The development of 6th-generation (6G) mobile networks brings advancements in wireless communications, including lower latency, higher data rates, and improved spectral efficiency, as well as enhanced facilitation for the use of Artificial Intelligence (AI) technologies, integrated communications, sensing, and renewable energy sources. To harness this potential, a scalable framework integrating edge and cloud computing is needed to provide a robust computing continuum for various IoT applications and services. This paper analyses the strengths and weaknesses of existing and emerging distributed computing frameworks—ranging from traditional centralized IoT architectures, through fog, edge, and local-edge architectures, to the full three-tier edge–cloud continuum—in terms of IoT service orchestration and diverse application requirements. We emphasize the latest evolutionary step, three-tier edge-cloud continuum, which aims to resolve the most significant limitations of the previous frameworks, recognized in the literature. It enables efficient IoT service orchestration by utilizing distributed resources and computation closer to end users, while taking full advantage of the centralized resources at data centers. We evaluate the maturity of these frameworks, considering factors such as scalability, resource efficiency, adaptability to changes, resource availability, security and privacy, as well as robustness and resilience. Overall, this study aims to serve as a roadmap for researchers, network architects, and industry stakeholders to make informed decisions on implementing the computing continuum in 6G networks.

As the complexity of communication networks continues to increase, driven by a diverse array of devices, services and applications, the adoption of Machine Learning (ML) has seen a significant rise to address various challenges ranging from management to security. Regarding network security, the application of ML ranges from preventive measures to detection and remediation due to its ability to dynamically learn and adapt to evolving threat landscapes. However, ML requires a significant amount of resources, mainly due to the fact that ML operates on data, and the volumes of data are consistently rising. This review article explores the resource consumption aspect of ML techniques used for network security and provides a comprehensive review of the current state of research. Moreover, we propose a taxonomy that can be used to classify the methods through which the resource consumption can be reduced for different ML-based network security implementations. The focus of the study encompasses several key aspects related to resource consumption, including energy, computing, memory, latency, bandwidth, and human resources. These resources are critical in improving the efficiency and optimizing the reliability and sustainability of network security solutions. Furthermore, based on an extensive literature review, we summarize key points regarding optimizing resource consumption in ML-based network security solutions. Finally, the challenges and future research directions for resource-efficient, ML-based network security solutions are outlined to aid in the advancement of research in this area.

Publisher Copyright: © 2024 The Author(s)

Unmanned aerial vehicles (UAVs), or drones, are transforming industries due to their affordability, ease of use, and adaptability. This emphasizes the need for reliable communication links, especially in beyond-line-of-sight scenarios. This paper investigates the feasibility of predicting future quality of service (QoS) in UAV payload communication links, with a special focus on 5G cellular technology. Through field tests conducted in a suburban environment, we explore challenges and trade-offs that cellular-connected UAVs face, particularly in the context of frequency band selection. We employed machine learning models to forecast uplink (UL) throughput for UAV payload communication, highlighting the significance of diverse training data for accurate predictions. The results reveal the effect of frequency band selection on UAV UL throughput rates at varying altitudes and the influence of integrating diverse feature sets, including radio, network, and spatial features, on ML model performance. These insights provide a foundation for addressing the complexities in UAV communications and enhancing UAV operations in modern networks.

The 5G networks revolutionize industrial connectivity, offering reliable, low-latency, and high-bandwidth communication for diverse critical environments, including mining. However, the current uplink capacity of 5G poses limitations for transmitting large data volumes, necessitating the exploration of concurrent wireless solutions. To address this, our paper proposes leveraging a Wi-Fi mesh network in conjunction with 5G and a Multipath Transmission Control Protocol (MPTCP) scheduler. To do so, we deploy a test network in a tunnel able to serve 5G as well as mesh Wi-Fi. We then evaluate how a moving vehicle equipped with a connectivity unit can take advantage of both networks even in challenging conditions by turning off part of the 5G base stations. Ultimately, we demonstrate that employing Wi-Fi and MPTCP can augment 5G in scenarios with limited uplink capacity or unreliable coverage.

Multi-access Edge Computing (MEC) has become inevitable in future communication networks due to its pivotal role in latency critical and distributed services. New emerging services seek to further push the concept of MEC to user environments for localized, private and even quicker processing of information. Micro MEC (µMEC) fulfills such needs of emerging services. In this article, we provide an overview of the security landscape of the µMEC philosophy and technologies. Security challenges, potential solutions for those challenges are discussed and the missing gaps are highlighted to stir further research in this direction.

The number of malware attempts that try to bypass the existing Network Intrusion Detection System (NIDS) is increasing. To detect illegal access to servers, deep analysis of the server-side network traffic has become increasingly important. However, the existing approaches have serious performance limitations in terms of real-time and accurate traffic detection. These limitations are mainly because of i) the rigid feature extraction and rule matching techniques of NIDS, which are insensitive to incremental network traffic, and ii) the strong correlation and coupling of malicious traffic to large normal traffic. To address these limitations, we propose a Few-shot Latent Dirichlet Generative Learning (FLAG) scheme for semantic-aware traffic detection in this paper. In FLAG, a Latent Dirichlet Allocation (LDA)-based pseudo samples generation algorithm is designated to augment the few-shot training data, which is essential to improve traffic classification accuracy. Furthermore, we propose a Fuzziness Recycle Method (FRM) to further improve the long short-term memory (LSTM)-based classifier’s robustness. Experimental results in real scenarios demonstrate that malicious traffic can be efficiently detected when only few-shot samples are learned. The results also reveal that the proposed scheme outperforms the state-of-the-art methods in detection accuracy.

Blockchain is so far well-known for its potential applications in financial and banking sectors. However, blockchain as a decentralized and distributed technology can be utilized as a powerful tool for immense daily life applications. Healthcare is one of the prominent applications area among others where blockchain is supposed to make a strong impact. It is generating wide range of opportunities and possibilities in current healthcare systems. Therefore, this paper is all about exploring the potential applications of blockchain technology in current healthcare systems and highlights the most important requirements to fulfill the need of such systems such as trustless and transparent healthcare systems. In addition, this work also presents the challenges and obstacles needed to resolve before the successful adoption of blockchain technology in healthcare systems. Furthermore, we introduce the smart contract for blockchain based healthcare systems which is key for defining the pre-defined agreements among various involved stakeholders.

Software Defined Network (SDN) is an advanced approach to designing dynamic, manageable, cost-effective, and adaptable network architectures. SDN will play a key role as an enabler for 5G and future networks. Transferring network monitoring functions to a software entity working in conjunction with configurable hardware accelerators through a scheme called Software Defined Monitoring (SDM) is one promising way to attain the dynamism necessary for the monitoring of the next generation-networks. In this paper, we propose a novel SDM architecture for future mobile backhual networks. As an SDN solution, the proposed architecture provides more granular and dynamic network management functions through its programmable interface, centralized control, and virtualized abstractions. At the same time, the SDM framework intuitively seem prone to various challenges that come with the separation of the control and data planes of middleboxes. This paper collects specific opportunities, vulnerabilities as well as challenges related to SDM. It also highlights how SDM can be used to solve the current limitations in legacy monitoring systems. The feasibility of the proposed SDM architecture is verified by using a testbed implementation.