Search

Federated Edge Learning (FEEL) is a promising distributed learning technique that aims to train a shared global model while reducing communication costs and promoting users' privacy. However, the training process might significantly occupy a long time due to the nature of the used data for training, which leads to higher energy consumption and therefore impacts the model convergence. To tackle this issue, we propose a data-driven federated edge learning scheme that tends to select suitable participating nodes based on quality data and energy. First, we design an unsupervised data-aware splitting scheme that partitions the node's local data into diverse samples used for training. We incorporate a similarity index to select quality data that enhances the training performance. Then, we propose a heuristic participating nodes selection scheme to minimize the communication and computation energy consumption, as well as the amount of communication rounds. The obtained results show that the proposed scheme substantially outperforms the vanilla FEEL in terms of energy consumption and the number of communication rounds., Comment: This paper has been accepted for publication by IEEE International Conference on Communications (ICC). The final version will be published by the IEEE

The evolution of the future beyond-5G/6G networks towards a service-aware network is based on network slicing technology. With network slicing, communication service providers seek to meet all the requirements imposed by the verticals, including ultra-reliable low-latency communication (URLLC) services. In addition, the open radio access network (O-RAN) architecture paves the way for flexible sharing of network resources by introducing more programmability into the RAN. RAN slicing is an essential part of end-to-end network slicing since it ensures efficient sharing of communication and computation resources. However, due to the stringent requirements of URLLC services and the dynamics of the RAN environment, RAN slicing is challenging. In this article, we propose a two-level RAN slicing approach based on the O-RAN architecture to allocate the communication and computation RAN resources among URLLC end-devices. For each RAN slicing level, we model the resource slicing problem as a single-agent Markov decision process and design a deep reinforcement learning algorithm to solve it. Simulation results demonstrate the efficiency of the proposed approach in meeting the desired quality of service requirements.

With the advent of the Internet of Things (IoT) and 5G networks, edge computing is offering new opportunities for business model and use cases innovations. Service providers can now virtualize the cloud beyond the data center to meet the latency, data sovereignty, reliability, and interoperability requirements. Yet, many new applications (e.g., augmented reality, virtual reality, artificial intelligence) are computation-intensive and delay-sensitivity. These applications are invoked heavily with similar inputs that could lead to the same output. Compute-less networks aim to implement a network with a minimum amount of computation and communication. This can be realized by offloading prevalent services to the edge and thus minimizing communication in the core network and eliminating redundant computations using the computation reuse concept. In this paper, we present matching-based services offloading schemes for compute-less IoT networks. We adopt the matching theory to match service offloading to the appropriate edge server(s). Specifically, we design, WHISTLE, a vertical many-to-many offloading scheme that aims to offload the most invoked and highly reusable services to the appropriate edge servers. We further extend WHISTLE to provide horizontal one-to-many computation reuse sharing among edge servers which leads to bouncing less computation back to the cloud. We evaluate the efficiency and effectiveness of WHISTLE with a real-world dataset. The obtained findings show that WHISTLE is able to accelerate the tasks completion time by 20%, reduce the computation up to 77%, and decrease the communication up to 71%. Theoretical analyses also prove the stability of the designed schemes.

The tremendous advancements in the Internet of Things (IoT) increasingly involve computationally intensive services. These services often require more computation resources than can entirely be satisfied on local IoT devices. Cloud computing is traditionally used to provide unlimited computation resources at distant servers. However, such remote computation may not address the short-delay constraints that many of today's IoT applications require. Edge computing allows offloading computing close to end users to overcome computation and delay issues. Nonetheless, the edge servers may suffer from computing inefficiencies. Indeed, some IoT applications are invoked multiple times by multiple devices. These invocations are often used with the same or similar input data, which leads to the same computational output (results). Still, the edge server willfully executes all received redundant tasks. In this work, we investigate the use of the computation reuse concept at the edge server. We design a network-based computation reuse architecture for IoT applications. The architecture stores previously executed results and reuses them to satisfy newly arrived similar tasks instead of performing computation from scratch. By doing so, we eliminate redundant computation, optimize resource utilization, and decrease task completion time. We implemented the architecture and evaluated its performance both at the networking and application levels. From the networking perspective, we reach up to an 80\% reduction in task completion time and up to 60\% reduction in resource utilization. From the application perspective, we achieve up to 90\% computation correctness and accuracy.

