Article
Computer Science, Hardware & Architecture
Rohit Kumar Gupta, Shashwat Mahajan, Rajiv Misra
Summary: This paper investigates the problem of network slicing in the Industrial Internet of Things and utilizes deep reinforcement learning techniques for resource allocation and optimization of environmental requirements. By optimizing objectives such as system throughput, spectral efficiency, service level agreement, low-power transmission, and transmission delay, the rewards of the Industrial Internet of Things are maximized.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Computer Science, Information Systems
Cheng Qian, Wei Yu, Chao Lu, David Griffith, Nada Golmie
Summary: This article investigates the impact of insufficient data on machine learning model training and proposes a framework using GAN and continuous learning to address the issue. The experimental results demonstrate that new data greatly improves model accuracy, and the proposed defensive mechanism safeguards the model learning process.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Information Systems
Jiadai Wang, Jiajia Liu
Summary: This article investigates attacks against forwarding nodes in software-defined IIoT, proposing a DRL-based general attack tolerance scheme, and using GAN to generate realistic network traffic for experimental verification.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Haodong Lu, Miao Du, Kai Qian, Xiaoming He, Kun Wang
Summary: This paper proposes a framework for detecting anomalies in industrial robotic sensors, and improves the detection accuracy by using an improved GANs to generate fake anomaly samples. The effectiveness of the framework is demonstrated through extensive experiments.
IEEE SENSORS JOURNAL
(2022)
Article
Computer Science, Artificial Intelligence
Yirui Zhou, Mengxiao Lu, Xiaowei Liu, Zhengping Che, Zhiyuan Xu, Jian Tang, Yangchun Zhang, Yan Peng, Yaxin Peng
Summary: Generative adversarial imitation learning (GAIL) treats imitation learning (IL) as a distribution matching problem between expert policy and learned policy. This paper focuses on the generalization and computational properties of policy classes. It is proven that GAIL can ensure generalization when policies are well controlled. By incorporating distributional reinforcement learning (RL) into GAIL, the greedy distributional soft gradient (GDSG) algorithm is proposed to solve GAIL. GDSG has advantages including alleviating Q-value overestimation problem and improving policy performance through sufficient exploration, as well as attaining a sublinear convergence rate to a stationary solution. Comprehensive experimental verification in MuJoCo environments demonstrates that GDSG outperforms previous GAIL variants in mimicking expert demonstrations.
Article
Computer Science, Information Systems
Mohammad Mehedi Hassan, Md Rafiul Hassan, Shamsul Huda, Victor Hugo C. de Albuquerque
Summary: The article addresses the challenging problem of trust-boundary protection in Industrial Internet of Things environments and proposes a cooperative data generator based on a downsampler-encoder to better capture the distribution of attack models. Experimental results demonstrate that this approach outperforms conventional deep learning and other ML techniques in terms of robustness against adversarial/noisy examples in the IIoT environment.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Information Systems
Merim Dzaferagic, Nicola Marchetti, Irene Macaluso
Summary: This article addresses the issue of reliability in the Industrial Internet of Things (IIoT) when there are missing sensor measurements due to network or hardware problems. The proposed solution involves using a generative model, specifically generative adversarial networks (GANs), to impute the missing data and enhance the performance of the fault detection and classification modules. The evaluation of the approach using real-world data shows that the GAN-imputed data can effectively mitigate the impact of missing measurements on the fault detection and classification modules, even for critical sensors.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Information Systems
Xiaoding Wang, Sahil Garg, Hui Lin, Jia Hu, Georges Kaddoum, Md Jalil Piran, M. Shamim Hossain
Summary: This paper proposes a reliable anomaly detection strategy for Industrial Internet of Things (IIoT) using federated learning. By training local models with deep reinforcement learning algorithm and introducing privacy leakage degree and action relation, the detection accuracy can be greatly improved, achieving privacy preservation.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Automation & Control Systems
Xiaokang Zhou, Yiyong Hu, Jiayi Wu, Wei Liang, Jianhua Ma, Qun Jin
Summary: In this article, we propose a distribution bias aware collaborative GAN (DB-CGAN) model for imbalanced deep learning in industrial IoT. By introducing a complementary classifier and a data augmentation algorithm, the model can effectively handle the distribution bias between the generated data and the original data, resulting in significantly improved classification accuracy.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Automation & Control Systems
Rahim Taheri, Mohammad Shojafar, Mamoun Alazab, Rahim Tafazolli
Summary: Fed-IIoT is a robust architecture for detecting Android malware applications in IIoT, utilizing federated learning on both participant and server sides to enhance security and privacy protection.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Hardware & Architecture
Shanshuo Ding, Liang Kou, Ting Wu
Summary: This paper proposes a metaverse network intrusion detection model based on the Internet of Things, which is a hybrid intrusion detection model based on Generative Adversarial Network (GAN). The model achieves Nash equilibrium by training the GAN generator and discriminator alternately based on gradient penalty Wasserstein distance, and enriches the dataset by generating data indistinguishable by the discriminator to improve its imbalance. In addition, the paper combines deep autoencoder (DAE) and random forest (RF) algorithms to optimize the model training process and improve training efficiency, achieving high accuracy in binary and multiple classification experiments.
MOBILE NETWORKS & APPLICATIONS
(2022)
Article
Chemistry, Analytical
Mattia Antonini, Miguel Pincheira, Massimo Vecchio, Fabio Antonelli
Summary: Industrial assets use multiple sensing devices to monitor physical parameters and detect potential failures through anomaly detection with machine learning tools. This paper proposes an adaptable anomaly detection system using IoT, edge computing, and Tiny-MLOps in extreme industrial environments like submersible pumps. The system runs on an IoT sensing Kit near the data source, and the anomaly detection model uses the isolation forest algorithm for fast training and detection. It also employs blockchain technology for transparent anomaly storage.
