology show that the proposed technique achieves promising results. That is, the measured burst loss ratio hovers around the desired burst loss ratio for all nodes.

This paper considers the use of closed loop feedback control theoretic techniques to improve the performance of Optical Burst Switching (OBS) networks. In OBS networks, the Burst Loss Ratio (BLR) is the ratio between the lost bursts to the sent bursts. The BLR is used as a performance metric. The desired burst loss ratio depends on the application using the network. Some applications might tolerate more burst loss ratios than other applications. Higher network link utilization could be achieved by having more control over the burst loss ratio. Burstification rate is the rate of injecting bursts into the OBS network. In this paper, a novel technique to control the burst loss ratio in OBS networks is proposed. The technique is based on classical control theory approaches to tune the burstification rate in order to achieve a desired burst loss ratio to satisfy the application requirements. Extensive experiments show that the proposed technique achieves promising results. That is, the measured burst loss ratio hovers around the desired burst loss ratio and higher utilization is observed. Empirical approaches are used to identify the proposed model. The empirical model fits the OBS network by a value that did not fall below 75%.

Active Queue Management (AQM) policies are mechanisms for congestion avoidance, which pro-actively drop packets in order to provide an early congestion notification to the sources. Random Early Detection (RED), the defacto standard and its different flavors have been proposed as simple solutions to the AQM problem. However, these approaches require manual tuning and fail to accurately capture variations in the input traffic, thereby resulting in unstable behavior. α_SNFAQM is a new AQM mechanism that uses a neurofuzzy prediction method (α_SNF) to capture traffic variation and accurately detect the future congestion. It distinguishes (i) severe congestion and (ii) light congestion. We compare the performance of α_SNFAQM with other AQM schemes like RED, PAQM and APACE in a bottleneck link. Simulation results have shown that α_SNFAQM outperforms other AQM schemes in stabilizing the instantaneous queue length, reducing packet loss ratio while keeping a high utilization of the link.

Optical burst switching (OBS) is a new optical switching paradigm where traffic can be switched and groomed at a lower level compared to optical circuit switching (OCS). Although research in OBS networks has evolved from theoretical investigations to proof-of-concept demonstrations, several key issues need to be investigated further before OBS prototypes can clearly outperformed OCS networks. While the burst loss rate is often used as the main performance parameter in OBS networks, burst losses come from two sources, contention and insufficient offset time (IOT) due to deflection routing. In this paper, we investigate further the assignment of the offsets and propose an adaptive offset determination that depends on both the bandwidth utilization of the links and nodes traversed by the bursts and some measurement of the burst losses due to insufficient offset time. Numerical results demonstrate that our approach is effective in reducing the network burst drop rate through a reduction of the burst losses due to insufficient offset time. Moreover, it proves to lead to a highly stable IOT burst drop even for dynamic traffic and can be easily controlled, so as to find the equilibrium between the IOT and contention burst drops leading to the minimum burst drop rate

This paper presents a new structure-preserving method of sampling self-similar traffic with direct applications to network monitoring and resource provisioning. Predicting the bandwidth required by upcoming traffic plays a key role for providing an efficient and intelligent resource provisioning, especially in the context of IP over WDM. To achieve this, we are proposing a periodic sampling method (called maximum-based sampling) that picks one measurement during a sampling interval of size T. Mathematical analysis and simulation results demonstrate that the proposed maximum-based sampling method preserves the self-similarity property of the original traffic over many time scales. The LMMSE (linear minimum mean square error) prediction method is used for traffic forecast. The numerical results show that the accuracy of traffic prediction performed on the proposed sampled process remains stable for different sampling interval size T.

There is a growing need for e2e lightpaths for high volume data transferring applications such as GridFTP and SAN. They wish to dynamically deploy lightpaths over multiple management domains. Research, sponsored by Canarie Incorporated, is underway to enable "customer-empowered networks" and to experiment them with the Canadian research network CA*Net4. New signaling and control approaches using Web services have been proposed. One difficulty is that each domain must retain the control of their optical network infrastructure and ensure proper allocation of optical resources. Hence, it is important that the signaling takes into account the management constraints imposed by the different domains. This paper presents a policy-based approach for user-controlled lightpath provisioning. The work builds on the research around the new signaling approaches for realizing customer-empowered networks. We present an architecture based on Web Services allowing users or Grid applications to establish e2e lightpaths over multiple autonomous systems. To tackle the problem of admission control and to address the resource allocation issue, we developed policy restricted signaling which allows customers to reserve lightpaths over multiple domains while ensuring that management rules of each domain are enforced. The signaling has been implemented and the experiment on a small network composed of Cisco equipment proves the viability of our approach.