Fuzzy evidence accrual for GPS navigation protection of UAVs
Kathleen Ann Kramer, Stephen Craig Stubberud · 2012
Military GPS, such as used for UAVs, is vulnerable to spoofing, and civilian GPS is even more vulnerable. Reliance on GPS, though, continues to increase across all sectors where moving people or products are involved or where mobile surveillance is needed. A research effort has begun to investigate approaches to identify and possibly counter GPS spoofing techniques. In this phase of the effort, an evidence accrual system that uses fuzzy logic and estimation has been applied to this identification. The evidence accrual system begins with the development of a tree structure. The leaf nodes of the tree represent atomic information, the lowest level of information, which is fused with other atomic information into higher concepts of understanding of what is happening to the GPS data. In this instance, these higher concepts are then fused into evidence levels for one of four possible states: no jamming, instantaneous jamming, jam and reset, and bias build-up. Unlike probabilistic techniques, where all of the states need to be normalized to sum to one, this approach allows each state to have its own level of evidence along with a quality or uncertainty measure.