Internet of Things (IoT) is an emerging paradigm that is turning and revolutionizing worldwide cities into smart cities. However, this emergence is accompanied with several cybersecurity concerns due mainly to the data sharing and constant connectivity of IoT networks. To address this problem, multiple Intrusion Detection Systems (IDSs) have been designed as security mechanisms, which showed their efficiency in mitigating several IoT-related attacks, especially when using deep learning (DL) algorithms. Indeed, Deep Neural Networks (DNNs) significantly improve the detection rate of IoT-related intrusions. However, DL-based models are becoming more and more complex, and their decisions are hardly interpreted by users, especially companies’ executive staff and cybersecurity experts. Hence, the corresponding users cannot neither understand and trust DL models decisions, nor optimize their decisions (users) based on DL models outputs. To overcome these limits, Explainable Artificial Intelligence (XAI) is an emerging paradigm of Artificial Intelligence (AI), that provides a set of techniques to help interpreting and understanding predictions made by DL models. Thus, XAI enables to explain the decisions of DL-based IDSs to make them interpretable by cybersecurity experts In this paper, we design a new XAI-based framework to give explanations to any critical DL-based decisions for IoT-related IDSs. Our framework relies on a novel IDS for IoT networks, that we also develop by leveraging deep neural network, to detect IoT-related intrusions. In addition, our framework uses three main XAI techniques (i.e., RuleFit, Local Interpretable Model-Agnostic Explanations (LIME), and SHapley Additive exPlanations (SHAP)), on top of our DNN-based model. Our framework can provide both local and global explanations to optimize the interpretation of DL-based decisions. The local explanations target a single/particular DL output, while global explanations focus on deducing the most important features that have conducted to each made decision (e.g., intrusion detection). Thus, our proposed framework introduces more transparency and trust between the decisions made by our DL-based IDS model and cybersecurity experts. Both NSL-KDD and UNSW-NB15 datasets are used to validate the feasibility of our XAI framework. The experimental results show the efficiency of our framework to improve the interpretability of the IoT IDS against well-known IoT attacks, and help the cybersecurity experts get a better understanding of IDS decisions.

Industry 4.0 or Industrial Internet of Things (IIoT), has become one of the most talked-about industrial business concepts in recent years. Thus, to efficiently integrate Internet of Things (IoT) technology into industry, the collected and sensed data from IIoT need to be scheduled in real time constraints; especially for big factories. To this end, we propose in this paper a hierarchical fog servers’ deployment at the network service layer across different tiers. Using probabilistic analysis models, we prove the efficiency of the proposed hierarchical fog computing compared with the flat architecture. In this work, IIoT data and requests are divided into both high priority and low priority requests; the high priority requests are urgent/emergency demands that need to be scheduled rapidly. Therefore, we use two-priority queuing model in order to schedule and analyze IIoT data. Finally, we further introduce a workload assignment algorithm to offload peak loads over higher tiers of the fog hierarchy. Using realistic industrial data from Bosh group, the benefits of the proposed architecture compared to the conventional flat design are proved using various performance metrics and through extensive simulations.

5G networks are anticipated to support a plethora of innovative and promising network services. These services have heterogeneous performance requirements (e.g., high-rate traffic, low latency, and high reliability). To meet them, 5G networks are entailed to endorse flexibility that can be fulfilled through the deployment of new emerging technologies, mainly software-defined networking (SDN), network functions virtualization (NFV), and network slicing. In this paper, we focus on an interesting automotive vertical use case: autonomous vehicles. Our aim is to enhance the quality of service of autonomous driving application. To this end, we design a framework that uses the aforementioned technologies to enhance the quality of service of the autonomous driving application. The framework is made of 1) a distributed and scalable SDN core network architecture that deploys fog, edge and cloud computing technologies; 2) a network slicing function that maps autonomous driving functionalities into service slices; and 3) a network and service slicing system model that promotes a four-layer logical architecture to improve the transmission efficiency and satisfy the low latency constraint. In addition, we present a theoretical analysis of the propagation delay and the handling latency based on GI/M/1 queuing system. Simulation results show that our framework meets the low-latency requirement of the autonomous driving application as it incurs low propagation delay and handling latency for autonomous driving traffic compared to best-effort traffic.

