Session A4b: Aviation, Aeronautics, and Uncrewed Aerial Applications 2

A Scalable Pipeline for Real-Time Global Detection and Localization of GNSS Interference Using ADS-B
Zixi Liu, Sherman Lo, Yu-Hsuan Chen, Todd Walter, Stanford University
Location: Holiday 6 (Second Floor)
Date/Time: Thursday, Sep. 11, 2:12 p.m.

GNSS jamming and spoofing have emerged as critical threats to aviation safety and navigation reliability. Current detection methods rely heavily on pilot/operator reports or regional ground sensors, which are not scalable or real-time. We present an automated, near real-time, globally scalable pipeline leveraging Automatic Dependent Surveillance–Broadcast (ADS-B) data. The system integrates Bayesian Online Changepoint Detection (BOCPD) for jamming detection, Kalman filter–based velocity monitoring for spoofing, DBSCAN clustering for event determination, and line-of-sight (LOS) analysis for localization. From 2022–2025, the system has continuously monitored global air traffic, detecting and localizing interference with localization accuracy within 0.01 deg and detection delays under 5 minutes. A public website (rfi.stanford.edu) and Python package make all detection results available for analysis, enabling transparency and broad access.