Time synchronization is a fundamental requirement for various industrial automation scenarios. Its use cases range from precise distributed control to deterministic Sequence of Events (SoEs) in case of plant faults. With the advent of Time-Sensitive Networking (TSN), the Generalized Precision Time Protocol (gPTP) has been the protocol of choice to enable sub- $\mu$ s time synchronization between network devices and has superseded the Network Time Protocol (NTP) in most mission-critical systems. However, gPTP requires hardware support in all devices and thus is not compatible with many existing system installations. In this paper, we revisit time synchronization for industrial networks and tackle the question of NTP’s synchronization accuracy in the light of network convergence. We examine the impact of recent capabilities of network devices such as strict priority scheduling and frame preemption on (S)NTP and show that accuracies below 1 ms can be achieved in large networks.

The Industrial Internet of Things (IIoT) has revolutionized industrial sectors with enhanced connectivity, data exchange, and predictive maintenance. However, it faces various challenges from non-IID data distributions and communication overheads, to the consistency and privacy of prediction models for maintenance. Federated Learning (FL) has been considered as a prevalent technique to address privacy concerns. On the other hand, Edge computing (EC) is being increasingly introduced to ensure low-latency data processing in IIoT systems, especially those with time-critical requirements, e.g., industrial robotics and motion control systems. Moreover, the complexity and design dimensions of today’s IIoT systems has lead to the development of large Machine Learning (ML) models with millions of parameters, e.g., using computer vision for field management. This introduces computational and privacy challenges in IIoT scenarios. Innovative FL optimization approaches such as pruning are aimed to tackle these challenges by reducing the number of training parameters, while they may impact accuracy. In this paper, we propose a new technique, called Adaptive Mean aBsolute devIaTion (AMbit), which is an innovative pruning approach optimizing data transmission without compromising model accuracy and inducing additional computation overhead. By dynamically comparing the difference between the current and the previous weight value, AMbit adapts better to parameter fluctuations at different stages, thereby accurately locating those parameters that have less impact on convergence. AMbit’s generality and efficiency are illustrated using MNIST and CIFAR-10 datasets, outperforming traditional Magnitude pruning in FL. For MNIST, AMbit reduces data uploads up to 43.75%, with an increase of 0.62% in accuracy. While for For CIFAR-10, AMbit achieved a 63.44% decrease with a 5.22% drop in accuracy.

Modern software development paradigms, including Service-Oriented Architecture (SOA), tend to make use of available services e.g., web service Application Programming Interfaces (APIs) to generate new software. Thus, for the further advancement of SOA, the development of accurate automatic tasks, such as service discovery and composition, is necessary. Most of these automated tasks rely heavily on web service metadata annotation. The lack of machine-readable documentation and structured metadata reduces the accuracy and volume of automatic data annotation, negatively affecting the performance of automated SOA tasks. This study aims to propose automatic code completion for improving web service-based systems by identifying and capturing service usage collected from public repositories that share Open Source Software (OSS). To this end, a Data-Centric Service Code Generation (DCServCG) model is proposed to improve old-fashioned, general-purpose code generators that neglect essential service-based code characteristics e.g., sequence overlap and bias issues. DCServCG takes advantage of the data-centric concept, i.e., conditional text generation, to overcome the mentioned issues. We have evaluated the approach from the point of view of language modeling metrics. The obtained results indicate that the usage of the data-centric approach reduces perplexity by 1.125. Moreover, the DCServCG model uses de-noising and conditional text generation, which is trained on the transformer by distilling the knowledge, DistilGPT2 (82M parameters) trained faster and its perplexity is 0.363 lower than ServCG (124M parameters) without de-noising and conditional text generation, which lower perplexity value indicates better model generalization performance.

