Abstract:

Traditional unmanned aerial vehicle (UAV) navigation spoofing techniques use a single antenna to transmit spoofing signals, while non-cooperative UAV equipped with array antennas has the ability to recognize the arriving directions of signals, which can detect, identify, and suppress spoofing signals from a single arriving direction, leading to spoofing failures. To address this issue, firstly, a density-based spatial clustering of applications with noise(DBSCAN)-Kmeans hybrid clustering algorithm is designed to cluster the signal directions based on the true navigation satellite relative to the target UAV azimuth, in order to obtain the optimal direction angle for the spoofing signal. Secondly, multiple cooperative UAVs carry spoofing payloads and transmit mutually self-consistent spoofing signals from the optimal arriving directions, which makes sure that the spoofing signals are nearly identical to the real signals arriving directions. Finally, experimental results show that the multi-directions spoofing signals transmitted by multiple spoofing payloads can successfully intrude into the navigation links of non-cooperative UAV equipped with array antennas, and the proposed method has good covertness and feasibility.

Key words: unmanned aerial vehicle (UAV) countermeasures, covert navigation spoofing, UAV collabo-ration, hybrid clustering