A maximum likelihood-based unscented Kalman filter for multipath mitigation in a multi-correlator based GNSS receiver

Abstract

OATAO is an open access repository that collects the work of Toulouse researchers and makes it freely available over the web where possible. ABSTRACTIn complex environments, the presence or absence of multipath signals not only depends on the relative motion between the GNSS receiver and navigation satellites, but also on the environment where the receiver is located. Thus it is difficult to use a specific propagation model to accurately capture the dynamics of multipath signal parameters when the GNSS receiver is moving in urban canyons or other severe obstructions. This paper introduces a statistical model for the line-of-sight and multipath signals received by a GNSS receiver. A multi-correlator based GNSS receiver is also exploited with the advantage to fully characterizing the impact of multipath signals on the correlation function by providing samples of the whole correlation function. Finally, a maximum likelihood-based unscented Kalman filter is investigated to estimate the line-of-sight and multipath signal parameters. Numerical simulations clearly validate the effectiveness of the proposed approach.

Robust Ranging in the Presence of Repeater Signals

“…This methodology has since been adopted in many works on GNSS parameter estimation for multipath scenarios. These works include estimation or tracking of multiple signal arrivals, optimality criteria being ML [18], Bayesian mean squared error [19] or mixtures of the two [20,21]. In [22], we presented a novel algorithm based on the EKF, which substitutes traditional DLLs and is able to robustly track the satellite signal code delay and simultaneously estimate the GNSS signal channel, with the help of a multicorrelator bank structure.…”

Classification of power quality disturbances based on KF‐ML‐aided S‐transform and multilayers feedforward neural networks

“…In statistics, maximum likelihood (ML) estimation is a well‐known method of estimating the parameters of a statistical model given data, and ML parameter estimation has been widely applied to many fields [17, 18]. For the optimal choice of the noise covariance matrices, this paper presents an adaptive Kalman filter based on the maximum likelihood (KF‐ML) for noise rejection by filtering.…”