Advanced KF-based methods for GNSS carrier tracking and ionospheric scintillation mitigation

Abstract

Ionospheric scintillation is the name given to the disturbance caused by electron density irregularities along the propagation path of electromagnetic waves through the iono sphere.

Performance analysis of multi-frequency GNSS carrier tracking for strong ionospheric scintillation mitigation

“…This paper leverages on this idea and proposes a carrier tracking method that jointly processes signals at multiple GNSS frequencies. The paper extends our previous work on single-frequency carrier tracking loops under scintillation conditions [4], [5], for which a state-space formulation and a Kalman filter (KF) were analyzed. Here, an augmented state-space formulation considering multiple frequencies is exploited and a tracking method proposed.…”

“…Such an approach allows us to distinguish between both phase contributions and estimate the scintillation process. It was shown in [4], [5] that these components can be fairly well approximated using an AR process as…”

“…The main problems arise in the AR model order selection, {q, p} in (2) and (3), and the time-series analysis to obtain the AR parameters. In [4], [5], the selection using synthetic data was done somehow heuristically, only via visual inspection of the power spectral densities, which lacks of mathematical rigor. In this contribution, we provide an analysis for the model order selection based on partial autocorrelation functions (PAF), which is a basic tool commonly used in time-series analysis for model order selection [7].…”

“…That is, all standard techniques keep track of the incoming signal complete phase, then being difficult to distinguish the phase contribution due to dynamics from the scintillation phase disturbance, thus providing poor scintillation mitigation capabilities. To overcome this limitation, a new tracking approach was proposed in [4], [5], where the scintillation physical phenomena is mathematically modeled using convenient AR processes and embedded in the filter formulation. In this article, we extend these results to the multi-frequency case.…”

“…Then after the acquisition stage, the samples at the output of the correlators can be modeled as [5] …”