LLM-Empowered IoT for 6G Networks: Architecture, Challenges, and Solutions
- 30 June 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Magazine
- Vol. 8 (6) , 34-41
Abstract
The Internet of Things (IoT) in the sixth generation (6G) network is envisioned to evolve towards intelligence, ubiquity, and self-optimization. Large language models (LLMs) have demonstrated remarkable generalization capabilities across diverse domains, including natural language processing, computer vision, etc. In this article, we propose an LLM-empowered IoT architecture for 6G networks to achieve intelligent autonomy and facilitate diversified IoT applications. On one hand, LLM-based solutions are tailored to satisfy the quality of service requirements of IoT applications, i.e., LLM for 6G IoT. On the other hand, LLMs are expected to be deployed in IoT environments through edge fine-tuning methods and edge inference techniques to support IoT applications, i.e., LLM on 6G IoT. Furthermore, we propose a memory-efficient split federated learning framework for LLM fine-tuning on heterogeneous IoT devices to alleviate memory consumption on both IoT devices and the edge server. Finally, a case study is presented, followed by a discussion about open issues.Keywords
All Related Versions
Funding Information
- Pengcheng Laboratory Major Key Project (2025QYA002)
- Pengcheng Laboratory Major Key Project (PCL2023AS1-5)
- Pengcheng Laboratory Major Key Project (PCL2024A01)
- National Natural Science Foundation of China (62201311)
- National Natural Science Foundation of China (62201071)
- National Natural Science Foundation of China (62192712)
- Young Elite Scientists Sponsorship Program by CAST (2023QNRC001)
This publication has 9 references indexed in Scilit:
- Unveiling the Vulnerability of Private Fine-Tuning in Split-Based Frameworks for Large Language Models: A Bidirectionally Enhanced Attack2024
- A Survey on Evaluation of Large Language ModelsACM Transactions on Intelligent Systems and Technology, 2024
- An Adaptive Compression and Communication Framework for Wireless Federated LearningIEEE Transactions on Mobile Computing, 2024
- Split Learning Over Wireless Networks: Parallel Design and Resource ManagementIEEE Journal on Selected Areas in Communications, 2023
- FedBERT: When Federated Learning Meets Pre-trainingACM Transactions on Intelligent Systems and Technology, 2022
- AI-Native Network Slicing for 6G NetworksIEEE Wireless Communications, 2022
- Holistic Network Virtualization and Pervasive Network Intelligence for 6GIEEE Communications Surveys & Tutorials, 2021
- Delay-Optimal Distributed Edge Computing in Wireless Edge Networks2020
- CARER: Contextualized Affect Representations for Emotion Recognition2018