Article
Engineering, Multidisciplinary
Sahar Soliman, Wed Oudah, Ahamed Aljuhani
Summary: The widespread deployment of the Industrial Internet of Things (IIoT) has raised the challenge of security and privacy. This paper proposes an intelligent detection system that effectively identifies and mitigates cyberattacks in IIoT networks using techniques such as singular value decomposition (SVD) and synthetic minority oversampling (SMOTE).
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Chemistry, Analytical
Segun I. Popoola, Bamidele Adebisi, Ruth Ande, Mohammad Hammoudeh, Kelvin Anoh, Aderemi A. Atayero
Summary: This paper proposes an efficient DL-based botnet attack detection algorithm that can handle highly imbalanced network traffic data, using Synthetic Minority Oversampling Technique (SMOTE) to achieve class balance. Experimental results show that this approach outperforms state-of-the-art ML and DL models in detecting botnet attacks in IoT networks.
Article
Computer Science, Information Systems
Henry Vargas, Carlos Lozano-Garzon, German A. Montoya, Yezid Donoso
Summary: This paper integrates Blockchain algorithms and Machine Learning techniques to create a comprehensive protection mechanism for IoT device networks, allowing for threat identification, activation of secure information transfer mechanisms, and adaptation to the computational capabilities of industrial IoT. The proposed solution achieves its objectives and is presented as a viable mechanism for detecting and containing intruders in an IoT network, surpassing traditional detection mechanisms such as an IDS in some cases.
Article
Automation & Control Systems
Tu N. Nguyen, Quoc-Dung Ngo, Huy-Trung Nguyen, Giang Long Nguyen
Summary: In recent years, attackers have increasingly targeted IoT devices in the industrial Internet of things (IIoT), with IoT botnet emerging as the most urgent security issue. The main methods for detecting IoT botnets are static, dynamic, and hybrid analysis. This article presents a novel method using dynamic analysis to enhance graph-based features generated from static analysis for IoT botnet detection. Experimental results show that this approach outperforms existing methods in terms of accuracy and complexity.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Automation & Control Systems
Mohamed Abdel-Basset, Victor Chang, Hossam Hawash, Ripon K. Chakrabortty, Michael Ryan
Summary: This article presents a forensics-based deep learning model, Deep-IFS, for intrusion detection in IIoT traffic. By utilizing local gated recurrent unit and multihead attention layers to learn local and global representations, and deploying and training the model in a fog computing environment, it effectively handles large-scale IIoT traffic data and achieves good distributed processing results.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Automation & Control Systems
Jing Long, Wei Liang, Kuan-Ching Li, Yehua Wei, Mario Donato Marino
Summary: In this study, a semi-supervised ladder network model is proposed for intrusion detection in the Industrial IoT, taking into account the security issues and the challenge of limited labeled data. This model considers the manifold distribution of high-dimensional data and incorporates a manifold regularization constraint in the decoder. Additionally, a random attention-based data fusion approach is proposed to generate global features for intrusion detection. Experimental results on the CIC-IDS2018 dataset show that the proposed approach achieves a lower false alarm rate and is time efficient for training.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Computer Science, Information Systems
Jawad Ahmad, Syed Aziz Shah, Shahid Latif, Fawad Ahmed, Zhuo Zou, Nikolaos Pitropakis
Summary: The Industrial Internet of Things (IIoT) increases efficiency and productivity in industrial environments by integrating smart sensors and devices with the internet. This paper proposes a fast and reliable attack detection scheme for IIoT using a Deep Random Neural Network (DRaNN) trained with hybrid particle swarm optimization and sequential quadratic programming. Experimental results demonstrate the promising performance of the proposed design in various configurations.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Segun I. Popoola, Bamidele Adebisi, Ruth Ande, Mohammad Hammoudeh, Aderemi A. Atayero
Summary: The study developed a memory-efficient DL method, LS-DRNN, leveraging the combined advantages of LAE, SMOTE, and DRNN, and evaluated its effectiveness using the Bot-IoT dataset. Results showed that the application of LAE and SMOTE methods significantly improved classification performance in minority classes, and the LS-DRNN model outperformed state-of-the-art models.
Article
Computer Science, Interdisciplinary Applications
Xianyu Zhang, Xinguo Ming
Summary: With the development of industrial Internet environment and intelligent technology, enterprises are focusing more on system platform, information sharing, network collaboration, personalized customization and service recommendation in designing, implementing and operating Industrial Internet Platforms (IIP). However, there is a lack of a comprehensive framework for studying the high-level planning of IIP implementation and few studies on the detailed path and steps of IIP implementation in specific industries. The research aims to study the general model, reference architecture, service evaluation index system, implementation path and application verification for IIP to provide guidance for government and industry in planning, designing, implementing and promoting IIP.
