The evolution of Distributed Denial-of-Service (DDoS) attack techniques on the Internet of Things (IoT) domain presents ongoing challenges as attackers increasingly emulate legitimate traffic patterns. This necessitates the continual adaptation of deep learning-based anomaly detection systems. Furthermore, the high cost of recurrently retraining deep learning models from scratch highlights the demand for adaptive detection approaches that can respond effectively to shifting threats in IoT environments. This paper investigates a range of Federated Continual Learning (FCL) techniques for identifying DDoS attacks within IoT systems, utilizing diverse federated learning approaches such as Prioritized Experience Replay (PER), Learning Without Forgetting (LWF), and Clustered Federated Learning (CFL). Continuous learning techniques, including Elastic Weight Consolidation (EWC), Agnostic Model Update (AMU), and Federated Proximal (FedProx), are also applied. The effectiveness of these methods is assessed across configurations with 16, 32, and 64 clients. Results indicate that LWF performed optimally in smaller client configurations, especially with FedAvg and FedProx, while EWC was more effective in larger setups. FedAvg and FedProx were consistently reliable strategies, whereas AMU and CFL demonstrated variable performance. This study highlights the critical role of advanced machine learning techniques in enabling real-time DDoS detection for IoT applications.

In this paper, we propose a federated continual learning-based IoT system for real-time monitoring of surface water quality, named FCL-IWQMS. This system improves surface water monitoring management by integrating sensor networks and predictive analytics, addressing the challenges of climate change and urbanization. FCL-IWQMS enables collaboration by allowing Internet of Things (IoT) devices to send only updates from their local models to a central server, consolidating them to generate an improved prediction model. This approach ensures data privacy while enhancing the accuracy of water quality predictions. The framework utilizes methods such as Prioritized Experience Replay (PER), Learning Without Forgetting (LWF), and Elastic Weight Consolidation (EWC) to refine local models in response to new data. Local updates are periodically sent to aggregation nodes, where techniques like Federated Averaging (FedAvg), Federated Trimmed Mean (FedTM), and Agnostic Model Update (AMU) are applied to consolidate updates. Evaluations with 40 clients using publicly available datasets show that the FedAvg-PER model outperforms others in predicting dissolved oxygen (DO). At the same time, AMU-LWF excels in pH predictions, and FedTM-PER leads in electrical conductivity (EC).

In many African countries, mobile payments are vital for financial transactions. However, enabling peer-to-peer payments is challenging due to limited network infrastructure and resource-constrained devices. Companies like Ejara are working to bring accessible financial services to underserved communities. However, existing routing protocols for Opportunistic Networks (OppNets), like traditional epidemic routing, create high routing overhead and latency from indiscriminate packet flooding, putting significant pressure on network and device resources. This study addresses these issues by introducing an optimized epidemic routing protocol with Reinforcement Learning algorithms integration to enhance packet forwarding in OppNets. The protocol dynamically adapts forwarding decisions based on delivery probability, latency, and resource constraints. Emulation results demonstrate that this approach significantly reduces routing overhead and latency while improving packet delivery reliability. This solution has meaningful implications for enabling efficient mobile payments and peer-to-peer interactions, especially in resource-constrained environments with intermittent connectivity, supporting broader accessibility to financial services.

This paper addresses Quality of Experience (QoE)-driven, self-optimizing storage for distributed file sharing—a field gaining increasing attention in cloud and edge systems research. We present a novel platform for secure file sharing, centered on QoE-driven optimization of the Zettabyte File System (ZFS). The proposed four-module architecture integrates ZFS with reinforcement learning (RL) to dynamically tune QoE metrics such as latency, throughput, and caching efficiency, adapting to evolving workloads and user expectations. By leveraging RL, the system continuously optimizes ZFS configurations for enhanced performance. The four-layer architecture provides a coherent end-to-end framework that links user-level QoE signals to low-level ZFS tunables, while incorporating blockchain-based traceability to ensure transparency and trust. Experimental evaluations demonstrate that the adaptive deep Q-learning strategy improves storage performance and QoE compared to static configurations, establishing a new benchmark for QoE-driven decentralized storage.

