Artificial Intelligence for Trustworthy Industrial Internet of Things

A special issue of Big Data and Cognitive Computing (ISSN 2504-2289).

Deadline for manuscript submissions: closed (31 October 2022) | Viewed by 9121

Special Issue Editors


E-Mail Website
Guest Editor
School of Information Technology, Halmstad University, Halmstad, Sweden
Interests: cybersecurity; IoT security; digital forensics; machine learning

E-Mail Website
Guest Editor
School of Engineering&Computing Sciences, Texas A&M University Corpus Christi, Corpus Christi, TX, USA
Interests: machine learning; deep learning; computer vision; natural language processing

Special Issue Information

Dear Colleagues,

The Industrial Internet of Things (IIoT) is a sensory structure open to the Internet to supervise and control critical industrial applications. Industrial IoT (IIoT) resolved various industrial platforms’ concerns by promoting self-controlling systems and assuring real-time operations. However, the fast and security-inconsiderate adoption of the IIoT has revealed several security weaknesses to industrial applications. Additionally, IIoT networks produce a vast amount of critical information, and where the corresponding data are not carried and examined securely, a privacy breach can occur.

Unfortunately, traditional security measures, such as data encryption, digital authentication, and access control, are not the optimum choice for IIoT assurance due to network heterogeneity and resource limitations. To utilize the IIoT efficiently, intelligent approaches are demanded to address IIoT security concerns reliably and efficiently.

Machine learning and deep learning (ML/DL) have received significant attention in academic and industrial areas in the last few years. They are considered one of the best computational models that can guarantee intelligent features for IIoT systems.

ML/DL can be extremely beneficial for data profiling, examination, and performance enhancement for the IIoT network components (i.e., sensors, actuators, controllers, and communication means), where the produced information can be used to locate and identify the system weaknesses and potential attacks. Threats and vulnerabilities can be detected at initial phases as the ML/DL-associated solutions have the precedence to discover novel attacks. Subsequently, they can provide advanced security methods for the IIoT context to make it more trustworthy and reliable than before.

Topics of interest in this Special Issue include but are not limited to the following research tracks and technologies:

AI-supported security/privacy methods for IIoT;

IIoT authentication and access control using ML/DL;

Intrusion detection and prevention for IIoT using ML/DL;

Cyberattacks detection and prevention for IIoT using ML/DL;

Privacy breaching detection and prevention for IIoT using ML/DL;

ML/DL threat modeling for IIoT;

Malicious behavior detection for IIoT using ML/DL;

Physical attacks detection in IIoT using ML/DL;

Hardware security using ML/DL for IIoT.

Dr. Mohamed Eldefrawy
Mr. Mahmoud Eldefrawy
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Big Data and Cognitive Computing is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Industrial Internet of Things
  • IIoT security and privacy
  • Artificial Intelligence
  • machine learning
  • deep learning
  • intrusion detection and prevention
  • hardware security

Published Papers (2 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

11 pages, 440 KiB  
Article
Exploring Ensemble-Based Class Imbalance Learners for Intrusion Detection in Industrial Control Networks
by Maya Hilda Lestari Louk and Bayu Adhi Tama
Big Data Cogn. Comput. 2021, 5(4), 72; https://doi.org/10.3390/bdcc5040072 - 06 Dec 2021
Cited by 18 | Viewed by 3401
Abstract
Classifier ensembles have been utilized in the industrial cybersecurity sector for many years. However, their efficacy and reliability for intrusion detection systems remain questionable in current research, owing to the particularly imbalanced data issue. The purpose of this article is to address a [...] Read more.
Classifier ensembles have been utilized in the industrial cybersecurity sector for many years. However, their efficacy and reliability for intrusion detection systems remain questionable in current research, owing to the particularly imbalanced data issue. The purpose of this article is to address a gap in the literature by illustrating the benefits of ensemble-based models for identifying threats and attacks in a cyber-physical power grid. We provide a framework that compares nine cost-sensitive individual and ensemble models designed specifically for handling imbalanced data, including cost-sensitive C4.5, roughly balanced bagging, random oversampling bagging, random undersampling bagging, synthetic minority oversampling bagging, random undersampling boosting, synthetic minority oversampling boosting, AdaC2, and EasyEnsemble. Each ensemble’s performance is tested against a range of benchmarked power system datasets utilizing balanced accuracy, Kappa statistics, and AUC metrics. Our findings demonstrate that EasyEnsemble outperformed significantly in comparison to its rivals across the board. Furthermore, undersampling and oversampling strategies were effective in a boosting-based ensemble but not in a bagging-based ensemble. Full article
(This article belongs to the Special Issue Artificial Intelligence for Trustworthy Industrial Internet of Things)
Show Figures

Figure 1

17 pages, 4438 KiB  
Article
Preparing Datasets of Surface Roughness for Constructing Big Data from the Context of Smart Manufacturing and Cognitive Computing
by Saman Fattahi, Takuya Okamoto and Sharifu Ura
Big Data Cogn. Comput. 2021, 5(4), 58; https://doi.org/10.3390/bdcc5040058 - 25 Oct 2021
Cited by 9 | Viewed by 4909
Abstract
In smart manufacturing, human-cyber-physical systems host digital twins and IoT-based networks. The networks weave manufacturing enablers such as CNC machine tools, robots, CAD/CAM systems, process planning systems, enterprise resource planning systems, and human resources. The twins work as the brains of the enablers; [...] Read more.
In smart manufacturing, human-cyber-physical systems host digital twins and IoT-based networks. The networks weave manufacturing enablers such as CNC machine tools, robots, CAD/CAM systems, process planning systems, enterprise resource planning systems, and human resources. The twins work as the brains of the enablers; that is, the twins supply the required knowledge and help enablers solve problems autonomously in real-time. Since surface roughness is a major concern of all manufacturing processes, twins to solve surface roughness-relevant problems are needed. The twins must machine-learn the required knowledge from the relevant datasets available in big data. Therefore, preparing surface roughness-relevant datasets to be included in the human-cyber-physical system-friendly big data is a critical issue. However, preparing such datasets is a challenge due to the lack of a steadfast procedure. This study sheds some light on this issue. A state-of-the-art method is proposed to prepare the said datasets for surface roughness, wherein each dataset consists of four segments: semantic annotation, roughness model, simulation algorithm, and simulation system. These segments provide input information for digital twins’ input, modeling, simulation, and validation modules. The semantic annotation segment boils down to a concept map. A human- and machine-readable concept map is thus developed where the information of other segments (roughness model, simulation algorithm, and simulation system) is integrated. The delay map of surface roughness profile heights plays a pivotal role in the proposed dataset preparation method. The successful preparation of datasets of surface roughness underlying milling, turning, grinding, electric discharge machining, and polishing shows the efficacy of the proposed method. The method will be extended to the manufacturing processes in the next phase of this study. Full article
(This article belongs to the Special Issue Artificial Intelligence for Trustworthy Industrial Internet of Things)
Show Figures

Figure 1

Back to TopTop