Session A2: GNSS Security: Interference, Jamming, and Spoofing 1

A Low Complexity Implementation of the MLE-Based GNSS Anti-Spoofing Method
Xiaoxuan Xu, Hong Li, and Mingquan Lu, Tsinghua University
Location: Beacon A

Peer Reviewed

Spoofing attacks pose a significant threat to Global Navigation Satellite Systems (GNSS). A previous effective spoofing detection technique, the MLE-Based GNSS anti-spoofing method, jointly processes satellite signals at the signal level and shows good performance in detecting spoofing attacks. However, due to the need to search for the optimal value of an eight-dimensional nonlinear function, the high computational complexity of this method limits its practical application. To address the high complexity issue of this anti-spoofing technique, this paper analyzes the specific causes of its computational complexity and proposes a low complexity optimization method, called Dual-Search using PSO and BFGS (DS-PB) method. By reducing the problem to two lower dimensional problems of delay search and Doppler frequency search, employing Particle Swarm Optimization (PSO) and pre-correlation table search for the delay search, and using Broyden-Fletcher-Goldfarb-Shanno (BFGS) method for the Doppler frequency search, the DS-PB method significantly reduces computational complexity while maintaining spoofing detection performance. Both simulation and practical experiments are conducted to verify that the optimized method can achieve spoofing detection in a three-dimensional (3D) position search space of ±2 km and a 3D velocity search space of ±50 m/s. And the time consumption can be greatly reduced from at least hundreds of seconds using grid search to just a few seconds. Compared with traditional grid search methods, it greatly reduces the computational complexity of the algorithm and makes it more practical. Furthermore, supplementary experiments are conducted to explore the effects of application of prior knowledge, the iteration times and the number of particles, which are the main three factors that affect the detection probability and time consumption of the DS-PB method, to provide guidance for practical application of the DS-PB method in different scenarios.