Detection And Mitigation Of Chirp-Style Jammer For Global Navigation Satellite Systems Receivers

The development of global navigation satellite system(GNSS)and its ever-increasing applications in numerous fields impose higher requirements related to the navigation performance in the GNSS receivers.However,the recently widespread chirp-style jammers,transmitting jamming signals at the GNSS frequencies,present a serious threat to the GNSS receivers.To protect the GNSS receivers,it is of paramount importance to develop effective detection and mitigation techniques for various classes of chirp-style jammers.This dissertation concentrates on the jammer detection and mitigation approaches for GNSS receivers,the main contents are summarized as follows:1.Aiming at detecting various classes of chirp-style jammers in GNSS receivers,time frequency and statistical inference based interference detection technique is proposed.Since the variance of the instantaneous frequency estimates extracted from the time-frequency distribution(TFD)of the received signal in the presence of interference is smaller than that in the interference-free case,two instantaneous frequency estimate sequences are respectively obtained from the received signal and a known interference-free signal.A two-population F-test is applied to compare the two variances of the instantaneous frequency estimates to detect the presence of interference.To reasonably determine the detection threshold for a given fixed false alarm probability,the probability density function(PDF)of the instantaneous frequency estimates obtained from an interference-free signal is analyzed and is shown to be uniform.For two uniform populations in the F-test,the parameters in the distribution function of the test statistic under the null hypothesis are theoretically derived.Extensive simulations are performed to verify the effectiveness of the proposed method.Results show that the proposed approach improves the detection performance for various classes of chirp-style jammers when compared with other approaches reported in the literature,and is insensitive to the jammer sweep rate.2.To further improve the chirp-style jammer detection performance in weak jammer environments,a pre-correlation interference detection method based on statistical analysis in the time-frequency domain for GNSS signals is presented.Based on the statistical analysis of the time-frequency domain points in the short-time Fourier transform(STFT)of the received signal,a chi-square goodness-of-fit test is applied to each frequency slice in the time-frequency domain for interference detection.Compared to the time-domain samples,the jammer-to-noise ratio of the time-frequency domain samples is improved,which is the basis of the performance improvement provided by the proposed method over the existing one.The auto-and cross-correlation functions of the canonical STFT results are analyzed,indicating that the PDF of the time-frequency domain samples deviates from the preset chi-square model.To alleviate the false alarm degradation caused by the observed PDF distortion,another version of the proposed method based on the block-wise STFT using non-overlapped samples is also proposed at the expense of slightly decreased detection capability.Extensive simulations are performed to verify the effectiveness of the proposed canonical and block-wise STFT-based methods.Compared to the existing approaches,the two versions of the proposed method improve the detection sensitivity for chirp-style jammers.3.For GNSS chirp-style jammers with abruptly-changing frequency characteristics,the instantaneous frequency estimates provided by the conventional TFD-based method deviate obviously from their real values around the instantaneous frequency turning points.To enhance the instantaneous frequency estimation accuracy for GNSS chirp-style jammers,an improved jammer characterization algorithm using short-term time-frequency Renyi entropy is first proposed.The instantaneous frequency turning point is extracted from the instantaneous number of components computed based on short-term Renyi entropy.The algorithm procedures for generating improved instantaneous frequency estimates using the instantaneous frequency turning point and the conventional instantaneous frequency estimates are respectively described for jammers with different saw-tooth properties.When the frequency is ramping upwards,the mean property of the quantization errors in the instantaneous frequency estimates is analyzed,and is shown to deviate from zero.Through setting the time-frequency observation window length based on the jammer sweep rate and displacing the instantaneous frequency estimates by a fixed frequency bin,the instantaneous frequency quantization errors are shifted down to oscillate around zero without any modification to the error waveform characteristics.Extensive simulations are performed to verify the effectiveness of the proposed algorithm,and show the improved estimation accuracy of the proposed algorithm over the conventional one.Since the anti-jamming scheme is generally designed taking into account the jammer class,a jammer classification procedure using the instantaneous frequency estimates is finally given.4.Due to the presence of large instantaneous frequency estimation errors around the jammer instantaneous frequency turning points,when applying the existing time-frequency domain jammer mitigation approaches,the processed signal still includes significant jammer suppression residuals,which would affect the navigation performance of the GNSS receivers.To enhance jammer mitigation capability with techniques that are insensitive to instantaneous frequency estimation errors,two new time-frequency domain jammer mitigation techniques with adaptive parameters are developed.1)The nonlinear prediction filter technique proposed for narrowband jammer mitigation in spread spectrum receivers is adapted to the GNSS applications.A three-coefficient finite impulse response filter with coefficients including the chirp rate information is used to model the jamming signal with abruptly-changing frequency characteristics.Since the process and measurement noises are both Gaussian,a linear Kalman filter can be used to estimate the state space which represents the received signal.The TFD-based method is used to provide jammer instantaneous frequency and chirp rate estimates for the updates of the Kalman filter.To alleviate the effect of instantaneous frequency estimation errors on the Kalman filter performance,the process and measurement noise variances are adaptively adjusted based on the jammer frequency sweep characteristics.2)To address the performance degradation of the conventional subspace projection based jammer suppression technique caused by the varying instantaneous phase estimation errors within the projection block,the partitioned subspace projection method is proposed,which however simultaneously causes significant distortion to the desired signal.To solve the contradiction between jammer mitigation and desired signal reservation,an adaptive algorithm for projection block length based on jammer frequency sweep characteristics is further developed.Extensive simulations are performed to demonstrate the effectiveness of the two proposed algorithms.Results indicate that the two proposed approaches provide improved jammer mitigation capability in high jammer power environments when compared to the techniques reported in the literature and unlike these techniques,are insensitive to instantaneous frequency estimation errors.