Supervised method maps 5G phone signals into interpretable device charts
Supervised Device Charting with CSI Measurements from Commercial 5G NR User Equipments
Information Theory
Summary
Distinguishing individual wireless devices by subtle hardware signal differences is tricky and usually only gives device names with no clear view of similarities or errors. The authors propose a way to transform 5G signal data from phones into simple visual maps that show how close or distinct each device’s signals are. They tested this on six commercial smartphones and found the charts help spot when devices are confused or stand out as unusual. The method also works across different days, revealing changes in signal patterns over time and grouping devices of the same phone model nearby on the map.
What this means in practice
- •For wireless security teams: Generate interpretable visual charts from 5G signals to improve identifying and distinguishing unauthorized or impersonated devices.
- •For mobile network operators: Enhance network monitoring by mapping device fingerprints over time to detect unusual device behavior or model grouping shifts.
Authors
Mischa Vasylyev, Frederik Zumegen, Reinhard Wiesmayr, Christoph Studer
Abstract
Radio frequency fingerprint identification (RFFI) is a promising approach to distinguish physical wireless devices using hardware-induced signal imperfections. Conventional RFFI methods only provide discrete device labels and no human-interpretable representation of the relations among received signals. We propose supervised device charting, which maps location-insensitive channel-state information (CSI) fingerprints to a low-dimensional chart that visualizes cluster compactness, overlap, and outliers. We evaluate the method with real-world 5G New Radio (5G NR) measurements from six commercial smartphones and introduce the neighbor label error rate (NLER) to quantify class-separation accuracy. Our results demonstrate that two- and three-dimensional device charts provide an interpretable visualization of the learned RFFI representation. For three-dimensional device charts, the NLER is 0.22% for same-day measurements and 7.38% for measurements from the next day. The device charts reveal a cross-day distribution shift and map the held-out device close to the known device of the same model. Increasing the device chart dimensions further improves cluster separation at the expense of interpretability.