Towards smarter cyberthreats detection model for industrial internet of things (IIoT) 4.0

Rejab Hajlaoui, Tarek Moulahi, Salah Zidi, Salim El Khediri, Bechir Alaya, Sherali Zeadally

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

Recently, Artificial Intelligence (AI) has been applied to the Internet of Things (IoT) as well as Industrial IoT 4.0 (IIoT). This has improved the autonomous behavior of smart things in terms of making independent and smart decisions. However, AI techniques are the target of many attacks like the poisoning of training data and the evasion attack. In this context, blockchain can be joined to AI methods to protect their execution. In this paper, we propose SmarterChain to safely detect cyberthreats in IoT. First, we train and generate machine learning models for cyberthreats detection. Next, we protect these models by embedding them on the blockchain. Fi- nally, the models are used by intelligent devices anywhere to facilitate safe decision-making regarding cyberthreats. In addition to identifying the type of attack, our proposed solution outperforms previous detection methods, using the same dataset, in terms of safety ensured by smart contracts, with low overhead due to deployment and runtime.

Original languageEnglish
Article number100595
JournalJournal of Industrial Information Integration
Volume39
DOIs
StatePublished - May 2024

Bibliographical note

Publisher Copyright:
© 2024 Elsevier Inc.

Keywords

  • Blockchain
  • Cyberthreats
  • Industrial internet of things (IIoT)
  • Machine learning (ML)
  • Smart contract

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Information Systems and Management

Fingerprint

Dive into the research topics of 'Towards smarter cyberthreats detection model for industrial internet of things (IIoT) 4.0'. Together they form a unique fingerprint.

Cite this