A Secured Industrial Internet-of-Things Architecture Based on Blockchain Technology and Machine Learning for Sensor Access Control Systems in Smart Manufacturing

In this paper, a layered architecture incorporating Blockchain technology (BCT) and Machine Learning (ML) is proposed in the context of the Industrial Internet-of-Things (IIoT) for smart manufacturing applications. The proposed architecture is composed of five layers covering sensing, network/protoc...

Full description

Saved in:
Bibliographic Details
Published in:Applied sciences 2022-05, Vol.12 (9), p.4641
Main Authors: Mrabet, Hichem, Alhomoud, Adeeb, Jemai, Abderrazek, Trentesaux, Damien
Format: Article
Language:English
Subjects:
Access control
Algorithms
Artificial intelligence
Bayesian analysis
Blockchain
Communication
Computer architecture
Control systems
Correlation coefficient
Correlation coefficients
Data analysis
Data integrity
Denial of service attacks
Energy
Evaluation
industrial IoT
Injection
Internet of Things
Learning algorithms
Learning theory
Machine learning
Manufacturing
Neural networks
Optimization techniques
Privacy
Radio frequency identification
security solutions
security threats
Sensors
smart manufacturing
Support vector machines
Technology
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, a layered architecture incorporating Blockchain technology (BCT) and Machine Learning (ML) is proposed in the context of the Industrial Internet-of-Things (IIoT) for smart manufacturing applications. The proposed architecture is composed of five layers covering sensing, network/protocol, transport enforced with BCT components, application and advanced services (i.e., BCT data, ML and cloud) layers. BCT enables gathering sensor access control information, while ML brings its effectivity in attack detection such as DoS (Denial of Service), DDoS (Distributed Denial of Service), injection, man in the middle (MitM), brute force, cross-site scripting (XSS) and scanning attacks by employing classifiers differentiating between normal and malicious activity. The design of our architecture is compared to similar ones in the literature to point out potential benefits. Experiments, based on the IIoT dataset, have been conducted to evaluate our contribution, using four metrics: Accuracy, Precision, Sensitivity and Matthews Correlation Coefficient (MCC). Artificial Neural Network (ANN), Decision Tree (DT), Random Forest, Naive Bayes, AdaBoost and Support Vector Machine (SVM) classifiers are evaluated regarding these four metrics. Even if more experiments are required, it is illustrated that the proposed architecture can reduce significantly the number of DDoS, injection, brute force and XSS attacks and threats within an advanced framework for sensor access control in IIoT networks based on a smart contract along with ML classifiers.
ISSN:2076-3417
2076-3417
DOI:10.3390/app12094641