A Robust Evolutionary Particle Filter Technique for Integrated Navigation in Urban Environments via GNSS and 5G Signals

This article focuses on integrated navigation systems using a combination of Global Navigation Satellite Systems (GNSS) and fifth-generation (5G) technology in urban environments. To address the challenge of accurately estimating the mobile terminal state in multipath environments, a robust evolutionary particle filter (REPF) technique is proposed. First, this article presents a modified clock compensation two-way pseudorange scheme that significantly improves pseudorange accuracy in nonideal line-of-sight/nonline-of-sight (LOS/NLOS) pseudorange environments. Then, this article derives the posterior belief conditioned on the obtained pseudorange measurements and velocity data from GNSS. Utilizing the aforementioned posterior distribution, we introduce a robust particle filter (RPF) algorithm to gauge both the sight state and localization in environments with multipath effects. To address the issue of particle degradation in the RPF algorithm, this article introduces a new evolutionary algorithm based on genetic theory to enhance the diversity of particle filtering. The proposed REPF technique is assessed in 5G ultradense networks, and simulation results demonstrate its achievement of accuracy in positioning and tracking at the meter level for moving targets in both LOS and NLOS environments.