Internet of Things (IoT) systems are rightly receiving considerable interest for many real-world applications, from in-body networks to satellite networks. Such a massive-scale system generates a considerable amount of traffic data, making IoT systems a distributed data source generator. For many reasons, such as the functionality of IoT applications and Quality of Service (QoS) provisioning, classifying these traffic data is of high importance. In the last few years, widespread interest has been expressed in applying Machine Learning (ML)-based techniques for Network Traffic Classification (NTC) tasks. However, the traditional centralized learning-based traffic classifiers pose serious challenges, especially in IoT networks. The centralized ML techniques call for collecting a large amount of data from various IoT devices, which in turn introduces data governance and privacy challenges. Furthermore, in the centralized ML, training data need to be transferred to the Cloud, which increases communication cost and latency. To address these problems, we propose Federated Learning (FL) Internet of Things (IoT) Traffic Classifier (FLITC)-a Federated Learning (FL)-based IoT traffic classification method which is based on the Multi-Layer Perception (MLP) neural network and holds the local data unimpaired on IoT devices by sending only the learned parameters to the aggregation server. Our experimental results show that the FLITC beats centralized learning in preserving the privacy of sensitive data and offers a better degree of accuracy at the cost of a longer training time.

Modern networks generate a massive amount of traffic data streams. Analyzing this data is essential for various purposes, such as network resources management and cyber-security analysis. There is an urgent need for data analytic methods that can perform network data processing in an online manner based on the arrival of new data. Online machine learning (OL) techniques promise to support such type of data analytics. In this paper, we investigate and compare the OL techniques that facilitate data stream analytics in the networking domain. We also investigate the importance of traffic data analytics and highlight the advantages of online learning in this regard, as well as the challenges associated with OL-based network traffic stream analysis, e.g., concept drift and the imbalanced classes. We review the data stream processing tools and frameworks that can be used to process such data online or on-the-fly along with their pros and cons, and their integrability with de facto data processing frameworks. To explore the performance of OL techniques, we conduct an empirical evaluation on the performance of different ensemble- and tree-based algorithms for network traffic classification. Finally, the open issues and the future directions in analyzing traffic data streams are presented. This technical study presents valuable insights and outlook for the network research community when dealing with the requirements and purposes of online data streams analytics and learning in the networking domain.

Internet of Things (IoT) refers to a system of interconnected heterogeneous smart devices communicating without human intervention. A significant portion of existing IoT networks is under the umbrella of ad-hoc and quasi ad-hoc networks. Ad-hoc based IoT networks suffer from the lack of resource-rich network infrastructures that are able to perform heavyweight network management tasks using, e.g. machine learning-based Network Traffic Monitoring and Analysis (NTMA) techniques. Designing light-weight NTMA techniques that do not need to be (re-) trained has received much attention due to the time complexity of the training phase. In this study, a novel pattern recognition method, called Trend-based Online Network Traffic Analysis (TONTA), is proposed for ad-hoc IoT networks to monitor network performance. The proposed method uses a statistical light-weight Trend Change Detection (TCD) method in an online manner. TONTA discovers predominant trends and recognizes abrupt or gradual time-series dataset changes to analyze the IoT network traffic. TONTA is then compared with RuLSIF as an offline benchmark TCD technique. The results show that TONTA detects approximately 60% less false positive alarms than RuLSIF.

Modern communication systems and networks, e.g., Internet of Things (IoT) and cellular networks, generate a massive and heterogeneous amount of traffic data. In such networks, the traditional network management techniques for monitoring and data analytics face some challenges and issues, e.g., accuracy, and effective processing of big data in a real-time fashion. Moreover, the pattern of network traffic, especially in cellular networks, shows very complex behavior because of various factors, such as device mobility and network heterogeneity. Deep learning has been efficiently employed to facilitate analytics and knowledge discovery in big data systems to recognize hidden and complex patterns. Motivated by these successes, researchers in the field of networking apply deep learning models for Network Traffic Monitoring and Analysis (NTMA) applications, e.g., traffic classification and prediction. This paper provides a comprehensive review on applications of deep learning in NTMA. We first provide fundamental background relevant to our review. Then, we give an insight into the confluence of deep learning and NTMA, and review deep learning techniques proposed for NTMA applications. Finally, we discuss key challenges, open issues, and future research directions for using deep learning in NTMA applications.

