AI-Enabled Intrusion Detection and Secure Data Management in IoT-Driven Connected Environments

The rapid expansion of Internet of Things (IoT) deployments across healthcare, smart cities, industrial facilities, and consumer environments has introduced a security challenge that conventional rule-based approaches cannot adequately address. Connected devices operate under tight resource constraints, generate continuous streams of sensitive data, and communicate across heterogeneous wireless and edge networks, conditions that create broad and difficult-to-monitor attack surfaces.Machine learning (ML) and deep learning have demonstrated strong potential for detecting anomalies, classifying attack patterns, and enabling adaptive security responses without the latency of centralised controls. However, important gaps remain. Models trained on benchmark datasets often generalise poorly to real deployment environments. Resource-constrained edge nodes struggle to run computationally intensive detection pipelines. Privacy regulations, particularly in healthcare and medical IoT (IoMT), impose additional constraints on how security data can be shared or used for model training. Federated and privacy-preserving approaches offer partial answers, but lack standardised evaluation criteria and reproducible implementation frameworks. Blockchain-based data integrity mechanisms show promise for audit trails in sensitive environments but have not been widely validated in constrained IoT contexts.This Research Topic addresses these open problems by bringing together applied and theoretical contributions at the intersection of AI-driven security, IoT network management, and secure data governance. Particular attention is given to health-connected environments, where the stakes of a security failure (compromised medical records, disrupted device communication, unauthorised access to patient data) are especially high. Contributions are welcome from network security, machine learning, distributed systems, and health informatics, to produce results that are not only technically sound but implementable in real deployment settings.We welcome original research papers, reviews, methods papers, and perspectives addressing, but not limited to, the following themes:- Machine learning and deep learning architectures for IoT intrusion detection and attack classification- Anomaly detection in resource-constrained, heterogeneous IoT and wireless sensor networks- Federated and privacy-preserving security frameworks for distributed IoT environments- Secure data management, provenance, and integrity in the Internet of Medical Things (IoMT)- Trust, authentication, and access control models for heterogeneous IoT ecosystems- Edge-cloud security architectures for real-time threat monitoring and incident response- Blockchain and cryptographic methods for IoT data integrity and auditability- Vulnerability assessment, risk prediction, and resilience in health-connected IoT systems- Benchmark datasets, evaluation metrics, and reproducibility standards for IoT security research. Topic editors: Nishtha Kesswani, Lalit Garg, Frank Lin Source: https://www.frontiersin.org/research-topics/82155/ai-enabled-intrusion-detection-and-secure-data-management-in-iot-driven-connected-environments