Research On Spoofing Detection Technology Of Satellite Navigation And Positioning

With the development of global navigation satellite system,satellite navigation and positioning technology has been widely used in people’s daily life,but due to the inherent vulnerability of civil signal,satellite navigation is easily affected by artificial interference.Artificial interference includes coerced interference and spoofing interference,among which spoofing interference has the characteristics of strong concealment and great harm,so this paper mainly focuses on spoofing interference research.This paper analyzes the feasible method of spoofing detection from the perspective of the result of positioning,proposes two different spoofing detection methods,and verifies the detection effect through simulation and experiment.The main work and innovation points of this paper are summarized as follows:(1)Taking GPS system as an example,this paper introduces the basic principles of satellite navigation and positioning,analyzes common spoofing jamming methods,summarizes existing anti-spoofing technologies,and compares their implementation complexity and application scenarios.(2)Aiming at the problem that the traditional spoofing detection methods based on signal characteristics cannot deal with complex spoofing,a spoofing detection algorithm based on base station assistance is proposed.The basic principle of this algorithm uses the high security base station location results to verify the satellite location results that may be spoofed.A consistency factor is proposed to measure the consistency degree of the two location results,and the higher the value is,the more likely it is to be spoofed.The threshold segmentation method is used for spoofing detection.The system model is established,and the consistency factors are analyzed theoretically under normal working and cheating scenarios,and the appropriate theoretical threshold is deduced.Finally,a dataset was generated by spoofing interference test and model simulation.Accuracy and precision were taken as evaluations,and the optimal actual threshold was found by ergodic threshold method.It was found that the theoretical threshold was basically consistent with the actual threshold.In addition,this algorithm was compared with the traditional machine learning classification methods,experiments show that the proposed algorithm has better performance in both detection performance and efficiency.(3)Aiming at the extra hardware cost caused by the base station assisted spoofing algorithm proposed in this paper,a spoofing detection algorithm based on series data prediction and anomaly detection is proposed.Several common prediction algorithms and anomaly detection algorithms for time series data are analyzed and compared.According to the characteristics of nonlinear and randomness of time series satellite positioning data,the spoofing detection algorithm in this paper is presented: First,the concept of jump coefficient is proposed to describe the degree of change between two location results.The jump coefficient is used to process the original data,reduce the dimension of nonlinear and random original data,extract features,and then normalize to speed up neural network learning.Then,LSTM was used for training prediction,and a set of prediction results were obtained,and then reversely normalized to the original dimension for comparison with the original input data.Finally,the upper threshold of the inverse normalized prediction results is calculated according to the dynamic threshold formula defined in this paper,and compared with the threshold point by point.If at least one point is larger than the threshold value,it is judged as cheating,otherwise,it is normal.In order to prove the detection performance of the algorithm,this paper uses the Taxi data set to generate the data set suitable for this paper for experiment.The experiment shows that the accuracy rate is up to 94% when the appropriate neural network structure and dynamic threshold parameters are selected.In addition,compared with other two detection algorithms,experiments show that the algorithm has better detection performance.