Quality of Service (QoS) provisioning is based on various network management techniques including resource management and medium access control (MAC). Various techniques have been introduced to automate networking decisions, particularly at the MAC layer. Deep reinforcement learning (DRL), as a solution to sequential decision making problems, is a combination of the power of deep learning (DL), to represent and comprehend the world, with reinforcement learning (RL), to understand the environment and act rationally. In this paper, we present a survey on the applications of DRL in QoS provisioning at the MAC layer. First, we present the basic concepts of QoS and DRL. Second, we classify the main challenges in the context of QoS provisioning at the MAC layer, including medium access and data rate control, and resource sharing and scheduling. Third, we review various DRL algorithms employed to support QoS at the MAC layer, by analyzing, comparing, and identifying their pros and cons. Furthermore, we outline a number of important open research problems and suggest some avenues for future research.

Many Internet of Things (IoT) networks are created as an overlay over traditional ad-hoc networks such as Zigbee. Moreover, IoT networks can resemble ad-hoc networks over networks that support device-to-device (D2D) communication, e.g., D2D-enabled cellular networks and WiFi-Direct. In thesead-hoctypes of IoT networks, efficienttopology managementis a crucial requirement, and in particular in massive scale deployments. Traditionally,clusteringhas been recognized as a common approach for topology management in ad-hoc networks, e.g., in Wireless Sensor Networks (WSNs). Topology management in WSNs and ad-hoc IoT networks has many design commonalities as both need to transfer data to the destination hop by hop. Thus, WSN clustering techniques can presumably be applied for topology management in ad-hoc IoT networks. This requires a comprehensive study on WSN clustering techniques and investigating their applicability to ad-hoc IoT networks. In this article, we conduct a survey of this field based on theobjectivesfor clustering, such as reducing energy consumption and load balancing, as well as the network properties relevant for efficient clustering in IoT, such as network heterogeneity and mobility. Beyond that, we investigate the advantages and challenges of clustering when IoT is integrated with modern computing and communication technologies such as Blockchain, Fog/Edge computing, and 5G. This survey provides useful insights into research on IoT clustering, allows broader understanding of its design challenges for IoT networks, and sheds light on its future applications in modern technologies integrated with IoT.

Wireless Sensor Networks (WSNs) typically include thousands of resource-constrained sensors to monitor their surroundings, collect data, and transfer it to remote servers for further processing. Although WSNs are considered highly flexible ad-hoc networks, network management has been a fundamental challenge in these types of networks given the deployment size and the associated quality concerns such as resource management, scalability, and reliability. Topology management is considered a viable technique to address these concerns. Clustering is the most well-known topology management method in WSNs, grouping nodes to manage them and/or executing various tasks in a distributed manner, such as resource management. Although clustering techniques are mainly known to improve energy consumption, there are various quality-driven objectives that can be realized through clustering. In this paper, we review comprehensively existing WSN clustering techniques, their objectives and the network properties supported by those techniques. After refining more than 500 clustering techniques, we extract about 215 of them as the most important ones, which we further review, catergorize and classify based on clustering objectives and also the network properties such as mobility and heterogeneity. In addition, statistics are provided based on the chosen metrics, providing highly useful insights into the design of clustering techniques in WSNs.

Trend change detection methods find trends in a dataset. Datasets based on Poisson distribution are important to analyze since they mimic many different applications such as computer networks. Our use-cases are simulations of computer networks. The last significant trend is the last predominant trend in a time-series dataset. Our method is a matrix based trend change detection that can analyze datasets with variable sizes. Reducing the time complexity and increasing the accuracy when determining the last significant trend are the goals of our method. We compare our method with RuLSIF, a basic change point detection method, to illustrate the benefits of our approach.