Research On Jamming Detection And Recognition Techniques Of Satellite Navigation System

In recent years,the positioning,navigation and timing services provided by the Global Navigation Satellite System(GNSS)have brought great convenience to people’s lives.However,the GNSS signals have extremely low ground power and are susceptible to various kinds of interference,which causes a great challenge to the security and availability of GNSS.In order to detect interference and identify the type of interference in a timely manner,this paper focuses on the research of jamming detection and recognition techniques for satellite navigation systems,summarizing the following main research contents:(1)Three blind jamming detection algorithms,namely single threshold energy detection algorithm,double threshold energy detection algorithm and Covariance Absolute Value(CAV)detection algorithm,are studied.The effects of sampling length,false alarm probability and signal-to-noise ratio on detection performance are analyzed through theoretical calculations and Monte Carlo simulation experiments.(2)To solve the problem that the fixed threshold of the CAV detection algorithm cannot meet the real-time changing channel environment,an improved CAV detection algorithm is proposed.By constructing a detection error function and deriving its optimal threshold under different signal-to-noise ratios,the simulation results show that the improved CAV detection algorithm has stronger interference detection ability in low signal-to-noise ratio environments.(3)To solve the problem of noise sensitivity and missed detection in the double threshold energy detection algorithm,a joint detection algorithm of double-threshold energy and improved CAV is proposed.The energy detection algorithm is used when the detection statistic falls outside the double threshold,and conversely,the improved CAV detection algorithm is used.The simulation results show that the detection probability of the proposed algorithm is higher than that of any four single detection algorithms,and in low signal-to-noise ratio environments,its detection probability is higher than that of the improved CAV detection algorithm.(4)To solve the problem that traditional recognition methods extract features cumbersomely and low recognition rate,a GNSS jamming recognition algorithm based on timefrequency analysis and convolutional neural network is proposed.The short-time Fourier transform is used to obtain the time-frequency diagram of five types of jamming signals,and a lightweight improved LeNet model(GJR-LeNet)is constructed for recognition.This network model has fewer parameters and faster recognition rate and can fully distinguish between interference and non-interference,but the recognition rate of specific interference types is low.To address this issue,a dual-path fusion network model based on self-attention and residual convolution(SAResFN)is designed,which takes different window lengths of the short-time Fourier transform to obtain long and short time-frequency signals,respectively.They are sent to the multi-head self-attention module and residual convolution module for feature extraction,and the two paths are fused.The experimental results show that the recognition performance of the SAResFN model is improved by 11.73%,0.8%and 1.27%compared with the GJR-LeNet model,the multi-head self-attention module and the residual convolution module,respectively.