Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
158,297
result(s) for
"Virtual networks"
Sort by:
Graph Neural Networks for Intelligent Modelling in Network Management and Orchestration: A Survey on Communications
The advancing applications based on machine learning and deep learning in communication networks have been exponentially increasing in the system architectures of enabled software-defined networking, network functions virtualization, and other wired/wireless networks. With data exposure capabilities of graph-structured network topologies and underlying data plane information, the state-of-the-art deep learning approach, graph neural networks (GNN), has been applied to understand multi-scale deep correlations, offer generalization capability, improve the accuracy metrics of prediction modelling, and empower state representation for deep reinforcement learning (DRL) agents in future intelligent network management and orchestration. This paper contributes a taxonomy of recent studies using GNN-based approaches to optimize the control policies, including offloading strategies, routing optimization, virtual network function orchestration, and resource allocation. The algorithm designs of converged DRL and GNN are reviewed throughout the selected studies by presenting the state generalization, GNN-assisted action selection, and reward valuation cooperating with GNN outputs. We also survey the GNN-empowered application deployment in the autonomous control of optical networks, Internet of Healthcare Things, Internet of Vehicles, Industrial Internet of Things, and other smart city applications. Finally, we provide a potential discussion on research challenges and future directions.
Journal Article
Securing 5G virtual networks: a critical analysis of SDN, NFV, and network slicing security
5G, the current generation of communication networks is based on the standards defined by 3GPP and other organizations (ETSI, ENISA, NGMN). These standards define virtual networks supported by three basic technologies, SDN, NFV, and Network Slicing. Virtual networks are primarily built using software and have clear advantages that appear to be reduced because of the corresponding loss in security due to the larger attack surface of this type of network. On the other hand, virtual networks can be made even more secure than hardware-based networks by leveraging the flexibility and adaptability of virtual functions and numerous articles have studied different aspects of their security. Current work goes from proposals for specific mechanisms to general studies of threats and defenses. Some of these are systematic literature reviews considering everything published on a specific theme. We prefer to analyze carefully selected papers considered significant and produce from them an overview of the status of the security of the network technologies used by 5G. After this analysis, we have found that although there are many studies of threats, they are not systematic and have confusions about concepts that may mislead implementers; we also found that the large variety of defenses can be confusing to designers. We have therefore conducted a critical analysis of threats and defenses to provide a clear perspective of how to secure these networks. Based on this perspective, we propose directions for research to improve or extend current defenses. We note that although virtual networks have special characteristics, they are examples of systems and much of the theory of systems security applies to them.
Journal Article
Improving Resources Management in Network Virtualization by Utilizing a Software-Based Network
Network virtualization is a way to simultaneously run multiple heterogeneous architectures on a shared substrate. The main issue in network virtualization is mapping virtual networks to substrate network. How to manage substrate resources in mapping phase will have an effective role in improving the use of infrastructure resources. Using software-based networks in network virtualization which separates control logic from data as a new technology, has led to efficient resource management in this context. In this article a software-based network approach has been presented to network virtualization and manage infrastructure resources efficiently. It optimizes mapping function by dynamic resource management of infrastructure resource. We have added a module in the controller to manage the resources dynamically. An initial mapping will be done for arriving new requests based on number of successful requests and arriving time slots. They will not be finalized by writing the rules in the switches before arriving n requests. If some remapping during the n time window is needed, remapping can be done by the controller and the final results are sent to the switches to write the flow rules. The simulation has been done using NS2 simulator showed based on different evaluation criteria such as acceptance rate, average link utilization, cost and delay.
Journal Article
Encyclopedia of networked and virtual organizations
\"This book documents the most relevant contributions to the introduction of networked, dynamic, agile, and virtual organizational models; definitions; taxonomies; opportunities; and reference models and architectures. It creates a repository of the main developments regarding the virtual organization, compiling definitions, characteristics, comparisons, advantages, practices, enabling technologies, and best practices\"--Provided by publisher.
Book
An E2E Network Slicing Framework for Slice Creation and Deployment Using Machine Learning
Network slicing shows promise as a means to endow 5G networks with flexible and dynamic features. Network function virtualization (NFV) and software-defined networking (SDN) are the key methods for deploying network slicing, which will enable end-to-end (E2E) isolation services permitting each slice to be customized depending on service requirements. The goal of this investigation is to construct network slices through a machine learning algorithm and allocate resources for the newly created slices using dynamic programming in an efficient manner. A substrate network is constructed with a list of key performance indicators (KPIs) like CPU capacity, bandwidth, delay, link capacity, and security level. After that, network slices are produced by employing multi-layer perceptron (MLP) using the adaptive moment estimation (ADAM) optimization algorithm. For each requested service, the network slices are categorized as massive machine-type communications (mMTC), enhanced mobile broadband (eMBB), and ultra-reliable low-latency communications (uRLLC). After network slicing, resources are provided to the services that have been requested. In order to maximize the total user access rate and resource efficiency, Dijkstra’s algorithm is adopted for resource allocation that determines the shortest path between nodes in the substrate network. The simulation output shows that the present model allocates optimum slices to the requested services with high resource efficiency and reduced total bandwidth utilization.
Journal Article
Decoding the metaverse : expand your business using Web3
Explores and breaks down the business potential of the metaverse.
BOOK
Secure Virtual Network Provisioning over Key Programmable Optical Networks
Virtual networks have emerged as a promising solution for enabling diverse users to efficiently share bandwidth resources over optical network infrastructures. Despite the invention of various schemes aimed at ensuring secure isolation among virtual networks, the security of data transfer in virtual networks remains a challenging problem. To address this challenge, the concept of evolving traditional optical networks into key programmable optical networks (KPONs) has been proposed. Inspired by this, this paper delves into the establishment of secure virtual networks over KPONs, in which the information-theoretically secure keys can be supplied for ensuring the information-theoretic security of data transfer within virtual networks. A layered architecture for secure virtual network provisioning over KPONs is proposed, which leverages software-defined networking to realize the programmable control of optical-layer resources. With this architecture, a heuristic algorithm, i.e., the key adaptation-based secure virtual network provisioning (KA-SVNP) algorithm, is designed to dynamically allocate key resources based on the adaption between the key supply and key demand. To evaluate the proposed solutions, an emulation testbed is established, achieving millisecond latencies for secure virtual network establishment and deletion. Moreover, numerical simulations indicate that the designed KA-SVNP algorithm performs superior to the benchmark algorithm in terms of the success probability of secure virtual network requests.
Journal Article