Design Principles of Zero-Shot Self-Supervised Unknown Emitter Detectors

TL;DR

This paper systematically evaluates zero-shot self-supervised unknown emitter detection methods, introduces a new 2D-Constellation data modality, and proposes interpretable models and SVD-based initialization to improve detection performance.

Contribution

It presents a comprehensive evaluation of detection systems, introduces a novel 2D-Constellation data modality, and proposes interpretable models and SVD initialization for enhanced performance.

Findings

2D-Constellation data improves detection metrics by up to 40%

SVD initialization enhances Deep Clustering performance by up to 40%

Evaluation across multiple architectures reveals key design insights

Abstract

The proliferation of wireless devices necessitates more robust and reliable emitter detection and identification for critical tasks such as spectrum management and network security. Existing studies exploring methods for unknown emitters identification, however, are typically hindered by their dependence on labeled or proprietary datasets, unrealistic assumptions (e.g. all samples with identical transmitted messages), or deficiency of systematic evaluations across different architectures and design dimensions. In this work, we present a comprehensive evaluation of unknown emitter detection systems across key aspects of the design space, focusing on data modality, learning approaches, and feature learning modules. We demonstrate that prior self-supervised, zero-shot emitter detection approaches commonly use datasets with identical transmitted messages. To address this limitation, we propose a 2D-Constellation data modality for scenarios with varying messages, achieving up to a 40\% performance improvement in ROC-AUC, NMI, and F1 metrics compared to conventional raw I/Q data. Furthermore, we introduce interpretable Kolmogorov-Arnold Networks (KANs) to enhance model transparency, and a Singular Value Decomposition (SVD)-based initialization procedure for feature learning modules operating on sparse 2D-Constellation data, which improves the performance of Deep Clustering approaches by up to 40\% across the same metrics comparing to the modules without SVD initialization. We evaluate all data modalities and learning modules across three learning approaches: Deep Clustering, Auto Encoder and Contrastive Learning.