A Civil GPS Anti-Spoofing and Recovering Method Using Multiple Tracking Loops and an Adaptive Filter Technique

Nowadays, with several evidences that a GPS (global positioning system) spoofing attack may work, a security of a civil GPS receiver is threatened. The GPS spoofing attacker try to deceive the GPS receiver with a resembled set of normal GPS signals, which should be seen at a position of the corresponding GPS receiver. At the same time, those spoofed signals are partly modified to provide the target GPS receiver with incorrect position information, which may be designed with a malicious intent. To protect the GPS receiver in that kind of difficult situation, this paper presents a civil GPS anti-spoofing method against the GPS spoofing attack by blocking the spoofed GPS signals and recovering the normal GPS signals. Multiple tracking loops for both the spoofed GPS signals and the normal GPS signals, and an adaptive filtering technique are used to achieve this goal. Result of each tracking loop is used to remove the other signals from the received GPS signals, which is a sum of the spoofed and normal GPS signals. To apply the method to the anti-spoofing operation, three different scenarios, where the GPS receiver can be deceived by the GPS spoofing attacker, are examined considering the condition of the spoofed GPS signals with respect to the normal GPS signals. The first scenario assumes that the spoofed GPS signals approach to the normal GPS signals in domains of the code phase or/and Doppler frequency, because the GPS spoofing attacker do not know an exact position and a dynamic characteristic of the target GPS receiver. In this case, an additional tracking loop for each channel is allocated and run to also trace the spoofed GPS signals once it is detected by an acquisition process. The estimation of the spoofed GPS signals is eliminated from the received GPS signals to protect the tracking loop for the normal GPS signals from the GPS spoofing attack. Since the acquisition process to detect the spoofed GPS signals prior to a valid attack needs much computing power and time, appropriate boundaries for both the Doppler frequency and the code phase are applied to the acquisition process carried out around the normal GPS signals periodically. For the second scenario, the situation where the spoofed GPS signals already overlap the normal GPS signals is considered. This means that the GPS spoofing attacker know the position and the dynamic characteristic of the target GPS receiver exactly. After a while, the spoofed GPS signals pretend to be the normal GPS signals by deceiving the tracking loops for the Doppler frequency and the code phase in the GPS receiver with a higher signal power in general. And it attempts to lead those tracking loops away from the normal GPS signals by deviating slowly. At the beginning of this process, a sum of two signals can be interpreted as a multipath. Then, the additional tracking loop is allocated to also trace the normal GPS signals. With a comparison between the tracking results of those tracking loops and the known characteristics of the spoofed GPS signals such as the Doppler frequency transition, the signal strength, the relative signal strength for satellites, and so on, the normal GPS signals can be recovered. The last scenario supposes a cooperation of a GPS jammer and the GPS spoofing attacker. In this case, after the GPS jammer let the target GPS receiver lost a signal lock for the normal GPS signals by broadcasting a jamming signal with an extremely high power, the GPS spoofing attacker broadcast the spoofed GPS signals, which have higher signal power than the normal GPS signals in general. Then, the GPS receiver may track the spoofed GPS signals. To solve this, the GPS receiver use logged information of the Doppler frequency and the code phase delay before the jamming attack not to mistake the spoofed GPS signal for the normal GPS signal. The closest signal to the logged one in acquisition result for both the Doppler frequency and the code phase can be selected as the normal GPS signal. For the case where the spoofed GPS signals have the same Doppler frequency and code phase to the normal GPS signals, the second scenario can be applied to solve the problem. The proposed method use multiple tracking loops, which operate independently for the spoofed and normal GPS signals by adopting the adaptive filter technique, to protect the GPS receiver from the GPS spoofing attack. Considered scenarios in this paper are expected to cover all conditions where the GPS anti-spoofing is needed. And, comparing to the other GPS anti-spoofing methods, the proposed method have a strength that it does not need any modification for the structure or data in the specification of GPS signal, and an extra hardware for the GPS receiver. The simulation with the software-based GPS spoofing signal generator and GPS receiver would be carried out to certify that the proposed method can be applied to the civil GPS anti-spoofing. With a selection of several parameters, such as the boundary for the rapid acquisition process, and the detection threshold for spoofed GPS signals, and the condition of additional tracking loop allocation, the proposed method is anticipated to block the spoofed GPS signals and recover the normal GPS signals effectively.