SA-FLIDS: secure and authenticated federated learning-based intelligent network intrusion detection system for smart healthcare DOI Creative Commons
Radjaa Bensaid, Nabila Labraoui, Ado Adamou Abba Ari

et al.

PeerJ Computer Science, Journal Year: 2024, Volume and Issue: 10, P. e2414 - e2414

Published: Dec. 13, 2024

Smart healthcare systems are gaining increased practicality and utility, driven by continuous advancements in artificial intelligence technologies, cloud fog computing, the Internet of Things (IoT). However, despite these transformative developments, challenges persist within IoT devices, encompassing computational constraints, storage limitations, attack vulnerability. These attacks target sensitive health information, compromise data integrity, pose obstacles to overall resilience sector. To address vulnerabilities, Network-based Intrusion Detection Systems (NIDSs) crucial fortifying smart networks ensuring secure use IoMT-based applications mitigating security risks. Thus, this article proposes a novel Secure Authenticated Federated Learning-based NIDS framework using Blockchain (SA-FLIDS) for fog-IoMT-enabled systems. Our research aims improve privacy reduce communication costs. Furthermore, we also weaknesses decentralized learning systems, like Sybil Model Poisoning attacks. We leverage blockchain-based Self-Sovereign Identity (SSI) model handle client authentication communication. Additionally, Trimmed Mean method aggregate data. This helps effect unusual or malicious inputs when creating model. approach is evaluated on real traffic datasets such as CICIoT2023 EdgeIIoTset. It demonstrates exceptional robustness against adversarial findings underscore potential our technique applications.

Language: Английский

Cybersecurity Solutions for Industrial Internet of Things–Edge Computing Integration: Challenges, Threats, and Future Directions DOI Creative Commons
Tamara Zhukabayeva, Lazzat Zholshiyeva, Nurdaulet Karabayev

et al.

Sensors, Journal Year: 2025, Volume and Issue: 25(1), P. 213 - 213

Published: Jan. 2, 2025

This paper provides the complete details of current challenges and solutions in cybersecurity cyber-physical systems (CPS) within context IIoT its integration with edge computing (IIoT–edge computing). We systematically collected analyzed relevant literature from past five years, applying a rigorous methodology to identify key sources. Our study highlights prevalent layer attacks, common intrusion methods, critical threats facing IIoT–edge environments. Additionally, we examine various types cyberattacks targeting CPS, outlining their significant impact on industrial operations. A detailed taxonomy primary security mechanisms for CPS is developed, followed by comparative analysis our approach against existing research. The findings underscore widespread vulnerabilities across architecture, particularly relation DoS, ransomware, malware, MITM attacks. review emphasizes advanced technologies, including machine learning (ML), federated (FL), blockchain, blockchain–ML, deep (DL), encryption, cryptography, IT/OT convergence, digital twins, as essential enhancing real-time data protection computing. Finally, outlines potential future research directions aimed at advancing this rapidly evolving domain.

Language: Английский

Citations

2
Multi-attention DeepCRNN: an efficient and explainable intrusion detection framework for Internet of Medical Things environments DOI

Nikhil Sharma,

Prashant Giridhar Shambharkar

Knowledge and Information Systems, Journal Year: 2025, Volume and Issue: unknown

Published: April 5, 2025

Language: Английский

Citations

0
Federated learning with Blockchain on Denial-of-Service attacks detection and classification of edge IIoT networks using Deep Transfer Learning model DOI
Monir Abdullah, Hanan Abdullah Mengash, Mohammed Maray

et al.

Computers & Electrical Engineering, Journal Year: 2025, Volume and Issue: 124, P. 110319 - 110319

Published: April 18, 2025

Language: Английский

Citations

0
A hierarchical blockchain architecture for federated learning in edge computing networks DOI
Shuyang Ren, Choonhwa Lee

The Journal of Supercomputing, Journal Year: 2025, Volume and Issue: 81(7)

Published: May 2, 2025

Language: Английский

Citations

0
A Theoretical Framework for Decentralized Intrusion Detection in Smart Networks Using Blockchain and Machine Learning DOI

Moinul Alam,

Mostafa Monzur Hasan,

Arvil Nath Akash

et al.

Lecture notes on data engineering and communications technologies, Journal Year: 2025, Volume and Issue: unknown, P. 245 - 256

Published: Jan. 1, 2025

Language: Английский

Citations

0
FBLearn: Decentralized Platform for Federated Learning on Blockchain DOI Open Access

Daniel Djolev,

Milena Lazarova,

Ognyan Nakov

et al.

Electronics, Journal Year: 2024, Volume and Issue: 13(18), P. 3672 - 3672

Published: Sept. 16, 2024

In recent years, rapid technological advancements have propelled blockchain and artificial intelligence (AI) into prominent roles within the digital industry, each having unique applications. Blockchain, recognized for its secure transparent data storage, AI, a powerful tool analysis decision making, exhibit common features that render them complementary. At same time, machine learning has become robust influential technology, adopted by many companies to address non-trivial technical problems. This adoption is fueled vast amounts of generated utilized in daily operations. An intriguing intersection AI occurs realm federated learning, distributed approach allowing multiple parties collaboratively train shared model without centralizing data. paper presents decentralized platform FBLearn implementation blockchain, which enables us harness benefits necessity exchanging sensitive customer or product data, thereby fostering trustless collaboration. As network introduced training replace centralized server, global aggregation approaches be utilized. investigates several techniques based on local average ensemble using either globally validation evaluation. The suggested are experimentally evaluated two use cases platform: credit risk scoring random forest classifier card fraud detection logistic regression. experimental results confirm adaptive weight calculation quality enhance robustness model. performance evaluation metrics ROC curves prove strategies successfully isolate influence low-quality models final proposed system’s ability outperform created with separate datasets underscores potential collaborative efforts improve accuracy compared models. Integrating forward-looking collaboration while addressing privacy concerns.

Language: Английский

Citations

1
SA-FLIDS: secure and authenticated federated learning-based intelligent network intrusion detection system for smart healthcare DOI Creative Commons
Radjaa Bensaid, Nabila Labraoui, Ado Adamou Abba Ari

et al.

PeerJ Computer Science, Journal Year: 2024, Volume and Issue: 10, P. e2414 - e2414

Published: Dec. 13, 2024

Smart healthcare systems are gaining increased practicality and utility, driven by continuous advancements in artificial intelligence technologies, cloud fog computing, the Internet of Things (IoT). However, despite these transformative developments, challenges persist within IoT devices, encompassing computational constraints, storage limitations, attack vulnerability. These attacks target sensitive health information, compromise data integrity, pose obstacles to overall resilience sector. To address vulnerabilities, Network-based Intrusion Detection Systems (NIDSs) crucial fortifying smart networks ensuring secure use IoMT-based applications mitigating security risks. Thus, this article proposes a novel Secure Authenticated Federated Learning-based NIDS framework using Blockchain (SA-FLIDS) for fog-IoMT-enabled systems. Our research aims improve privacy reduce communication costs. Furthermore, we also weaknesses decentralized learning systems, like Sybil Model Poisoning attacks. We leverage blockchain-based Self-Sovereign Identity (SSI) model handle client authentication communication. Additionally, Trimmed Mean method aggregate data. This helps effect unusual or malicious inputs when creating model. approach is evaluated on real traffic datasets such as CICIoT2023 EdgeIIoTset. It demonstrates exceptional robustness against adversarial findings underscore potential our technique applications.

Language: Английский

Citations

0