COMPUTERS & INDUSTRIAL ENGINEERING
(2021)
Article
Computer Science, Information Systems
Jun Cai, Qi Wang, Jianzhen Luo, Yan Liu, Liping Liao
Summary: Efficient anomaly detection methods are urgently needed in the Industrial Internet of Things (IIoT) to prevent attacks in the application layer. The proposed CapBad detector utilizes a phase-aware hidden semi-Markov model (pHSMM) to model industrial control protocol packets and automatically learn payload characteristics, along with employing a probabilistic suffix tree to analyze contextual similarity. Results show that CapBad has excellent performance in detecting abnormal packets in the application layer.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Information Systems
Louai A. Maghrabi, Ibrahim R. Alzahrani, Dheyaaldin Alsalman, Zenah Mahmoud Alkubaisy, Diaa Hamed, Mahmoud Ragab
Summary: Artificial intelligence has gained significant attention in the cybersecurity field of Industry 4.0 and has shown immense benefits in various applications. This paper focuses on a deep learning-based method for cyberattack detection and classification and demonstrates its superior performance through extensive simulations.
Article
Automation & Control Systems
Zixiao Zhao, Qinghe Du, Houbing Song
Summary: In this article, a learning network is proposed to timely discover intrusion in the fifth generation network for Industrial internet of things (IoT), and can identify two types of intrusion. By extracting traffic load information from the states (success, collision, and idle) of access resources observed at media access control and physical layers, the learning network can effectively capture the number of active devices, provide reasonable prediction using history records, and achieve more accurate detection compared with baseline approaches.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Computer Science, Information Systems
Cristina Paniagua, Jerker Delsing
Summary: The Internet of Things (IoT) is gaining popularity in industrial applications, requiring support from flexible and scalable systems. With the increasing number of available frameworks, selecting a suitable one for industrial applications has become difficult. Therefore, researching the characteristics of each framework to simplify the selection process is crucial.
IEEE SYSTEMS JOURNAL
(2021)
Article
Green & Sustainable Science & Technology
Hani Alshahrani, Attiya Khan, Muhammad Rizwan, Mana Saleh Al Reshan, Adel Sulaiman, Asadullah Shaikh
Summary: The Industrial Internet of Things (IIoT) is the use of IoT in industrial management, linking and synchronizing machines and devices through software programs and third platforms to improve productivity. Despite the benefits, security remains a major concern due to the lack of reliable security mechanisms and the magnitude of security features. Attacks exploiting vulnerabilities in IIoT networks have caused financial losses, reputational damage, and theft of important information. This paper proposes an SDN-based framework with machine learning techniques for intrusion detection in an industrial IoT environment, achieving an accuracy of 99.7% in detecting attacks.
Article
Engineering, Multidisciplinary
Hakan Can Altunay, Zafer Albayrak
Summary: The security of Industrial IoT (IIoT) networks is crucial, and it is important to develop Intrusion Detection Systems (IDS) to protect them against cyber threats and intrusions. In this study, three different models were proposed to detect intrusions in the IIoT network using deep learning architectures (Convolutional Neural Network, Long Short Term Memory, and CNN + LSTM), and they were validated on the UNSW-NB15 and X-IIoTID datasets. The results showed that the hybrid CNN + LSTM model achieved the highest accuracy for intrusion detection in both datasets.
ENGINEERING SCIENCE AND TECHNOLOGY-AN INTERNATIONAL JOURNAL-JESTECH
(2023)
Article
Computer Science, Information Systems
Marwa Keshk, Nour Moustafa, Elena Sitnikova, Benjamin Turnbull
Summary: This paper studies the role of big data component analysis in protecting sensitive information from illegal access. The technique of independent component analysis is used to transform CPS information while preserving data utility, and the results demonstrate that it is more effective than other privacy-preservation techniques.