Web applications have become integral to daily life due to the migration of applications and data to cloud-based platforms, increasing their vulnerability to attacks. This paper addresses the need for robust intrusion detection systems by proposing a system grounded in Zero Trust architecture, which mandates continuous monitoring and multi-layered defenses. The Zero Trust principles ensure ongoing threat assessment and comprehensive protection against various attack vectors. Building on these foundational Zero Trust principles, our study introduces a system designed to not only distinguish normal HTTP requests from well-known attack patterns but also detect emerging types of anomalous attacks. Our system consists of two models that integrate Natural Language Processing approaches, Deep Learning techniques, and Transfer Learning strategies. The first model is employed to detect new anomalous HTTP requests that differ from normal requests. HTTP requests identified as anomalous are transmitted to the second model in charge of classifying specific categories of both well-known and novel attacks. Experiments show that our end-to-end system achieves the average F1-score of 89% on the combination of the CAPEC dataset and the zero-shot CSIC dataset. The proposed system proves also to be able to identify anomalous requests with a minimal latency of 4.8 milliseconds in production settings.

In the realm of IoT platforms, susceptibility to cyber-attacks is a pressing concern, necessitating the deployment of Intrusion Detection Systems (IDS). Constructing a scalable, accurate, and lightweight model without compromising data privacy poses a formidable challenge. This study assesses classical and novel approaches employing federated learning (FL) to train IDS models. Optimization through Knowledge Distillation (KD) techniques aims to enhance computational efficiency. Experimental results reveal the efficacy of federated learning, achieving an 84.5% accuracy for 15 attack types, and an impressive performance for binary network attack classification. Notably, these models exhibit shorter inference times compared to cutting-edge machine learning models trained on the Edge-IIoTset dataset, offering promising advancements in IoT security.

In the past decades, the use of urban rivers for recreational and sporting activities has gained increasing interest. However, bathing in urban surface waters is not without health risks due to short-term pollution of fecal origin, which may have an important impact on the overall population health within a region where bathing in water streams is possible. Therefore, EU member states are required to lower the contamination risk of such areas through active water quality management, as defined by the Bathing Water directory (BWD, 2006/7/EC). This paper develops and evaluates a cost-effective IoT-based water quality monitoring system, based on low-cost water quality sensors coupled with machine-learning approaches. By monitoring spatiotemporal dynamics of several physical and chemical parameters correlated with bacterial indicators, managers can more easily decide if the water quality of a bathing site is enough for usage. To determine the water suitability at particular river sites, the system employs a convolutional neural network (CNN) deep learning classifier, integrating federated learning (FL) with knowledge distillation (FedKD) to streamline model architecture, reduce communication costs, and preserve data privacy. The system is tested on the Seine and the Marne rivers (Paris area, France) and results demonstrate that FedKD outperforms centralized knowledge distillation (KD) and FL algorithms such as FedAvg and UFedAVG. Using the current features, it achieves a satisfactory average accuracy of 90.74% at Marne station.

Techniques for Deep Learning (DL) and Natural Language Processing (NLP) are rapidly advancing. In addition, we notice that the access and utilisation of web applications is expanding in almost all fields in conjunction with related technologies. Web applications include a wide range of use cases involving personal, financial, military, and political data. This renders web-based applications a desirable target for cyber-attacks. To address this problem, we propose, in this study, a novel model capable of differentiating normal HTTP requests from different types of anomalous HTTP requests. Our model combines NLP techniques, the Bidirectional Encoder Representations from Transformers (BERT) model, and DL techniques. The pre-trained BERT model is able to operate on unprocessed data and therefore does not require manually extracted features. Our experimental results show that the proposed method achieves an F1 score of more than 98.90% in the classification of multiple categories of anomalous requests and normal requests on CAPEC dataset. Furthermore, we leverage Transfer Learning in order to detect new types of anomalous requests or new attack patterns that are similar to training anomalous patterns. With Transfer Learning techniques, our proposed model achieves an F1-score of 61.50% on unseen types of anomalous HTTP requests.

Fifth-generation wireless technology, or 5G, promises increased data speeds, lower latency and greater capacity for mobile communications, enabling the growth of the Internet of Things (IoT). Network slicing is an advantageous feature of 5G that enables the creation of multiple virtual networks to meet the performance, security, and reliability needs of different services. This feature, combined with blockchain technology, provides secure and transparent data sharing between devices and networks. In addition, blockchain-enabled network slicing improves 5G network management and creates new business models for industries such as healthcare, transportation, and manufacturing. However, most current blockchain systems are limited in their ability to handle high throughput, which is not equivalent to what 5G can do. Therefore, sharding-based blockchain is proposed as a possible solution to improve the scalability and throughput of blockchain networks. By dividing the network into parts or shards, transactions can be processed simultaneously, allowing for faster transaction verification times and increased network capacity. Although current sharding-based blockchain networks have some limitations, such as static sharding policies that cannot cope with the dynamic blockchain environment, an adaptive sharding blockchain system using Deep Reinforcement Learning (DRL) methods has been proposed to address this situation. The approach allows the system to change or adapt the shard specifications, such as shard size or block size, whenever necessary to ensure maximum throughput while maintaining the security and integrity of the blockchain network.

