Generalizable IoT Traffic Representations for Cross-Network Device Identification

2026-01-27

AI Summary

This work addresses the limited generalizability of existing IoT device identification methods, which often rely on labeled data and specific deployment environments. To overcome this, the authors propose a universal traffic representation learning framework that requires no task-specific fine-tuning. The approach leverages an unsupervised encoder–decoder architecture to learn compact flow-level embeddings from unlabeled IoT traffic, followed by attaching a lightweight classifier atop the frozen encoder for device-type identification. Trained on over 18 million real-world traffic samples, the method achieves a macro F1-score exceeding 0.9 in cross-network scenarios, significantly outperforming existing pre-trained encoders. These results demonstrate that small models can simultaneously achieve high efficiency, strong generalization, and robustness across diverse deployment environments in IoT settings.

Technology Category

• Application Domains: Internet of Things, Sensor Networks & Smart Cities
• Machine Learning: Learning on the Edge & Model Compression
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Machine learning models have demonstrated strong performance in classifying network traffic and identifying Internet-of-Things (IoT) devices, enabling operators to discover and manage IoT assets at scale. However, many existing approaches rely on end-to-end supervised pipelines or task-specific fine-tuning, resulting in traffic representations that are tightly coupled to labeled datasets and deployment environments, which can limit generalizability. In this paper, we study the problem of learning generalizable traffic representations for IoT device identification. We design compact encoder architectures that learn per-flow embeddings from unlabeled IoT traffic and evaluate them using a frozen-encoder protocol with a simple supervised classifier. Our specific contributions are threefold. (1) We develop unsupervised encoder--decoder models that learn compact traffic representations from unlabeled IoT network flows and assess their quality through reconstruction-based analysis. (2) We show that these learned representations can be used effectively for IoT device-type classification using simple, lightweight classifiers trained on frozen embeddings. (3) We provide a systematic benchmarking study against the state-of-the-art pretrained traffic encoders, showing that larger models do not necessarily yield more robust representations for IoT traffic. Using more than 18 million real IoT traffic flows collected across multiple years and deployment environments, we learn traffic representations from unlabeled data and evaluate device-type classification on disjoint labeled subsets, achieving macro F1-scores exceeding 0.9 for device-type classification and demonstrating robustness under cross-environment deployment.

Problem

• generalizable representations
• IoT device identification
• cross-network
• traffic representation
• unsupervised learning

Innovation

• generalizable representation
• unsupervised traffic embedding
• IoT device identification
• frozen encoder
• cross-environment robustness