Smart grids (SG) energy management system and electric vehicle (EV) have gained considerable reputation in recent years. This has been enabled by the high growth of EVs on roads; however, this may lead to a significant impact on the power grids. In order to keep EVs far from causing peaks in power demand and to manage building energy during the day, it is important to perform an intelligent scheduling for EVs charging and discharging service and buildings areas by including different metrics, such as real-time price and demand-supply curve. In this paper, we propose a real-time dynamic pricing model for EVs charging and discharging service and building energy management, in order to reduce the peak loads. Our proposed approach uses a decentralized cloud computing architecture based on software define networking (SDN) technology and network function virtualization (NFV). We aim to schedule user's requests in a real-time way and to supervise communications between microgrids controllers, SG and user entities (i.e., EVs, electric vehicles public supply stations, advance metering infrastructure, smart meters, etc.). We formulate the problem as a linear optimization problem for EV and a global optimization problem for all microgrids. We solve the problems by using different decentralized decision algorithms. To the best of our knowledge, this is the first paper that proposes a pricing model based on decentralized Cloud-SDN architecture in order to solve all the aforementioned issues. The extensive simulations and comparisons with related works proved that our proposed pricing model optimizes the energy load during peak hours, maximizes EVs utility, and maintains the microgrid stability. The simulation is based on real electric load of the city of Toronto.

Distributed denial-of-service (DDoS) attacks continue to grow at a rapid rate plaguing Internet Service Providers (ISPs) and individuals in a stealthy way. Thus, intrusion detection systems (IDSs) must evolve to cope with these increasingly sophisticated and challenging security threats. Traditional IDSs are prone to zero-day attacks since they are usually signature-based detection systems. The recent advent of machine learning and deep learning (ML/DL) techniques can help strengthen these IDSs. However, the lack of up-to-date labeled training datasets makes these ML/DL based IDSs inefficient. The privacy nature of these datasets and widespread emergence of adversarial attacks make it difficult for major organizations to share their sensitive data. Federated Learning (FL) is gaining momentum from both academia and industry as a new sub-field of ML that aims to train a global statistical model across multiple distributed users, referred to as collaborators, without sharing their private data. Due to its privacy-preserving nature, FL has the potential to enable privacy-aware learning between a large number of collaborators. This paper presents a novel framework, called MiTFed, that allows multiple software defined networks (SDN) domains ( $i.e.,$ collaborators) to collaboratively build a global intrusion detection model without sharing their sensitive datasets. In particular, MiTFed consists of: (1) a novel distributed architecture that allows multiple SDN based domains to securely collaborate in order to cope with sophisticated security threats while preserving the privacy of each SDN domain; (2) a novel Secure Multiparty Computation (SMPC) scheme to securely aggregate local model updates; and (3) a blockchain based scheme that uses Ethereum smart contracts to maintain the collaboration in a fully decentralized, trustworthy, flexible, and efficient manner. To the best of our knowledge, MiTFed is the first framework that leverages FL, blockchain and SDN technologies to mitigate the new emerging security threats in large scale. To evaluate MiTFed, we conduct several experiments using real-world network attacks; the experimental results using the well-known public network security dataset NSL-KDD show that MiTFed achieves efficiency and high accuracy in detecting the new emerging security threats in both binary and multi-class classification while preserving the privacy of collaborators, making it a promising framework to cope with the new emerging security threats in SDN.

Industry 5.0 is rapidly growing as the next industrial evolution, aiming to improve production efficiency in the 21 ^st century. This evolution relies mainly on advanced digital technologies, including Industrial Internet of Things (IIoT), by deploying multiple IIoT devices within industrial systems. Such a setup increases the possibility of threats, especially with the emergence of IIoT botnets. This can provide attackers with more sophisticated tools to conduct devastating IIoT attacks. Besides, machine learning (ML) and deep learning (DL) are considered as powerful techniques to efficiently detect IIoT attacks. However, the centralized way in building learning models and the lack of up-to-date datasets that contain the main attacks are still ongoing challenges. In this context, multiaccess edge computing (MEC) and federated learning (FL) are two promising complementary technologies. MEC brings computing capabilities at the edge of the industrial systems, while FL leverages the edge resources to enable a privacy-aware collaborative learning, especially in multiindustrial systems context. In this article, we design a novel MEC-based framework to secure IIoT applications leveraging FL, called FedGame. Specifically, FedGame enables multiple MEC domains to collaborate securely to deal with an IIoT attack, while preserving the privacy of IIoT devices. Moreover, a noncooperative game is formulated on the top of FedGame, to enable MEC nodes acquiring the needed virtual resources from the centralized MEC orchestrator, to deal with each type of IIoT attacks. We evaluate FedGame using real-world IIoT attacks; the experimental results show not only the accuracy of FedGame against centralized ML/DL schemes while preserving the privacy of Industrial systems but also its efficiency in providing required MECs resources and, thus, dealing with IIoT attacks.