Privacy-Preserving and Secure Federated Learning for IoT and Cyber-Physical Systems

Dear Colleagues,

The rapid proliferation of Internet of Things devices and cyber–physical systems in domains such as smart manufacturing, intelligent transportation, healthcare, and critical infrastructure is generating unprecedented volumes of distributed data at the network edge. Harnessing this data is essential for building intelligent, responsive, and resilient services. However, centralizing such data raises serious concerns regarding privacy, security, regulatory compliance, and communication overhead.

Federated learning has emerged as a promising paradigm that enables collaborative model training across devices and systems without sharing raw data. It thus offers a natural fit for privacy-aware and communication-efficient intelligence in large-scale IoT deployments and cyber–physical environments. At the same time, federated learning in these settings faces unique challenges, including heterogeneous devices and networks, non-independent and unbalanced data distributions, real-time constraints, attacks on models and protocols, and stringent requirements on safety and reliability.

To address these challenges, there is a pressing need for privacy-preserving and secure federated learning methods that can be deployed over constrained IoT platforms and safety-critical cyber–physical systems. Such methods must integrate advances in cryptography, differential privacy, secure and robust aggregation, trusted execution, and anomaly and attack detection, as well as system-level co-design with edge, fog, and cloud computing infrastructures.

This Special Issue focuses on privacy-preserving and secure federated learning for IoT and cyber–physical systems. It seeks original research contributions on theory, algorithms, architectures, implementations, and real-world applications that enable trustworthy, efficient, and scalable collaborative intelligence at the edge.

Topics of interest include, but are not limited to, the following:

• Privacy-preserving federated learning frameworks for IoT and cyber–physical systems;
• Secure aggregation and communication protocols for federated learning over resource-constrained devices;
• Differential privacy mechanisms and privacy budget management in federated learning;
• Robust and attack-resilient federated learning against poisoning, inference, and backdoor attacks;
• Federated learning for safety-critical cyber–physical systems such as industrial control, autonomous driving, and smart grids;
• System and hardware co-design for efficient federated learning on edge and embedded platforms;
• Energy-aware and communication-efficient federated learning for large-scale IoT deployments;
• Trust management, reputation, and incentive mechanisms in federated learning ecosystems;
• Integration of federated learning with edge, fog, and cloud architectures in cyber–physical infrastructures;
• Formal verification, reliability analysis, and safety assurance for federated learning in control loops;
• Benchmarking, performance evaluation, and real-world testbeds for federated learning in IoT environments;
• Standards, governance models, and regulatory compliance for privacy-preserving federated learning;
• Application case studies of federated learning in smart healthcare, intelligent transportation, industrial IoT, and smart cities;
• Human-centric, ethical, and socio-economic aspects of federated learning in networked cyber–physical systems.

This Special Issue aims to provide a timely forum for disseminating novel ideas, methodologies, and practical experiences that will shape the next generation of trustworthy and intelligent IoT and cyber–physical systems. High-quality submissions from both academia and industry are warmly invited.

Dr. Zhe Sun
Dr. Yuanyuan He
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 250 words) can be sent to the Editorial Office for assessment.

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. Future Internet 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 1600 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.

• federated learning
• Internet of Things
• cyber–physical systems
• privacy-preserving machine learning
• security and privacy
• edge and fog computing
• distributed artificial intelligence
• differential privacy
• secure aggregation
• adversarial robustness
• trustworthy AI