In the past, Quality of Service (QoS) was taken into account to evaluate the performance of multimedia services (e.g., video streaming, file transfer, etc.). However, it cannot reflect the user's perception, which is considered a crucial consideration by these services nowadays. Therefore, the emergence of Quality of Experience (QoE) is a potential solution. QoE can be measured via many parameters provided by Internet Service Providers (ISP), Application Service Providers (ASP), or end-users. However, privacy concerns hinder data sharing between the parties involved. To address these limitations, this paper proposes a QoE estimation mechanism that leverages Federated Learning. This mechanism aims to guarantee data privacy when no party needs to disclose their data to others. Moreover, the proposed mechanism incorporates the concept of a Decentralized Autonomous Organization (DAO) to mitigate the risk of a single point of failure in the centralized architecture of Federated Learning. It enables all participants to evaluate and select the model efficiently. The experimental results illustrate that the proposal surpasses the centralized solutions and guarantees data privacy.

The monitoring of surface water in smart cities can be enhanced with the Internet of Things (IoT) and the use of transfer learning. The former allows the increase in the coverage area and to better exploit this data. The latter has the potential to reduce the need of data collection, which may be costly. In this work, we discuss the potential use of these two domains for the estimation of surface water quality in the Marne River (France). The assessment is made using physico-chemical data from sensors to predict the concentration of fecal indicator bacteria. The results show that the use of transfer learning has the potential to enhance water quality monitoring in smart cities.

Nowadays, blockchain, revolutionary technology with high security and decentralization is widely applied in many industry segments including healthcare, cryptocurrency and so on. Despite its benefit, it is a challenge for blockchain to effectively operate in large-scale networks without compromising security and decentralization properties due to low throughput. Therefore, in this paper, we consider sharding-based blockchain as a solution to overcome this limitation. The objective is to split the network into different shards to process transactions in parallel and reduce quorum size to reach consensus quickly. However, existing sharding-based blockchain networks implement a consensus algorithm for both main and sub-shards, so it is ineffective with different purposes (e.g., to improve security, reduce resource consumption, etc.). Hence, we propose a system with heterogeneous consensus algorithms in sharding-based blockchain to meet different requirements. Experimental results show that the proposed approach improves up to 54 percent of throughout in comparison with benchmarks.

Software-Defined Networking (SDN) architecture was developed to address the shortcomings of traditional network architectures. It allows system administrators to easily manage and configure the network by separating and abstracting the control plane from the data plane. All the knowledge and intelligence of SDN is concentrated in a software entity called the SDN controller, making the network programmable. However, a large-scale SDN architecture, particularly in the IoT domain, requires the implementation of a physically distributed control mechanism. Such a mechanism, based on the East/West interface raises many challenges in terms of scalability, reliability, security, consistency, and traceability. The development of the Blockchain allows addressing some of these challenges. In this paper, we present a design using Blockchain technology to improve SDNs in terms of trackability and discuss the adaptations required for large-scale deployment. Experimental results clearly show that the use of a proof-of-authority consensus algorithm in combination with a Merkle tree approach reduces the impact in terms of latency as well as in terms of Gas consumption.

Software-Defined Networking (SDN) and its extension Intent-Based Networking (IBN) are network paradigms that enable dynamic, programmatically efficient network configuration. IBN allows network operators to express an outcome or business objective without the low-level configurations necessary to program the network to achieve these demands. Existing research proposals for IBN introduce several systems to translate users intents into network infrastructure configurations. Despite the positive aspects of these proposals, they still suffer from many drawbacks. Some require users to learn a new intent definition language. Some others may lack the appropriate grammar to make these frameworks recognize the intent correctly. In this paper, we introduce a framework leveraging the capabilities of Natural Language Processing (NLP) for network management from an operator utterances. In order to understand natural language, our framework uses the sequence-to-sequence (seq2seq) learning model based on recurrent neural networks (LSTM). The model has been improved by using word embedding and user feedback. As a proof of concept, we implement our framework for network management in a Global Internet eXchange Network and evaluate its practicality regarding NLP accuracy and network performance.

