Analyzing the Effects of Detected Spoofing Patterns on an Aircraft

Alex Romano, Sherman Lo, Zixi Liu, Yu Hsuan Chen, Todd Walter

Peer Reviewed

Abstract:	GNSS Radio Frequency (RF) Interference (RFI) in the form of spoofing and jamming is becoming increasingly commonplace, transitioning from theoretical vulnerability to daily operational reality. This trend is most visible in the Eastern Mediterranean, Southeast Asia, and across Russia–from the Baltic to the Black Sea, the Ural Mountains, the Caucasus, and Central Asia. A notable characteristic of current GNSS spoofing incidents is that the observed spoofing manifests as distinct geometric patterns rather than singular, static positions. This contrasts with the simpler spoofing attacks targeting airports previously documented in Russia (2018) and Israel. The spoofing observed today takes the form of several notable dynamic trajectories: perfect circles, polygons, and semi-circles transitioning into straight lines with false glide slopes. Some patterns, namely those that transition from a semi-circle to a false glide slope, seem to mimic an aircraft’s final approach to land. This paper documents these patterns based on analysis of Automatic Dependent Surveillance Broadcast (ADS-B) data from flights in the last two years. The introduction and spread of spoofing in Russia are likely to counter autonomous drone attacks stemming from the RussiaUkraine war. To understand the efficacy of these patterns, the second part of this paper assesses the effect of these dynamic spoofing patterns on an aircraft model. Using the Air Force Research Laboratory’s (AFRL) AeroBench F-16 flight model, we can scale the simulation and its parameters to a more appropriate factor for a fixed wing drone. Then, from the observed spoofing patterns observed in the ADS-B data, we inject the spoofed location into the aircraft model to observe its effects on both the airframe and its flight trajectory. Within the simulation, the geometry of the spoofing patterns, the speed at which the patterns are propagated, and the location of the spoofed coordinates can be altered. Lastly, we vary the aircraft’s dependencies on the following information: position, speed (derived from a pitot static tube), heading (from a magnetic compass), and barometric altitude. Either the aircraft has full knowledge of its true values for one of these sets of information, or it relies upon the information delivered by the spoofed GNSS. Altogether, by varying these parameters of the spoofing patterns and the aircraft’s sensor dependencies, we can simulate and quantify the effects of spoofing on an autonomous, fixed wing aerial vehicle, better understand the reason these patterns are specifically employed, and propose possible safeguards against spoofing and its effects. This paper first discusses the use of ADS-B for GNSS spoofing detection and examines spoof trajectories observed in several hotspots around the world, including Russia, Myanmar, and the India-Pakistan border region. It then details the aircraft model 2 and simulation architecture used to analyze the effects of these trajectories, including the parameterized sweep of pattern geometries, spoof speeds, and sensor-dependency modes that vary which navigation channels the autopilot derives from GNSS versus independent onboard sensors. Results are presented across four tested patterns, circles, n-sided polygons, figure-8s, and false approach trajectories, followed by a discussion of operational intent, simulation limitations, and implications for spoof-resilient flight control design.
Published in:	Proceedings of the ION 2026 Pacific PNT Meeting April 13 - 16, 2026 Hilton Waikiki Beach Honolulu, Hawaii
Pages:	576 - 597
Cite this article:	Romano, Alex, Lo, Sherman, Liu, Zixi, Chen, Yu Hsuan, Walter, Todd, "Analyzing the Effects of Detected Spoofing Patterns on an Aircraft," Proceedings of the ION 2026 Pacific PNT Meeting, Honolulu, Hawaii, April 2026, pp. 576-597. https://doi.org/10.33012/2026.20637
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In