相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Chemistry, Analytical
Patricia Cardoso et al.
Summary: This study proposes a solution that incorporates software-defined networking (SDN), network function virtualization (NFV), and fog computing (FC) to autonomously activate or deactivate embedded virtualized resources in response to clients' requests for edge services. Extensive tests show the superior performance of the proposed elastic edge resource provisioning algorithm, resulting in higher flow rate, smaller delay, and reduced loss compared to non-proactive controller operation. Furthermore, the controller records the duration of each edge service session, enabling resource accounting per session.
Article
Computer Science, Information Systems
Daniela M. Casas-Velasco et al.
Summary: Traditional routing protocols have limitations in terms of routing decision speed, adaptability to traffic changes, and support for quality of service requirements. To address these limitations, a deep reinforcement learning-based intelligent routing solution in software-defined networking (SDN) called DRSIR is proposed. DRSIR overcomes the limitations of reinforcement learning-based solutions and provides efficient and intelligent routing by considering path-state metrics.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2022)
Proceedings Paper
Computer Science, Information Systems
Zain Ul Abideen Tariq et al.
Summary: Network Intrusion Detection systems (NIDS) are crucial for ensuring the safety and security of communication and information networks. This paper discusses the limitations of signature-based IDS and proposes the use of anomaly-based IDS and Reinforcement Learning for intrusion detection. It also explores the application of Multi-Armed Bandit approaches for hyper-parameter optimization in unsupervised anomaly detection in smart homes, achieving significant improvements in accuracy and performance.
2022 INTERNATIONAL CONFERENCE ON CYBER WARFARE AND SECURITY (ICCWS)
(2022)
Article
Computer Science, Information Systems
Samarth Gupta et al.
Summary: In this study, a unified approach was developed to leverage the correlations between rewards in a multi-armed bandit framework, resulting in fundamental generalizations of classic algorithms. Rigorous analysis showed that the proposed algorithms outperformed traditional methods, achieving order-optimal regret in correlated settings. Experimentation on real datasets further confirmed the significant improvement over classical bandit algorithms.
IEEE TRANSACTIONS ON INFORMATION THEORY
(2021)
Article
Computer Science, Information Systems
Xi Huang et al.
Summary: In this study, a solution is proposed to minimize costs and ensure queue stability in SDN systems through integrated learning and control design. Combining multi-armed bandit problem and optimal control techniques, the research achieves a balance between exploration and exploitation as well as long-term stability constraints.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2021)
Article
Computer Science, Information Systems
Rashid Amin et al.
Summary: The introduction of Software-Defined Networking (SDN) separated the control plane and data plane in traditional networks, providing additional features and tools to address issues such as routing optimization. Researchers are leveraging machine learning (ML) techniques to enhance SDN routing applications, with a focus on supervised learning, unsupervised learning, and reinforcement learning.
Article
Computer Science, Theory & Methods
Ryan Heartfield et al.
Summary: The modern Internet of Things (IoT)-based smart home presents a challenging security environment due to changing devices, new vulnerabilities, and different user interactions. MAGPIE is the first smart home intrusion detection system that autonomously adjusts its decision function based on changing conditions. It considers both cyber and physical data sources, detects human presence, and achieves high accuracy in intrusion detection.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2021)
Article
Computer Science, Information Systems
Fengxiao Tang et al.
Summary: End-to-end quality of service (QoS) and quality of experience (QoE) guarantee is crucial for network optimization, especially in 5G and future 6G networks. Machine learning algorithms are seen as key solutions for optimizing 6G networks, but there are still many challenges and open issues to be addressed.
IEEE COMMUNICATIONS SURVEYS AND TUTORIALS
(2021)
Article
Computer Science, Information Systems
Justus Rischke et al.
Article
Multidisciplinary Sciences
Jose Moura et al.
SN APPLIED SCIENCES
(2020)
Article
Computer Science, Information Systems
Fatima Hussain et al.
IEEE COMMUNICATIONS SURVEYS AND TUTORIALS
(2020)
Article
Computer Science, Information Systems
Mingzhe Chen et al.
IEEE COMMUNICATIONS SURVEYS AND TUTORIALS
(2019)
Article
Computer Science, Hardware & Architecture
Sebastien Henri et al.
IEEE-ACM TRANSACTIONS ON NETWORKING
(2018)
Article
Computer Science, Information Systems
Raouf Boutaba et al.
JOURNAL OF INTERNET SERVICES AND APPLICATIONS
(2018)
Article
Computer Science, Theory & Methods
P Auer et al.
SIAM JOURNAL ON COMPUTING
(2003)
Article
Computer Science, Artificial Intelligence
P Auer et al.