Nowadays, the rapid growth of the Internet makes network management more complex due to various and com-plicated network problems. In the past, network administrators implemented troubleshooting approaches (e.g., ping, traceroute, etc.) manually to identify the root cause of problems. However, it is not effective due to human intervention and an increase of network devices. Consequently, the root cause analysis is considered by the research community. There are existing studies for the root cause analysis without human intervention (e.g., statistical approaches, heuristic algorithms, etc.). However, these approaches show limited performance (e.g., due to complex threshold identifcation, etc.). The emerging of machine learning (ML) and deep learning is a potential solution to overcome this obstacle, offering an opportunity to develop an effective root cause analysis approach. Therefore, in this paper, we propose a root cause analysis approach using ML and time-series network parameters to identify the root cause of problems in the network. In this approach, we consider balancing the accuracy and the time complexity of ML algorithms to select an appropriate ML technique. Moreover, we contribute troubleshooting datasets to identify three kinds of root causes including link failure, switch failure and buffer overload. The experimental results show that the proposal can achieve approximately 97 percent of precision, recall and f1-score in considered scenarios and require less processing time (only require 0.00143 ms for a sample) in comparison with other ML algorithms.

An ingress router often has multiple potential egress points in an extensive network where it can transmit traffic to external networks. The traditional solution is choosing the closest node (with the shortest path) to the ingress node. This paper claims the drawbacks of this approach in a flexible network system and introduces our proposal called MAB-based Egress Selection. Our approach uses several Reinforcement Learning techniques, which are commonly used to resolve Multi-Armed Bandit (MAB) problem, to allow the ingress router to periodically re-pick egress point, hence optimize the long-term performance of traffic transmission. To formalize the egress selection process as a MAB problem, we use a combined score of delay and loss representing link status as a reward. However, capturing those network metrics encounters some issues due to the distributed control and restricted local view of network nodes. For this purpose, a centralized control architecture, e.g., Software-defined Network (SDN), is a promising candidate. We applied four common algorithms, ε-greedy, Softmax, UCB1 and Single Pull UCB2 (SP-UCB2) for egress selection process. The models are evaluated in two simulated network topologies with different scenarios of network traffic condition. The experimental results show that the UCB algorithms produce the best performance, especially in busy network.

Traffic Matrix (TM) clearly describes the volume and the distribution of traffic flows inside a network. TM plays an important role in many network management fields, such as traffic accounting, short-time traffic scheduling or re-routing, network design, anomaly detection, etc. Hence, an accurate TM prediction strategy is essential to handle those tasks effectively. Fortunately, Artificial Intelligence (AI) has been developing very strongly, thanks to computer technology developments such as GPU and TPU. That offers an opportunity to apply AI to TM prediction methods. However, applying Machine Learning techniques in traditional networks encounters some issues due to the distributed control and restricted local view of network nodes. For this purpose, a centralized control architecture, e.g., Software-defined Network (SDN), is a promising candidate. In this paper, we apply Long Short-Term Memory (LSTM) and its two variants, Bidirectional LSTM (BiLSTM) and Gate Recurrent Unit (GRU), for TM prediction mechanisms of an SDN architecture network. The prediction models have been evaluated using two datasets: the popular GÉANT backbone network traffic data and our dataset generated through a testbed. The experimental results show that our approach yielded promising traffic prediction accuracy.

For the past decade, IP (Internet Protocol) routing approaches utilize TCAM (Ternary Content Addressable Memory) for the rule matching in the switches. These approaches are expensive and require more power consumption. Fortunately, the emerging of segment routing can resolve this drawback by encoding a routing path into the packet header to forward the packets to a destination. However, the standard segment routing algorithm has encountered a main problem. Using the shortest path to forward the packets can lead to a high traffic load on these paths and a performance reduction. It results in a decrease in user's perception and some negative economic impacts for ISPs (Internet Service Providers). Therefore, in this paper, we propose a novel service-centric segment routing mechanism using reinforcement learning in the context of encrypted traffic. Our proposal aims to help ISPs to decrease the influence of the network problems and meet the strict user's requirement related to QoE (Quality of Experience). The obtained results under the considered conditions demonstrate that our approach out-performs the standard segment routing algorithm and requires reasonable computational cost.