The exponential growth of distributed energy resources is enabling the transformation of traditional consumers in the smart grid into prosumers. Such transition presents a promising opportunity for sustainable energy trading. Yet, the integration of prosumers in the energy market imposes new considerations in designing unified and sustainable frameworks for efficient use of the power and communication infrastructure. Furthermore, several issues need to be tackled to adequately promote the adoption of decentralized renewable-oriented systems, such as communication overhead, data privacy, scalability, and sustainability. In this article, we present the different aspects and challenges to be addressed for building efficient energy trading markets in relation to communication and smart decision-making. Accordingly, we propose a multi-level pro-decision framework for prosumer communities to achieve collective goals. Since the individual decisions of prosumers are mainly driven by individual self-sufficiency goals, the framework prioritizes the individual prosumers' decisions and relies on the 5G wireless network for fast coordination among community members. In fact, each prosumer predicts energy production and consumption to make proactive trading decisions as a response to collective-level requests. Moreover, the collaboration of the community is further extended by including the collaborative training of prediction models using Federated Learning, assisted by edge servers and prosumer home-area equipment. In addition to preserving prosumers' privacy, we show through evaluations that training prediction models using Federated Learning yields high accuracy for different energy resources while reducing the communication overhead.

Emerging applications such as augmented reality and tactile Internet are compute-intensive and latency-sensitive, which hampers their running in constrained end devices alone or in the distant cloud. The stringent requirements of such application drove to the realization of Edge computing in which computation is offloaded near to users. Compute-less networking is an extension of edge computing that aims at reducing computation and abridging communication by adopting in-network computing and computation reuse. Computation reuse aims to cache the result of computations and use them to perform similar tasks in the future and, therefore, avoid redundant calculations and optimize the use of resources. In this paper, we focus on the correctness of the final output produced by computation reuse. Since the input might not be identical but similar, the reuse of previous computation raises questions about the accuracy of the final results. To this end, we implement a proof of concept to study and gauge the effectiveness and efficiency of computation reuse. We are able to reduce task completion time by up to 80% while ensuring high correctness. We further discuss open challenges and highlight future research directions., Comment: Accepted for publication by the IEEE Network Magazine

Biomechanics is a scientific discipline that studies the forces acting on a body and the effects they produce. In this paper, we bring together biomechanists and networking researchers to shed light into how research efforts in biomechanics, primarily related to the study of the human body, can be facilitated through networking and computing technologies, such as edge and cloud computing, Software Defined Networking, and Information-Centric Networking. We first present challenges related to networking and computing that biomechanists face today and we then describe how networking and computing technologies can address them. Finally, we identify directions for future networking research with a focus on biomechanics to facilitate and encourage interdisciplinary collaborations between biomechanists and networking researchers., Comment: Accepted for publication by the IEEE Communications Magazine

Information-Centric Networking (ICN) has emerged as a paradigm to cope with the lack of built-in security primitives and efficient mechanisms for content distribution of today's Internet. However, deploying ICN in a wireless environment poses a different set of challenges compared to a wired environment, especially when it comes to security. In this paper, we present the security issues that may arise and the attacks that may occur from different points of view when ICN is deployed in wireless environments. The discussed attacks may target both applications and the ICN network itself by exploiting elements of the ICN architecture, such as content names and in-network content caches. Furthermore, we discuss potential solutions to the presented issues and countermeasures to the presented attacks. Finally, we identify future research opportunities and directions., Comment: This paper has been accepted for publication by the IEEE Wireless Communications Magazine. The final version will be published by the IEEE

In recent years, there has been a massive increase in the amount of Internet of Things (IoT) devices as well as the data generated by such devices. The participating devices in IoT networks can be problematic due to their resource-constrained nature, and integrating security on these devices is often overlooked. This has resulted in attackers having an increased incentive to target IoT devices. As the number of attacks possible on a network increases, it becomes more difficult for traditional intrusion detection systems (IDS) to cope with these attacks efficiently. In this paper, we highlight several machine learning (ML) methods such as k-nearest neighbour (KNN), support vector machine (SVM), decision tree (DT), naive Bayes (NB), random forest (RF), artificial neural network (ANN), and logistic regression (LR) that can be used in IDS. In this work, ML algorithms are compared for both binary and multi-class classification on Bot-IoT dataset. Based on several parameters such as accuracy, precision, recall, F1 score, and log loss, we experimentally compared the aforementioned ML algorithms. In the case of HTTP distributed denial-of-service (DDoS) attack, the accuracy of RF is 99%. Furthermore, other simulation results-based precision, recall, F1 score, and log loss metric reveal that RF outperforms on all types of attacks in binary classification. However, in multi-class classification, KNN outperforms other ML algorithms with an accuracy of 99%, which is 4% higher than RF.

