Improving Tracking Robustness Through Interference Using Pilot Signals with a Deeply Coupled Estimator

This paper shows the viability of improving tracking robustness of global navigation satellite systems (GNSS) pilot signals in high interference and/or jamming conditions by deep coupling with inertial sensors using a Kalman filter. In this work, we confront the limiting factors of typical tracking loops, including the dependency on pre-filtering or coherent averaging (Julien 2014), the adverse correlation effects that would otherwise come from integrating over the Doppler frequency of the incoming signals (Borio et al. 2014, Julien 2014), biased inertial measurement sensor (IMU) accelerometer/gyroscope noise inputs, and local oscillator (LO) phase noise (Misra and Enge 2001). Our deeply coupled Kalman filter is designed to specifically confront these limitations. The use of a deeply coupled Kalman filter also allows for a well-defined analysis of the integrity of the filter’s best state estimate, which can be used to expose noise sources which most quickly degrade estimate quality. Using this analysis, the robustness of this and similar estimators to all noise levels given all available hardware can be extended and defined, and thus provide a valuable asset not only to robustness, but also to estimator and sensing scheme design. We show that this early version of our tracking algorithm is able to maintain signal lock in carrier to noise density ratios as low as 4 dBHz.