IET Networks最新文献

The efficiency improvement of healthcare systems is a major national goal across the world. However, delivering scalable and reliable healthcare services to people, while managing costs, is a challenging problem. The most promising methods to address this issue are based on smart healthcare (s-health) technologies. Furthermore, the combination of edge computing and s-health can yield additional benefits in terms of delay, bandwidth, power consumption, security, and privacy. However, the strategic placement of edge-servers is crucial to achieve further cost and latency benefits. This article is divided into two parts: an AI-based priority mechanism to identify urgent cases, aimed at improving quality of service and quality of experience is proposed. Then, an optimal edge-servers placement (OESP) algorithm to obtain a cost-efficient architecture with lower delay and complete coverage is presented. The results demonstrate that the proposed priority mechanism algorithms can reduce the latency for patients depending on their number and level of urgency, prioritising those with the greatest need. In addition, the OESP algorithm successfully selects the best sites to deploy edge-servers to achieve a cost-efficient system, with an improvement of more than 80%. In sum, the article introduces an improved healthcare system with commendable performance, enhanced cost-effectiveness, and lower latency.

An experimental framework for managing 5G and beyond networks through cloud-native deployments and end-to-end monitoring is presented. The framework uses containerised network functions in a Kubernetes cluster across a multi-domain network spanning cloud and edge hosts. End-to-end monitoring is demonstrated through Grafana dashboards that showcase both infrastructure resources and radio metrics in two scenarios involving UPF re-selection and user mobility. As a third scenario, the authors demonstrate how a decision engine interacts with the experimental platform to perform zero-touch containerised application relocation, highlighting the potential for enabling dynamic and intelligent management of 5G networks and beyond.

Given the observed developments of novel communication modes and the establishment of next-generation cellular networks, mobility modelling and ad hoc routing requirements have emerged. Flying ad hoc networks are key pivots in enabling technological leaps in the domain of on-demand communications, especially in emergency scenarios; as such, resorting to application- and mobility-aware routing is a promising enabler of this emerging set of use cases. This article investigates swarm mobility modelling, and applicable routing protocols, conducting comparative analysis that leads to the introduction of the new Anchored Self-Similar 3D Gauss-Markov Mobility Model (ASSGM-3D), which incorporates a novel set of spatio-temporal statistical metrics.

Cognitive radio (CR) systems are configured to dynamically assess the spectrum utilisation and contribute towards an improved spectrum efficiency. Hence, accurate detection of the incumbent signal in a given channel, popularly known as spectrum sensing (SS), is essential for CR. Here, in the domain of SS, the authors introduce a new goodness-of-fit test (GoFT) founded on p-norm of the observations at the receiver node. To capture the heavy-tailed nature of noise distribution in practical communication channels, the authors utilise generalised Gaussian distribution (GGD) as a noise model. A novel p-norm detector (PND) and a geometric power detector (GPD) is proposed and corresponding probability density function (PDF) under GGD is derived. Via Monte Carlo simulations, the authors show a match of the derived PDFs with the simulation results. Using Neyman-Pearson framework the performances of PND and GPD are compared with an existing differential entropy detector (DED), the well-known energy detector (ED) and joint correlation and energy detector (CED) under GGD noise model. Evaluation of proposed PND and GPD utilising Monte Carlo simulations indicate a superior performance. Further, the experiments employing real-world data establish superiority of the proposed detectors as compared to existing techniques. The authors derive and implement an optimised threshold for PND, providing further improvement in performance.

The satisfaction of the Quality of Service (QoS) levels during an entire service life-cycle is one of the key targets for Service Providers (SP). To achieve this in an optimal way, it is required to predict the exact amount of the needed physical and virtual resources, for example, CPU and memory usage, for any possible combination of parameters that affect the system workload, such as number of users, duration of each request, etc. To solve this problem, the authors introduce a novel architecture and its open-source implementation that a) monitors and collects data from heterogeneous resources, b) uses them to train machine learning models and c) tailors them to each particular service type. The candidate solution is validated in two real-life services showing very good accuracy in predicting the required resources for a large number of operational configurations where a data augmentation method is also applied to further decrease the estimation error up to 32%.

