GNSS Positioning Enhancement Based on NLOS Multipath Biases Estimation Using Gaussian Mixture Noise

Abstract

Global navigation satellite systems (GNSS) have been widely used in many applications where positioning plays an important role. However, the performances of these applications can be degraded in urban canyons, due to Non-Line-Of-Sight (NLOS) and Multipath interference affecting GNSS signals. In order to ensure high accuracy positioning, this article proposes to model the NLOS and Multipath biases by Gaussian Mixture noise using Expectation Maximization (EM) algorithm. In this context, an approach to estimate the Multipath and NLOS biases for real time positioning is presented and statistical tests for searching the probability distribution of NLOS and Multipath biases are illustrated. Furthermore, a hybrid approach based on PF (Particle Filter) and EM algorithm for estimating user position in hard environment is presented. Using real GPS (Global Positioning System) signal, the efficiency of the proposed approach is shown, and a significant improvement of the positioning accuracy over the simple PF estimation is obtained.