Article
Automation & Control Systems
Suparna De, Maria Bermudez-Edo, Honghui Xu, Zhipeng Cai
Summary: This article reviews the latest technologies and applications of deep generative models (DGMs) in industrial Internet of Things (IIoT), categorizing the discussed works into application areas such as anomaly detection, trust-boundary protection, network traffic prediction, and platform monitoring. After analyzing existing implementations, challenges that need to be addressed are identified, and potential research directions are proposed.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Engineering, Civil
Faisal Naeem, Sattar Seifollahi, Zhenyu Zhou, Muhammad Tariq
Summary: The key challenge in CIoV systems is designing a smart agent for ultra-reliable low latency communication, and we propose a scheduling algorithm based on SDN that uses GAN-DDQN to learn the action-value distribution for intelligent transmission scheduling.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2021)
Article
Automation & Control Systems
Yen-Ting Chen, Chia-Yi Hsu, Chia-Mu Yu, Mahmoud Barhamgi, Charith Perera
Summary: This article discusses the potential of using deep learning methods to improve industrial technologies and introduces synthetic datasets and information network structures as a solution. Through empirical research, it is found that classifiers generated by private information network structures are more accurate.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Engineering, Electrical & Electronic
Mohammed Jouhari, Khalil Ibrahimi, Hamidou Tembine, Mohammed Benattou, Jalel Ben Othman
INTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS
(2019)
Article
Computer Science, Information Systems
Mohammed Jouhari, Abdulla Khalid Al-Ali, Emna Baccour, Amr Mohamed, Aiman Erbad, Mohsen Guizani, Mounir Hamdi
Summary: The article proposes a method for distributing DNNs within UAVs to enable data classification on resource-constrained devices, and avoid delays introduced by server-based solutions. The simulation results showed that the optimization solution outperformed existing approaches.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Hafsa Benaddi, Khalil Ibrahimi, Abderrahim Benslimane, Mohammed Jouhari, Junaid Qadir
Summary: This paper proposes a DRL-based IDS for network traffics using MDP and analyzes the IDS behavior through modeling the interaction between the IDS and attacker players. The proposed DRL-IDS outperforms existing models in terms of detection rate, accuracy, and false alarms reduction.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2022)
Proceedings Paper
Automation & Control Systems
Yaya Etiabi, Mohammed Jouhari, Andreas Burg, El Mehdi Amhoud
Summary: In this paper, a novel LoRa RSSI fingerprinting approach considering the spreading factor (SF) is proposed. Performance evaluation using deep neural networks and deep reinforcement learning models shows significant improvement in localization accuracy for LoRa networks.
2023 IEEE 97TH VEHICULAR TECHNOLOGY CONFERENCE, VTC2023-SPRING
(2023)
Proceedings Paper
Computer Science, Information Systems
Hafsa Benaddi, Mohammed Jouhari, Khalil Ibrahimi, Abderrahim Benslimane, El Mehdi Amhoud
Summary: This paper proposes a method to improve the training process of intrusion detection systems (IDS) using conditional generative adversarial networks (cGANs) to handle unbalanced data and missing specific class samples. By applying adversarial training in experiments, the results show that this method can significantly improve the detection accuracy of theft attacks.
2022 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM 2022)
(2022)
Proceedings Paper
Telecommunications
Boutaina Jebari, Khalil Ibrahimi, Mohammed Jouhari, Mounir Ghogho
Summary: This paper analyzes the profitability of selfish mining attacks using game theory and the monetary award as the evaluation criterion. It models the interactions between mining pools and discusses possible outcomes of the game, highlighting scenarios where the system could be compromised. To the best of our knowledge, this is the first work that models the selfish mining attack as a stochastic game.
IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC 2022)
(2022)
Article
Acoustics
Hamza Zradgui, Khalil Ibrahimi
Summary: This research proposes an energy scarcity-aware routing protocol for efficient underwater wireless sensor networks, aiming to improve energy efficiency and prolong network life. The proposed protocol shows significant improvements in performance parameters compared to other protocols.
Proceedings Paper
Engineering, Electrical & Electronic
Hamza Zradgui, Mohammed Jouhari, Khalil Ibrahimi
Summary: This research focuses on underwater wireless sensor networks (UWSNs) and proposes introducing autonomous underwater vehicles (AUVs) with low energy consumption and a clustering protocol to enhance network efficiency.
2021 INTERNATIONAL CONFERENCE ON ADVANCED COMMUNICATION TECHNOLOGIES AND NETWORKING (COMMNET'21)
(2021)
Proceedings Paper
Computer Science, Information Systems
Hafsa Benaddi, Mohammed Jouhari, Khalil Ibrahimi, Abderrahim Benslimane
Summary: This study utilizes signaling game approach to address the impact of double-spending attacks on critical IoT applications like Bitcoin, with simulations showing that signaling game can reduce this impact.
2021 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM)
(2021)
Proceedings Paper
Computer Science, Information Systems
Hafsa Benaddi, Khalil Ibrahimi
2020 8TH INTERNATIONAL CONFERENCE ON WIRELESS NETWORKS AND MOBILE COMMUNICATIONS (WINCOM 2020)
(2020)
Proceedings Paper
Computer Science, Theory & Methods
Khalil Ibrahimi, Ouafa Ould Cherif, Mohammed ELkoutbi, Imane Rouam
2019 INTERNATIONAL CONFERENCE ON WIRELESS NETWORKS AND MOBILE COMMUNICATIONS (WINCOM)
(2019)
Article
Computer Science, Information Systems
Mohammed Jouhari, Khalil Ibrahimi, Hamidou Tembine, Jalel Ben-Othman