Assessing spoofing of GPS systems

• Autor(a) principal: Dias, Rui Filipe Pereira
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/22211
• topic: Radionavigation; Anti spoofing; Spoofing; GNSS; Radionavegação; Defesa contra spoofing; Falsificação

Lately, plenty of self navigation vehicles have been developed, as drones, or in the future, self driving cars. However, it has become easier to forge radionavigation signals, which can be a problem. With the growing risk of this threat, there has to be way to solve it and this thesis goal is to study various ways to mitigate this problem. For this effect, an u-blox evk-m8t GNSS (Global Navigation Satellite System) receiver was used, which is capable of returning raw unprocessed data from radio navigation signals. A raspberry pi was also used to analyze the data. This is not a linear problem, since each spoofer is unique, it is necessary to pay attention to transitions, comparing old with new data. Since each scenario is a different scenario, the variations will be observed in order to try to find a variation pattern. These variations will be tested in a neural network in order to find if it is viable to detect forged signals this way. Spoofing as a whole also has specific variations that should not be there, the unstable clock variation is the most influenceable factor. This work managed to conclude that it is possible to implement a calibration algorithm that is able to detect patterns in forged signals and distinguish them from legitimate signals. Forged signals, normally, are more incoherent in variations of signal properties and its functioning as a whole, for example, the position that would be calculated by removing a satellite from the equation. These signals also present unpredicted variations in the clock delay.