Emerging Trends and Techniques for Integrated IoT Systems towards Ubiquitous Artificial Intelligence

Introduction:

Emerging AI techniques like large language models and embodied AI have empowered the degree of intelligence and usability of Internet-of-Thing (IoT) systems. Accordingly, fundamental requirements are raised in the adoption and implementation of integrated IoT systems to provide ubiquitous data and computing resources.

Integrated IoT systems have invariably permeated various facets of human society, with applications ranging from fitness apps that leverage sensing data to provide personalized exercise plans to smart grids that optimize city-wide electric power supply through advanced metering data processing. Nevertheless, current IoT systems often operate in isolated domains, where data collection and processing, decision-making, and service delivery are controlled by a limited number of devices or service providers. These isolated silos hinder the widespread adoption of IoT technologies and fall short of the original vision for IoT — to create comprehensive, seamless, and integrated connections between the physical and digital worlds. To overcome this barrier, it is crucial for academia, industry and government to collectively explore innovative methods and techniques to integrate these IoT systems. One example would be integrating data from smart grids and smart traffic systems with smart home systems to provide more efficient and timely services to residents.

Ideally all devices, data, and resources should be shared among participants to support numerous services efficiently. However, the heterogeneous ownership and interconnections of service providers make this ideal scenario challenging to achieve. Obstacles exist at all levels, from the deployment of bottom-layer devices to data sharing across heterogeneous systems. Devices owned by different entities must be managed wisely to balance efficiency and benefits.

Data sharing can occur at multiple levels, with service providers exchanging information or data contributors uploading content to various IoT systems. The ownership, profitability, privacy and authentication of this data need to be carefully managed. Additionally, the extent of how IoT ecosystems will benefit their participants, including service providers and society as a whole, should be considered.

Given the promising potential of integrated IoT ecosystems towards ubiquitous AI and the significant challenges they present, research in this area is still in its infancy. There have been pioneering efforts in areas like crowdsensing and edge computing, where data and computing resources are shared across different entities. However, these efforts address only a small portion of the broader challenge of integrating IoT systems and often focus on limited or impractical scenarios. This warrants a need to bride the gap between current solutions and the envisioned scope of integrated IoT ecosystems is both meaningful and imperative.

To that end, this special issue serves as both a repository of novel solutions and ongoing research focused on integrated IoT ecosystems and a platform to highlight the scientific and technological challenges and opportunities that may influence the future development of IoT. Additionally, this special issue will function as a workshop where experts from academia and industry can share their pioneering insights and ideas in this research field.

We welcome novel contributions, including but not restricted to following topics. 

Topics covered include, but are not limited to:

• Architecture design and models for integrated IoT systems for ubiquitous AI
• Communication techniques in integrated IoT systems for ubiquitous AI
• Intelligent-service-centric network modeling for integrated IoT systems
• Device deployment and management in integrated IoT systems for ubiquitous AI
• Distributed data processing in integrated IoTsystems for ubiquitous AI
• Resource allocation and task offloading in integrated IoTsystems for ubiquitous AI
• Distributed machine learning in integrated IoT systems
• New design of integrated IoTsystems for large language models
• Data privacy and security for integrated IoT systems for ubiquitous AI
• Blockchains and authentication in integrated IoT systems for ubiquitous AI
• Fairness and rationality among participants of integrated IoT systems for ubiquitous AI
• Crowdsensing for heterogeneous and integrated IoT systems for ubiquitous AI
• Edge computing for integrated IoT systems for ubiquitous AI
• Lightweighted deep learning models for intelligent services in integrated IoT systems
• New design of integrated IoTsystems for embodied AI
• Investigation on functionalities and benefits of integrated IoT systems for ubiquitous AI

 Important dates:

• Submission deadline: February 15, 2025
• First notification: April 15, 2025
• Revised version deadline: May 15, 2025
• Final notification: July 15, 2025
• Publication date: August, 2025

Submission instructions: 

Please read the Guide for Authors before submitting. All articles should be submitted online via the editorial management system; please select article type Integrated IoT for AI.

Guest editors:

• Zhipeng Cai, IEEE Fellow, Georgia State University, USA. Email: zcai@gsu.edu
• Suparna De, University of Surrey, UK, Email: s.de@ieee.org
• Liran Ma, Miami University, USA. Emai: liranma@gmail.com
• Daehee Seo, Sangmyung University, Korea, Email: daehseo@smu.ac.kr