Special Issue: Advances in Federated Learning for Secure Internet of Things

Theme and Scope

The advent of the Internet of Things (IoT) and smart devices greatly influences our everyday lives in several ways. This tremendous growth in IoT devices possess several benefits, and it potentially transforms the world. For example, a connected healthcare device provides a deeper and complete individual's health status in a more precise manner. But with all of its benefits, there comes the increased risk of security and privacy issues. This is because the IoT, in general, has a broader scope, and it has significant contributions to various streams such as business, manufacturing, transportation, home automation, and so on. The privacy and security risks are higher as they are closely related to consumer-oriented devices. Also, the number and diversity of consumer IoT devices are growing rapidly, and in the future, this will likely increase more, and it will function autonomously without human interventions. Besides, when converged with machine learning and data analytics, it provides proactive actions and offers more useful and interesting patterns to make appropriate decisions. Hence, it has become crucial to find out the best practices to preserve the security and privacy of IoT devices.

Federated learning, also called the next generation of artificial intelligence, is mainly introduced to revolve around the core idea called "data privacy." It is all about moving the computations to the data. With conventional artificial intelligence algorithms, the most prominent issue is inadequate data transaction practices. In contrast, the use of federated learning promises to solve most of the issues faced by conventional machine learning and artificial intelligence algorithms across the IoT networks. It creates machine learning models that make use of the datasets that are cross-platforms while offering the necessary measures to prevent data leaks and security threats. The use of federated learning enables the devices to learn from each other with the objective to train the models across the distributed channels consisting of the large number of devices functioning as the clients. This helped to resolve the privacy challenges as the raw data was not sent to the server itself. That is, the technique of moving computation to the data is the most powerful concept to build any intelligent system while protecting the privacy of any users associated with it. But however, the real-time implementation of federated learning requires more sophisticated tools. In addition, it poses challenges such as computing power, storage, speed and sophistication measures. More advanced research in this stream will significantly transform the IoT security for better and enhance the future of IoT applications in a more significant way than ever before. In this context, this special issue intends to bring out the advances in federated learning for IoT applications.

Probable themes include, but are not limited to:

  • Blockchain assisted federated learning for secure IoT
  • New innovative federated learning frameworks and architectures for IoT applications
  • IoT assisted smart networking with federated learning
  • Advances in federated learning for emerging IoT applications
  • Future use cases of federated learning for IoT from a security perspective
  • Federated learning for secure IoT: challenges, opportunities, and future research directions
  • Enabling technologies and frameworks for secure IoT using federated learning
  • Collaborative machine learning for decentralized IoT systems
  • Energy efficient secure federated learning for IoT
  • Distributed machine learning for emerging IoT applications

Instructions for Manuscripts

Each paper, written in English, the maximum words number in each paper should be below 8,000 words, including references and illustrations. More information can be found at https://jit.ndhu.edu.tw/. Before submission authors should carefully read over the journal’s Author Guidelines, which are located at https://jit.ndhu.edu.tw/about/submissions#authorGuidelines.

Prospective authors should submit an electronic copy of their complete manuscript through the journal Manuscript Tracking System at https://mc04.manuscriptcentral.com/internet-tech. When submitting papers, authors should specify that the manuscript is for Special Issue on “Advances in Federated Learning for Secure Internet of Things”. The manuscript template can be found at https://jit.ndhu.edu.tw/about/submissions#authorGuideline.


Important Dates:

Manuscript Submission Deadline Date:

05th June, 2022

First Notification Date:

07th August, 2022

Revised Papers Due Date:

08th October, 2022

Final notification Date:

07th December, 2022

Tentative Publication:

the 4th Quarter of 2023

Supervising Editor:

Dr. Wei-Che Chien

Department of Computer Science & Information Engineering, Dong Hwa University, Taiwan

E-mail: wcc@gms.ndhu.edu.tw


Guest Editors:

Dr. Oscar SanjuánMartínez

Universidad Internacional de la Rioja (UNIR), Spain.

E-mail: Oscar.sanjuan@ieee.org


Dr. Mariacristina Gallo

Dipartimento di Scienze Aziendali - Management & Innovation Systems,

University of Salerno, Italy

E-mail: mgallo@unisa.it


Dr. B. Cristina Pelayo García-Bustelo

Department of Computer Science, University of Oviedo, Spain

E-mail: crispelayo@uniovi.es