Article
Computer Science, Information Systems
Muna AL-Hawawreh, Elena Sitnikova
Summary: Achieving security for brownfield IIoT systems is a significant challenge due to their legacy devices and integration with new IoT technologies. A new generic end-to-end IIoT security testbed is proposed, which is easily reproducible and configurable for testing various security scenarios. Experiments demonstrate the effectiveness of the testbed for operation and security testing compared to existing testbeds.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Artificial Intelligence
Muna Al-Hawawreh, Nour Moustafa, Jill Slay
Summary: This study presents a new threat intelligence framework to examine and model attacks on the CoAP protocol in IIoHT systems, introducing RDoS as a new threat and utilizing deep learning for real-time discovery of attack network behaviors. The experiment results show that the proposed discovery model outperforms other conventional machine learning algorithms in revealing RDoS and effectively protecting SmartSat-IIoHT networks.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Engineering, Electrical & Electronic
Muna Al-Hawawreh, Ibrahim Elgendi, Kumudu Munasinghe
Summary: The Internet of Things (IoT) is facing challenges related to data redundancy and energy consumption. To address this, we propose an AI-powered solution that utilizes autocorrelation and deep reinforcement learning to make smart decisions about transmitting data, thereby reducing data redundancy and minimizing sensor power consumption.
IEEE SENSORS JOURNAL
(2022)
Proceedings Paper
Computer Science, Information Systems
Ramtin Ranji, Elena Sitnikova, Frank den Hartog
Summary: Excessive usage of wirelessly networked equipment in densely populated areas can lead to performance degradation due to interference and traffic congestion. Enabling constructive collaboration between actors for spectrum sharing based on trading and consensus is a possible solution, but defining utility remains a complex issue that requires consideration of various factors.
2021 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS WORKSHOPS (ICC WORKSHOPS)
(2021)
Article
Computer Science, Information Systems
Muna Al-Hawawreh, Elena Sitnikova, Neda Aboutorab
Summary: This paper introduces a targeted ransomware detection model tailored for IIoT systems, utilizing Asynchronous Peer-to-Peer Federated Learning and Deep Learning techniques to effectively detect known and unknown attacks in these systems with their heterogeneous and distributed nature.
Article
Education & Educational Research
Keith F. Joiner, Leanne Rees, Britt Levett, Elena Sitnikova, Dijana Townsend
Summary: This research examined the learning environments of five university postgraduate subjects taught through distance education. Significant differences were found in the environment across most demographics, with older students showing preferences for more Involvement and Student Cohesiveness, while students with lower or average prior academic achievement demonstrated better environmental fit. The study highlights the importance of considering demographic factors in designing effective online learning environments.
LEARNING ENVIRONMENTS RESEARCH
(2021)
Article
Computer Science, Hardware & Architecture
Marwa Keshk, Elena Sitnikova, Nour Moustafa, Jiankun Hu, Ibrahim Khalil
Summary: The paper introduces a new privacy-preserving anomaly detection framework called PPAD-CPS, which protects confidential information and detects malicious observations in power systems and their network traffic. Experimental results show that the framework is more effective than four recent techniques and outperforms seven peer anomaly detection techniques in terms of detection rate, false positive rate, and computational time.
IEEE TRANSACTIONS ON SUSTAINABLE COMPUTING
(2021)
Article
Computer Science, Information Systems
Merna Gamal, Hala M. Abbas, Nour Moustafa, Elena Sitnikova, Rowayda A. Sadek
Summary: The paper introduces a new IDS based on Few-Shot Deep Learning, known as CNN-IDS, which automatically identifies zero-day attacks and protects IoT systems. By utilizing a filtered Information Gain method and a one-dimensional CNN algorithm, the proposed model enhances attack detection rates.
CMC-COMPUTERS MATERIALS & CONTINUA
(2021)
Article
Computer Science, Information Systems
Marwa Keshk, Benjamin Turnbull, Elena Sitnikova, Dinusha Vatsalan, Nour Moustafa
Summary: Cyber-Physical Systems (CPS) are crucial for global critical infrastructure and are vulnerable to Advanced Persistent Threats (APTs), necessitating the development of efficient privacy-preserving techniques. This paper provides a comprehensive review of current privacy-preserving techniques for protecting CPS systems and data from cyber attacks, discussing the importance of privacy preservation and CPS components.
Article
Education & Educational Research
Keith. F. Joiner, Leanne Rees, Britt Levett, Elena Sitnikova, Dijana Townsend
Summary: This research examined the effectiveness of structured peer critiquing for students with different levels of prior achievement in postgraduate courses delivered online. It found that structured peer critiquing was more effective for students with lower or average prior achievement, and forums were a more subtle means of structuring critiquing and engagement than direct exchange. Care is needed to ensure online debate is incisive.
STUDIES IN HIGHER EDUCATION
(2021)