Information-Centric Networking (ICN) has recently emerged as a prominent candidate for the Future Internet Architecture (FIA) that addresses existing issues with the host-centric communication model of the current TCP/IP-based Internet. Named Data Networking (NDN) is one of the most recent and active ICN architectures that provides a clean slate approach for Internet communication. NDN provides intrinsic content security where security is directly provided to the content instead of communication channel. Among other security aspects, Access Control (AC) rules specify the privileges for the entities that can access the content. In TCP/IP-based AC systems, due to the client-server communication model, the servers control which client can access a particular content. In contrast, ICN-based networks use content names to drive communication and decouple the content from its original location. This phenomenon leads to the loss of control over the content causing different challenges for the realization of efficient AC mechanisms. To date, considerable efforts have been made to develop various AC mechanisms in NDN. In this paper, we provide a detailed and comprehensive survey of the AC mechanisms in NDN. We follow a holistic approach towards AC in NDN where we first summarize the ICN paradigm, describe the changes from channel-based security to content-based security and highlight different cryptographic algorithms and security protocols in NDN. We then classify the existing AC mechanisms into two main categories: Encryption-based AC and Encryption-independent AC. Each category has different classes based on the working principle of AC (e.g., Attribute-based AC, Name-based AC, Identity-based AC, etc). Finally, we present the lessons learned from the existing AC mechanisms and identify the challenges of NDN-based AC at large, highlighting future research directions for the community., Comment: This paper has been accepted for publication by the ACM Computing Surveys. The final version will be published by the ACM

The promising vision of Information-Centric Networking (ICN) and of its realization, Named Data Networking (NDN), has attracted extensive attention in recent years in the context of the Internet of Things (IoT) and Wireless Sensor Networks (WSNs). However, a comprehensive NDN/ICN-based architectural design for WSNs, including specially tailored naming schemes and forwarding mechanisms, has yet to be explored. In this paper, we present single-Channel Cluster-based Information-Centric WSN (CCIC-WSN), an NDN/ICN-based framework to fulfill the requirements of cluster-based WSNs, such as communication between child nodes and cluster heads, association of new child nodes with cluster heads, discovery of the namespace of newly associated nodes, and child node mobility. Through an extensive simulation study, we demonstrate that CCIC-WSN achieves 71-90% lower energy consumption and 74-96% lower data retrieval delays than recently proposed frameworks for NDN/ICN-based WSNs under various evaluation settings., Comment: This paper has been accepted for publication by the IEEE Internet of Things Journal. The copyright is with IEEE and the final version will be published by IEEE

Vehicular Ad Hoc Network has attracted both research and industrial community due to its benefits in facilitating human life and enhancing the security and comfort. However, various issues have been faced in such networks such as information security, routing reliability, dynamic high mobility of vehicles, that influence the stability of communication. To overcome this issue, it is necessary to increase the routing protocols performances, by keeping only the stable path during the communication. The effective solutions that have been investigated in the literature are based on the link prediction to avoid broken links. In this paper, we propose a new solution based on machine learning concept for link prediction, using LR and Support Vector Regression (SVR) which is a variant of the Support Vector Machine (SVM) algorithm. SVR allows predicting the movements of the vehicles in the network which gives us a decision for the link state at a future time. We study the performance of SVR by comparing the generated prediction values against real movement traces of different vehicles in various mobility scenarios, and to show the effectiveness of the proposed method, we calculate the error rate. Finally, we compare this new SVR method with Lagrange interpolation solution.

Based on the dominant paradigm, all the wearable IoT devices used in the healthcare sector also known as the internet of medical things (IoMT) are resource constrained in power and computational capabilities. The IoMT devices are continuously pushing their readings to the remote cloud servers for real-time data analytics, that causes faster drainage of the device battery. Moreover, other demerits of continuous centralizing of data include exposed privacy and high latency. This paper presents a novel Federated Filtering Framework for IoMT devices which is based on the prediction of data at the central fog server using shared models provided by the local IoMT devices. The fog server performs model averaging to predict the aggregated data matrix and also computes filter parameters for local IoMT devices. Two significant theoretical contributions of this paper are the global tolerable perturbation error (${To{l_F}}$) and the local filtering parameter ($\delta$); where the former controls the decision-making accuracy due to eigenvalue perturbation and the later balances the tradeoff between the communication overhead and perturbation error of the aggregated data matrix (predicted matrix) at the fog server. Experimental evaluation based on real healthcare data demonstrates that the proposed scheme saves upto 95\% of the communication cost while maintaining reasonable data privacy and low latency., Comment: 6 pages, 6 Figures, accepted for oral presentation in IEEE ICC 2019, Internet of Things, Federated Learning and Perturbation theory

Internet of Things (IoT) and Network Softwarization are fast becoming core technologies of information systems and network management for next generation Internet. The deployment and applications of IoT ranges from smart cities to urban computing, and from ubiquitous healthcare to tactile Internet. For this reason the physical infrastructure of heterogeneous network systems has become more complicated, and thus requires efficient and dynamic solutions for management, configuration, and flow scheduling. Network softwarization in the form of Software Defined Networks (SDN) and Network Function Virtualization (NFV) has been extensively researched for IoT in recent past. In this article we present a systematic and comprehensive review of virtualization techniques explicitly designed for IoT networks. We have classified the literature into software defined networks designed for IoT, function virtualization for IoT networks, and software defined IoT networks. These categories are further divided into works which present architectural, security, and management solutions. In addition, the paper highlights a number of short term and long term research challenges and open issues related to adoption of software defined Internet of things., Comment: Initial version. An update is in progress