GNSS spoofing detection through spatial processing/ Fabian Rothmaier, Yu-Hsuan Chen, Sherman Lo and Todd Walter

In this paper, we present an algorithmic framework for signal-geometry-based approaches of GNSS spoofing detection. We formulate a simple vs. simple hypothesis test independent of nuisance parameters that results in significantly reduced missed detection probability compared to prior approaches. It is highly tractable such that it can be computed online by the receiver. We employ a hypothesis iteration framework that finds spoofed subsets of satellites efficiently and accounts for the presence of weak multipath, for a provable decision behavior in safety-of-life applications. We support the theoretical derivations by showing results on previously published simulated and on-air data sets. We validate the measurement model and show robustness to multipath with flight data from a Dual Polarization Antenna (DPA) mounted on a C12 aircraft. Finally, we show the algorithm's benefit on data recorded during a government-sponsored live spoofing event.