Network security protocols are implemented to address network security challenges. Computer networks and applications have advanced and developed significantly in recent years, but consumers ‘excitement for network technology and high-tech devices has been dampened by continual exposure to data security vulnerabilities. As of now, some individuals refuse to use smart devices due to concerns about the authenticity, confidentiality and integrity of data security leaks. This not only prompts Internet service providers to follow market protection mechanisms but also requires software developers to apply appropriate security protocols to protect computer network security. These applications’ dependability and integrity are dependent not just on the effectiveness of cryptographic algorithms, but also on key management protocols. Understanding network security protocols and implementing high-quality standards to govern the transmission of data in the network are critical components of guaranteeing network security. The article explores data security, primarily at the application layer, various attack methods for different network security protocols and highlights the potential security implications. The study also looks at the corresponding, practical security measures and future research prospects for certain kinds of attacks. Finally, some technical challenges that remain unsolved at the time of writing are summarised, and future trends in cybersecurity are discussed.

The advent of the Internet of Things (IoT) enables different customized services to ease the day-to-day life activities of users by utilizing information attained through the internet connectivity of low-powered sensing devices. Due to device diversity and resource constraints of participating devices, IoT is vulnerable to security attacks. Consequently, authentication is the fundamental measure for using IoT services in the context of network security. IoT devices’ resource captivity makes designing robust and secure authentication mechanisms challenging. Besides, existing user authentication mechanisms are designed assuming a user always accesses an IoT environment using a particular device. However, nowadays, most users employ multiple devices to access the internet; subsequently, it needs an authentication mechanism to handle this diversity. This paper addresses this limitation and proposes a new One-Time Password (OTP)-based user authentication scheme supporting user access from multiple devices in an IoT environment. We verify the proposed scheme using widely used BAN logic, AVISPA tool, and informal security analysis, guaranteeing that our scheme preserves the necessary security features. Comparative performance analysis shows that our scheme achieves comparable computation, storage, and communication costs concerning existing works. Moreover, simulation results demonstrate that the proposed method also sustains satisfactory network performance.

Within Edge and Fog computing, edge and fog nodes must be optimally located at the network edge to minimise the network's overall latency. This survey addresses all aspects of these nodes' placement problems. Literature on edge and fog nodes' placement is collected from reputable databases (IEEE Xplore digital library, Scopus, ScienceDirect, and Web of Science) using a search query. Manual search using keywords and the snowball method is also used to get as many related papers as possible. According to defined inclusion criteria, retrieved documents are filtered to 64 articles for eight years (2015–2022). Depending on the optimisation method used, literature is classified into six categories. The first relies on Integer programming, accounting for 20.3% (13/64). The second category depends on heuristic and metaheuristic methods, accounting for 20.3% (13/64). The third category depends on hybrid methods between the two aforementioned categories accounting for 18.7% (12/64). Forth category depends on clustering methods, accounting for 11% (7/64). The fifth category depends on reinforcement learning, accounting for 6.3% (4/64). And the final category depends on the hybrid methods between two or more methods mentioned above, accounting for 23.4% (15/64). Papers have been analysed to get information like the optimisation problem, the method used for solving it, considered parameters, objectives, constraints, implementation tools, and evaluation methods.

Internet-wide traffic redirection attacks have been reported for long, and are mainly caused by Border Gateway Protocol route hijacking. Such attacks can be quite harmful, impairing access to popular Internet sites for long periods. This work addresses the use of machine learning techniques (both unsupervised and supervised) leveraging from a distributed monitoring infrastructure of probes that measure the round trip time to Internet sites under surveillance. The detection process is separated into two stages: per-probe classification and a combination of individual probe decisions. Our results show that the best strategy is to classify using an unsupervised technique based on Tukey's method and to combine using Hidden Markov Models, due to its performance and adaptability to different attack types.