Frequency tracking methods for GPS chirp-type interference detection and mitigation

Abstract

The potential threat of GPS interference has arisen with the increased reliability on GPS and the open availability of portable jammers online at present. In order to deal with this threat, detecting and tracking interference are important for safe GPS operations in all countries. Previous works have focused on detecting the existence of interference using an adaptive notch filter. These approaches are limited when used to detect and track chirp-type interference because the fast sweep rate of this type of interference degrades the signal tracking performance of an adaptive notch filter. Nevertheless, when the adaptation parameters are properly selected, the adaptive notch filter can track and mitigate the jamming signals. This method also has limitations when the quality of the measurement suddenly deteriorates or the sweep rate changes. In this case, the tracking and mitigation performance of the adaptive notch filter is degraded on account of the simple filter structure, which does not include a robust algorithm. Thus, it is necessary to use a model-based tracking algorithm for chirp-type interference when the measurement noise increases. In this dissertation, two Kalman filtering based methods are proposed to design the model-based tracking algorithm for chirp-type interference as well as other continuous wave interference. By using the estimated frequency, mitigation algorithm is also proposed and is based on the second order digital notch filter. The frequency of GPS interference can be obtained using the properties of the trigonometric functions of received signal samples, but these values contain numerous errors caused by measurement noise and frequency changes associated with the interference. In order to reduce these errors, an adaptive fading Kalman filter with a low-pass differentiator (LPD) and a pattern enhancement algorithm is used to estimate the sweep period of chirp-type interference, which is used to reset the filter parameter for estimating the frequency of the interference accurately. By estimating the sweep period, the interference identification logic is designed to select the proper system model of the Kalman filter. However, due to the limited performance of LPD which is used to estimate the sweep period of the interference, the algorithm can only track linear chirp-type interference which has a dramatic change at the end of sweep period. In order to deal with the problem, the revised frequency tracking algorithm which does not depend on estimating the sweep period is needed to track the various chirp-type interferences. Thus, the frequency of chirp-type interference is modeled by a Fourier series, which is always valid regardless of the sweep period and which can maintain tracking performance better than the previous methods when nonlinear chirp-type interference is received. In addition, an optimization technique based on Powells method is applied to the main algorithm in order to select the optimal number of coefficients. Finally, in order to mitigate the interference, the estimated frequency from the filter is used to design a notch filter which eliminates the interference in the received signal. The mitigation performance of the proposed algorithm is evaluated by means of Monte-Carlo simulations. The performance of the proposed algorithm is simulated for scenarios of GPS signals in the presence of various chirp-type interference and is analyzed by using software GPS and interference simulator data. Through theoretical analysis and by comparing simulation results with conventional algorithms, the feasibility and performance of the proposed methods are shown.