GPS Interference Mitigation Using Derivative-free Kalman Filter-based RNN

Abstract

The global positioning system (GPS) with accurate positioning and timing properties has become integral part of all applications around the world. Radio frequency interference can significantly decrease the performance of GPS receivers or even completely prohibit the acquisition or tracking of satellites. The approaches of system performances that can be further enhanced by preprocessing to reject the jamming signal will be investigated. A recurrent neural network (RNN) predictor for the GPS anti-jamming applications will be proposed. The adaptive RNN predictor is utilized to accurately predict the narrowband waveform based on an unscented Kalman filter (UKF)-based algorithm. The UKF algorithm as a derivative-free alternative to the extended Kalman filter (EKF) in the framework of state-estimation is adopted to achieve better performance in terms of convergence rate and quality of solution. The adaptive RNN filter can be successfully applied for the suppression of interference with a number of different narrowband formats, i.e. continuous wave interference (CWI), multi-tone CWI, swept CWI and pulsed CWI, to emulate realistic circumstances. Simulation results show that the proposed UKF-based scheme can offer the superior performances to suppress the interference over the conventional methods by computing mean squared prediction error (MSPE) and signal-to-noise ratio (SNR) improvements.