Data Driven Security for Industrial IoT and CPS

Erscheinungsjahr 2027
ISBN: 978-3-032-44830-9
Verlag: Springer

Buch, Englisch, Format (B × H): 155 mm x 235 mm

Reihe: Studies in Systems, Decision and Control

This book presents an in-depth analysis of how advanced data analytics, deep learning, and generative artificial intelligence (AI) can be used to strengthen the security, reliability, and resilience of Industrial Internet of Things (IIoT) devices and Cyber-Physical Systems (CPS). As IoT devices become more prevalent across various industries such as manufacturing, healthcare, agriculture, and transportation, they enable greater efficiency, automation, and data-driven decision-making. However, this book introduces new security challenges that require innovative solutions. This book focuses on using data-driven models and algorithms to address the security concerns inherent in IIoT environments, where the complexity and diversity of devices and systems complicate the application of traditional cybersecurity measures. Key technologies explored in the book include deep learning, explainable AI (XAI), and generative AI, which offer dynamic, adaptive approaches to detecting and mitigating cyber threats. By leveraging large datasets from IoT devices, machine learning algorithms can identify abnormal patterns in network traffic, sensor data, and device behavior, helping to prevent attacks like unauthorized access, DDoS, and man-in-the-middle attacks. This book discusses the unique security issues faced by IoT systems, such as device vulnerabilities, weak encryption, and insufficient authentication mechanisms. It emphasizes the need for tailored security models that consider the limitations of IoT devices while providing robust protection. The integration of Intrusion Detection Systems (IDS) powered by machine learning is presented as a critical component in real-time threat detection and response. A significant portion of the book highlights the use of generative AI, particularly Generative Adversarial Networks (GANs), for simulating potential attack scenarios and predicting future threats. This proactive approach allows security teams to better understand risks and strengthen defense mechanisms.

Zielgruppe

Research

Autoren/Hrsg.

Guezzaz, Azidine

Ravi, Vinayakumar

Fachgebiete

• Interdisziplinäres Wissenschaften Wissenschaften: Forschung und Information Kybernetik, Systemtheorie, Komplexe Systeme
• Technische Wissenschaften Elektronik | Nachrichtentechnik Nachrichten- und Kommunikationstechnik
• Mathematik | Informatik EDV | Informatik Computerkommunikation & -vernetzung
• Technische Wissenschaften Elektronik | Nachrichtentechnik Elektronik

Inhaltsverzeichnis

Open-Source Honeypot Solutions for Cybersecurity: A Benchmark Analysis of Architectures, Capabilities, and Deployment Scenarios.- Design of a Secure Layered OT-IT Architecture for Smart Grids Based on the Purdue Model and IEC 62443.- Toward Trustworthy Federated Healthcare IoT: A Layered Edge Blockchain Architecture.- AI-Driven Real-Time Threat Detection and Autonomous Response Framework for Secure Industrial IoTand Cyber-Physical Systems.- A Multi-Protocol ICS Dataset and Unsupervised Behavior Audit for Machine Learning-Based SDN Security.- Intrusion Detection in IoMT Using Ensemble Boosting.- Adaptive Intrusion Detection in IoT Networks with Proximal Policy Optimization.- Machine Learning-Based Anomaly Detection for Computer Network Traffic .- Privacy-Preserving Federated Learning for Heterogeneous Agricultural IoT Data: Techniques, Trade-offs, and a Decision Framework.- I2ACS: Toward an Immune-Inspired Action and Access Control Security System.- Zero-Trust Mutual Authentication for IoMT Using PUF-Derived ECC Keys and Blockchain.- From Deception to Detection, A Unified LLM-Based Honeypot and Splunk SIEM Framework for Enterprise Information System Survivability.- Privacy-Preserving IDS for Cyber-Physical Systems: A Federated Learning and Knowledge Distillation Approach.- A Two-Tier Hard-Threshold/ML Architecture for Real-Time Multi-Protocol ICS Intrusion Response in SDN.- Adaptive Federated DDoS Detection with Concept Drift Awareness in SDN-IoT Networks.- AutoMOOC 3.0: A Blockchain-Enabled Multi-Agent Architecture for Autonomous, Personalized, and Trustworthy Online Learning.- HCCF-FS: A Two-Stage Feature Reduction Framework for Lightweight AI-Based Network Intrusion Detection Systems.- EdgecareAI: A Plug-and-Play, Privacy-Preserving Edge Intelligence Platform for Real-Time Multimodal Healthcare Diagnostics.- Deep Learning-Driven Multimodal Emotion Recognition by Integrating EEG Signals, Thermal and Digital Facial Images.- 3D Trichoscopic Pattern Analysis and Mobile-Based Hair Disorder Assessment for Young Adults.- Hypergraph Spectral Analysis for Coordinated Attack Detection in IIoT Networks.- UAV-SDIW: Secure Digital Image Watermarking for Optimized Aerial Image Transmission.- Generative AI for Threat Simulation and Prediction in Industrial IoT and Cyber-Physical Systems.- Hybrid Federated Learning and Blockchain Framework for Secure and Scalable IoT Threat Detection.- Federated Deep Learning with Explainable AI for Adaptive Intrusion Detection in Industrial IoT Cyber-Physical Systems.- Convolutional Bidirectional Long Short-Term Memory Networks for Resilient Intrusion Detection within IIoT Cyber-Physical System Infrastructures.

Über Autor(innen)

Azidine Guezzaz is currently an associate professor of computer science and mathematics at Technology Higher School Essaouira at Cadi Ayyad Marrakech University Morocco. He is a member in Laboratory Mathematics, Computer Science, and Modeling of Complex Systems (MIMSC). He received his Ph.D. from IbnZohr University Agadir, Morocco, in 2018. His research interest is computer security, cryptography, artificial intelligence, intrusion detection, and smart cities. He is also a reviewer of various scientific journals. Cyber security, big data analytics, Network Security Computer, Networking, Network Architecture, Cryptography, Routing, Intrusion Detection Intrusion Prevention, Applied Artificial Intelligence, Data Mining, Data Warehouse, Advanced Machine Learning, Business Intelligence.

Vinayakumar Ravi is an assistant research professor at Center for Artificial Intelligence, Prince Mohammad Bin Fahd University, Khobar, Saudi Arabia. His previous position was a postdoctoral research fellow in developing and implementing novel computational and machine learning algorithms and applications for big data integration and data mining with Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA. His current research interests include applications of data mining, artificial intelligence, machine learning (including deep learning) for biomedical informatics, cyber security, image processing, and natural language processing. He has more than 100 research publications in reputed IEEE conferences, IEEE Transactions, and Journals. His publications include prestigious conferences in the area of cyber security, like IEEE S&P and IEEE Infocom. Dr. Ravi has received a full scholarship to attend Machine Learning Summer School (MLSS) 2019, London.