Abstract
The aim of this paper was to propose a method to characterize the ability of a GNSS user to detect a spoofing attack from the behavior of the clock bias. Spoofing interference is not a new issue, especially in military GNSS, although it is a new challenge for civil GNSS, since it is currently implemented and used in many everyday applications. For this reason, it is still a topical issue, especially for receivers that only have access to high-level data (PVT,CN0). To address this important issue, after conducting a study of the receiver clock polarization calculation process, this led to the development of a very basic Matlab model that emulates a spoofing attack at the computational level. Using this model, we were able to observe that the clock bias is affected by the attack. However, the amplitude of this disturbance depends on two factors: the distance between the spoofer and the target and the synchronization between the clock that generates the spoofing signal and the reference clock of the constellation. To validate this observation, more or less synchronized spoofing attacks were carried out on a fixed commercial GNSS receiver with the use of GNSS signal simulators and also with a moving target. We propose then a method to characterize the capacity of detecting a spoofing attack with the clock bias behavior. We present the application of this method for two commercial receivers of the same manufacturer from different generations.