Robust signal processing for GNSS

Digital signal processing for Global Navigation Satellite System (GNSS) receivers is mostly based on the assumption that the noise at the receiver input is Gaussian. This assumption leads to a non-linear Least Squares (LS) problem where GNSS signal parameters are estimated by minimizing a quadratic cost function. The receiver performance can be however significantly degraded by non-Gaussian phenomena such as interference and jamming. This paper reviews robust signal processing concepts and describes two approaches for the design of robust techniques for GNSS signal reception. Specific focus is devoted to Mestimators where the standard quadratic cost function considered by acquisition and tracking is replaced by the sum of less rapidly increasing functions. Using the M-estimator approach, robust versions of the standard acquisition and tracking blocks are obtained. The second approach is based on non-linear filters such as the median and the myriad filters. These filters are used to design robust correlation blocks. A case study involving real data is used to demonstrate the effectiveness of robust signal processing techniques which have the potential to significantly improve the performance of GNSS receivers in the presence of non-Gaussian phenomena such as interference.