This paper addresses the knowledge dissemination problem in distributed SDN control by proposing an adaptive and continuous consistency model for the distributed SDN controllers in large-scale deployments. We put forward a scalable and intelligent replication strategy following Quorum-replicated consistency: It uses the read and write Quorum parameters as adjustable control knobs for a fine-grained consistency level tuning. The main purpose is to find, at runtime, appropriate partial Quorum configurations that achieve, under changing network and workload conditions, balanced trade-offs between the application's continuous performance and consistency requirements. Our approach was implemented for a CDN-like application that we designed on top of the ONOS controllers. When compared to ONOS's static consistency model, our model proved efficient in minimizing the application's inter-controller overhead while satisfying the SLA-style application requirements.

In the coming years, the development of the Internet of Things (IoT) will have relevance for transport, environment, health care, smart cities and also multimedia services (Multimedia Internet of Things (MIoT)). Nowadays, many ISP (Internet Service Provider) encrypt the data to make it secure during the transmission. However, it imposes some obstacles for the NSP (Network Service Provider) because of the lack of visibility for operators into network traffic. To resolve these issues, we proposed the Quality Estimation Framework for Encrypted Traffic (Q2ET) containing a classification module and a QoE assessment module. The first module inherited from our previous research works to classify the encrypted network traffic using CNN (Convolutional Neural Network). The second one applies the objective and subjective methods based on the statistical analysis and machine learning methods that combine application and network parameters to calculate user's QoE (Quality of Experience) in terms of MOS (Mean Opinion Score). The Q2ET allows the NSP to monitor the user's QoE to take the appropriate decisions when the QoE degradation happens in the network systems.

Logically-centralized but physically-distributed SDN controllers are mainly used in large-scale SDN networks for scalability, performance and reliability reasons. These controllers host various applications that have different requirements in terms of performance, availability and consistency. Current SDN controller platform designs employ conventional strong consistency models so that the SDN applications running on top of the distributed controllers can benefit from strong consistency guarantees for network state updates. However, in large-scale deployments, ensuring strong consistency is usually achieved at the cost of generating performance overheads and limiting system availability. That makes weaker optimistic consistency models such as the eventual consistency model more attractive for SDN controller platform applications with high-availability and scalability requirements. In this paper, we argue that the use of the standard eventual consistency models, though a necessity for efficient scalability in modern SDN systems, provides no bounds on the state inconsistencies tolerated by the SDN applications. To remedy that, we propose an adaptive consistency model for the distributed ONOS controllers following the notion of continuous and compulsory (per-controller) eventual consistency, where network application states adapt their eventual consistency level dynamically at runtime based on the observed state inconsistencies under changing network conditions. When compared to the ONOS approach to static eventual consistency, our approach proved efficient in minimizing state synchronization overheads while taking into account application state consistency SLAs and without compromising the application requirements of high-availability, in the context of large-scale SDN networks.

Nowadays, network traffic classification plays an important role in many fields including network management, intrusion detection system, malware detection system, etc. Most of the previous research works concentrate on features extracted in the non-encrypted network traffic. However, these features are not compatible with all kind of traffic characterization. Google's QUIC protocol (Quick UDP Internet Connection protocol) is implemented in many services of Google. Nevertheless, the emergence of this protocol imposes many obstacles for traffic classification due to the reduction of visibility for operators into network traffic, so the port and payload- based traditional methods cannot be applied to identify the QUIC- based services. To address this issue, we proposed a novel technique for traffic classification based on the convolutional neural network which combines the feature extraction and classification phase into one system. The proposed method uses the flow and packet-based features to improve the performance. In comparison with current methods, the proposed method can detect some kind of QUIC-based services such as Google Hangout Chat, Google Hangout Voice Call, YouTube, File transfer and Google play music. Besides, the proposed method can achieve the microaveraging F1-score of 99.24 percent.

