Specific Emitter Identification via Active Learning

2026-01-08

arXiv.org

AI Summary

This work addresses the challenge of specific emitter identification, which typically relies on large volumes of labeled data that are costly to acquire. To mitigate this dependency, the authors propose a three-stage semi-supervised training framework that integrates self-supervised contrastive learning, supervised fine-tuning, and an active learning sampling strategy based on both uncertainty and representativeness. The framework further incorporates dynamic dictionary updates and a joint contrastive–cross-entropy loss for optimization. Evaluated under extremely limited labeling budgets, the method significantly enhances model generalization and identification accuracy, consistently outperforming existing supervised and semi-supervised approaches on both ADS-B and WiFi datasets, thereby substantially reducing reliance on annotated data.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

With the rapid growth of wireless communications, specific emitter identification (SEI) is significant for communication security. However, its model training relies heavily on the large-scale labeled data, which are costly and time-consuming to obtain. To address this challenge, we propose an SEI approach enhanced by active learning (AL), which follows a three-stage semi-supervised training scheme. In the first stage, self-supervised contrastive learning is employed with a dynamic dictionary update mechanism to extract robust representations from large amounts of the unlabeled data. In the second stage, supervised training on a small labeled dataset is performed, where the contrastive and cross-entropy losses are jointly optimized to improve the feature separability and strengthen the classification boundaries. In the third stage, an AL module selects the most valuable samples from the unlabeled data for annotation based on the uncertainty and representativeness criteria, further enhancing generalization under limited labeling budgets. Experimental results on the ADS-B and WiFi datasets demonstrate that the proposed SEI approach significantly outperforms the conventional supervised and semi-supervised methods under limited annotation conditions, achieving higher recognition accuracy with lower labeling cost.

Problem

• Specific Emitter Identification
• Active Learning
• Labeled Data
• Annotation Cost
• Communication Security

Innovation

• Specific Emitter Identification
• Active Learning
• Self-supervised Contrastive Learning
• Semi-supervised Learning
• Dynamic Dictionary Update