Title: GNSS Interference Detection and Localization using ADS-B Data: An Automated Pipeline for Global Coverage
Abstract: The Global Navigation Satellite System (GNSS) plays a critical role in aviation safety, enabling precise navigation and collision avoidance. However, its vulnerability to radio frequency interference (RFI), including jamming and spoofing, poses severe threats, particularly during critical flight phases such as approach and landing. This thesis introduces a novel and scalable approach for the rapid detection and localization of GNSS interference events using ADS-B data. By leveraging wide-area, crowd-sourced observations from aircraft and integrating a suite of published methodologies, we developed a fully automated pipeline to provide continuous, global GNSS interference monitoring.
This system improves situational awareness, enhances safety, and supports the timely identification and mitigation of interference sources. The pipeline includes two core components: global RFI awareness and local RFI onset monitoring. Global detection and localization employ changepoint detection, DBSCAN clustering, CNNs, and nonlinear least squares localization. The onset component provides near real-time alerts through a Bayesian algorithm that enables continuous online updates. Operating 24/7, the system detects hundreds of global GNSS interference events daily and visualizes them at https://rfi.stanford.edu/. It is capable of identifying events within 5 minutes of onset, achieving real-world localization accuracy within a 10 km radius at a 95% confidence level. The website provides rapid and reliable surveillance of global GNSS interference.
Speaker: Zixi Liu is a PhD candidate at Stanford University in the GPS Lab. She received her M.S. degree in Aeronautics and Astronautics from Stanford University in 2020. Her research focuses on GNSS interference detection and localization. Her work lies in the intersection of statistics, optimization, and machine learning.
Location: Virtual Zoom Meeting
Please contact a member of the executive for the link.
Date: Thursday, May 29, 2025
Time: Meeting will open at 11:30am, presentation beginning at noon
Cost: Free