Quick UDP Internet Connections (QUIC) is a new transport protocol developed by Google in 2012. QUIC is considered as a combination of TCP, TLS and HTTP on the top of UDP with some advantages such as reducing connection establishment time, improving congestion control, multiplexing without heads of line blocking and connection migration. In video streaming, Dynamic Adaptive Streaming over HTTP (DASH) is tied with TCP in many years, but the video streaming using HTTP on TCP has some disadvantages in terms of head of line blocking, connection migration, etc. The emergence of QUIC resolves these drawbacks and provides some solutions to reduce the latency and improve the quality of network service with respect to QoE. Therefore, in this paper, we investigate and evaluate the performance of QUIC and traditional transport protocols in the context of video streaming using DASH services. Some QUIC parameters such as maximum congestion window, buffer size and number of emulated connections are considered to choose the appropriate parameters for video streaming. Besides, we compare the performance of QUIC with TCP in terms of some network parameters and some DASH parameters. The experimental results showed that the performance of QUIC with 2 emulated connections is not as good as TCP. When the number of emulated connections is set to 6, the number of changes in quality level and stalling events are lower than the figure for TCP. Consequently, the quality level of QUIC with 6 emulated connections is better than TCP. Moreover, the QoE score of QUIC with 6 emulated connections is higher than the figure for QUIC with 2 emulated connections and TCP.

The decoupling of control and data planes in Software-Defined Networking (SDN) brings benefits in terms of logically centralized control and application programming. But, the single point of management in physically centralized SDN architectures is a potential point of failure and a bottleneck that compromises network reliability and performance. Such centralized designs may also face scalability challenges especially in networks with a large number of hosts (e.g. IoT-like networks). To avoid such concerns, SDN control architectures are usually designed as physically distributed systems. This raises practical challenges about the best approach to decentralizing the control plane while maintaining the logically centralized network view. In particular, determining the number of controllers and locating them in the network is a hard task that should be addressed appropriately. This paper proposes two novel strategies that cover different aspects of the controller placement problem with respect to performance and reliability criteria. These strategies use two types of heuristics that are compared and assessed on large-scale topologies to provide operators with guidelines on how to find their optimal controller placement that meets their specific needs.

Video streaming has become a main contributor in an ever increasing Internet traffic, and meets the users expectation is a challenging task for both the Network service Provider (NsP) and Content service Provider (CsP). In this context, a new metric called: Quality of Experience (QoE) is evolved to measure the user satisfaction using video service, and it becomes a key driver for achieving the business goal of NsP and CsP. In this perspective, we have proposed a novel framework that considers the user QoE to adapt the video quality, named Optimized Quality of DASH (OQD). The objective of the proposed OQD framework is to optimize users experience, and maximize the bandwidth usage. A Machine Learning (ML) approach based on GRadient Boosting (GRB) method is implemented to predict the user QoE that considers three important network and application QoE Influence Factors (QoE IFs). We use the Reinforcement Learning (RL) approach to select the optimal video quality segment, which improves the user QoE. The performance of the proposed method is evaluated and compared against Greedy adaptive bit-rate method in terms of re-buffering, bandwidth utilization, average MOS, and standard deviation MOS. The results clearly show that proposed method performs well, as it considers the user’s perceived video quality as a regulator to optimize the overall video delivery network.

Spectrum sensing security in cooperative cognitive radio networks with continuously mobile secondary users becomes a critical challenge. Thus, we propose a trust game-based model to ensure the spectrum detection while the mobility of the SUs is taken into account. Our proposal ensures both the attacks detection and the punishment of mobile malicious users launching the Spectrum Sensing Data Falsification (SSDF) attacks. Extensive simulations prove that the proposed model outperforms the AND-rule, OR-rule and Game-Based Secure Sensing (GSS) models in terms of correct decision probability, throughput and error probability with the random and linear mobility models and under four types of SSDF attacks.

The user satisfaction measurement has gained high attention from Network Operators (NOs) and Service Providers (SPs) because their businesses are highly dependent on the user's satisfaction. Generally, the traditional strategies to measure the user's perception are based on Quality of Service (QoS), which is not sufficient to reflect the real user's perceived quality. Therefore, NOs and SPs start to develop new strategies based on the Quality of Experience (QoE) metric to analyze the relationship between the user's satisfaction and influence factors (QoE IFs). In this paper, a new method to build a predictive model to estimate user's satisfaction in terms of Mean Opinion Score (MOS) is proposed. The proposed method uses the dataset collected using the controlled testbed based on the YouTube video service. In the proposed model, the correlation matrix is used to develop a new heuristic method that used back-jumping technique to select the most beneficial factors to predict the optimal user's satisfaction.

