A Survey of Foundation Models for IoT: Taxonomy and Criteria-Based Analysis

2025-06-13

AI Summary

Existing IoT foundation model research is predominantly task-specific, lacking cross-domain comparability and principled guidance for adapting to novel tasks. Method: We propose the first taxonomy framework centered on four core performance objectives—efficiency, context awareness, security, and privacy preservation—overcoming the limitations of conventional task- or architecture-based classifications. Through bibliometric analysis, multi-dimensional evaluation, cross-domain technical mapping, and systematic literature review, we integrate key techniques including self-supervised learning, lightweight inference, federated learning, and trustworthy AI to construct a comprehensive technology map spanning mainstream IoT scenarios. Contribution/Results: The framework explicitly identifies representative methods, standardized benchmarks, and applicability boundaries for each objective, significantly enhancing objectivity in method comparison and practicality in novel task design. It serves as an actionable, reference-ready guide for researchers and practitioners in model selection and system design.

Technology Category

• Application Domains: Internet of Things, Sensor Networks & Smart Cities
• Machine Learning: Learning on the Edge & Model Compression
• Cognitive Modeling & Cognitive Systems: Other Foundations of Cognitive Modeling & Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Foundation models have gained growing interest in the IoT domain due to their reduced reliance on labeled data and strong generalizability across tasks, which address key limitations of traditional machine learning approaches. However, most existing foundation model based methods are developed for specific IoT tasks, making it difficult to compare approaches across IoT domains and limiting guidance for applying them to new tasks. This survey aims to bridge this gap by providing a comprehensive overview of current methodologies and organizing them around four shared performance objectives by different domains: efficiency, context-awareness, safety, and security&privacy. For each objective, we review representative works, summarize commonly-used techniques and evaluation metrics. This objective-centric organization enables meaningful cross-domain comparisons and offers practical insights for selecting and designing foundation model based solutions for new IoT tasks. We conclude with key directions for future research to guide both practitioners and researchers in advancing the use of foundation models in IoT applications.

Problem

• Survey compares foundation models across IoT domains
• Organizes methods by efficiency, context-awareness, safety, security
• Provides insights for applying models to new IoT tasks

Innovation

• Foundation models reduce labeled data reliance
• Objective-centric organization enables cross-domain comparisons
• Survey reviews techniques for IoT performance objectives