In Heterogeneous Wireless Networks, mobile users use a terminal with multiple access interfaces for non-real-time or real-time applications (services). In such environment, the major issue is Always Best Connected (ABC), which means that the mobile nodes rank the network interfaces and select the best one at anytime and anywhere. To meet the ABC requirements, many network interface selection strategies have been proposed in the literature, using various technologies. This paper surveys existing approaches and discusses their advantages and limitations. The paper also highlights open issues in this area of research and proposes a new QoE-based approach for interface selection based on TOPSIS algorithm for e-Health use case. The effectiveness of our approach is evaluated through simulations. Obtained results show clearly that our approach ensures the best QoE for user, and eliminates a major inconvenient due to rank reversal (ranking abnormality).

Traditionally, companies host their own services, platforms and infrastructures on their own servers. This policy results in high costs in terms of material and human resources. It may also be inadequate to the real needs of the company. In this context, one solution is to use cloud computing to outsource their services. The latter is defined by making available to the customer high-performance servers and high bandwidth. The cloud is also defined by renting software and hardware infrastructure to customers according to their needs. Cloud computing is made possible by the improvement of computer networks infrastructures. Indeed, broadband connections have reduced latency and thus enabled the use of remote resources. The success of cloud computing has led to a significant increase in the providers number offering many and varied cloud services. While the access to these services is made possible through a simple subscription, no technique is currently available to select the cloud provider that best fits their needs. Selecting a provider is an optimization problem that has been studied in several areas. Given the large number of parameters and actors in the cloud, this problem is known as NP-complete one. In this work, we propose a new developed platform which plays the role of a broker between clients and cloud providers. Based on a set of benchmark tasks on provider services, it performs an adaptive cloud provider selection in accordance with the client needs. The experimental results show that the proposed approach gives benefits to subscribers in terms of QoE.

The tremendous growth in video services, specially in the context of mobile usage, creates new challenges for network service providers: How to enhance the user's Quality of Experience (QoE) in dynamic wireless networks (UMTS, HSPA, LTE/LTE-A). The network operators use different methods to predict the user's QoE. Generally to predict the user's QoE, methods are based on collecting subjective QoE scores given by users. Basically, these approaches need a large dataset to predict a good perceived quality of the service. In this paper, we setup an experimental test based on crowdsourcing approach and we build a large dataset in order to predict the user's QoE in mobile environment in term of Mean Opinion Score (MOS). The main objective of this study is to measure the individual/global impact of QoE Influence Factors (QoE IFs) in a real environment. Based on the collective dataset, we perform 5 testing scenarios to compare 2 estimation methods (SVM and ANFIS) to study the impact of the number of the considered parameters on the estimation. It became clear that using more parameters without any weighing mechanisms can produce bad results.

A main objective of an Information Centric Network (ICN) is to improve the network by placing the knowledge in center of the network design. This vision of the network needs an efficient distributed and decentralized knowledge plane. So, an important amount of knowledge should be disseminated over the supervised network, which remains an open problem. Indeed, the dissemination infrastructure must be able to ensure the transport of all information types including knowledge information, throughout the network, and guarantee its freshness. Another crucial aspect of the problem is related to the network robustness with respect to network failures. In this paper, we propose a new model of knowledge dissemination based on super peers architecture. We formalize the super peer selection problem as a K-medoids clustering task. Furthermore, to handle the dynamicity of the network and especially the changes of the end-user network topology, we improved the selection mechanism by adding an adaptive mechanism based on the Page-Hinkley statistical test. Experimental results show that the proposed approach significantly improves performances compared to other current approaches.

Autonomic computing is a new paradigm inspired by the biological world. It aims at making a network independent of any human monitoring. To reach such autonomy, knowledge should be disseminated over the network, which remains an open problem. Our solution consists in proposing a new model of knowledge dissemination based on three key ideas: a hierarchical architecture, a specific-service overlay network (SSON) and a multi-criteria selection of a subset of nodes responsible for knowledge management. The simulation results show that the proposed approach significantly improves performances compared to other approaches.

Autonomic Networks represent a concept inspired by the biological world that aims at making a network independent of any human monitoring. To reach such autonomy, knowledge should be disseminated over the network, which remains an open problem. In fact, disseminating the knowledge over all nodes leads to a big overhead. That is why we need to select a subset of nodes in charge of knowledge management. A single criterion-based selection mechanism has been proposed in a previous work but such a mechanism seems to be very simplistic. In this paper, we present a new multi-criteria selection mechanism based on Pareto. The simulation results show that the proposed approach significantly improves performances compared to single criterion-based selection mechanism.

International audience