Joint Localization and Synchronization via User Cooperation in Non-Terrestrial Networks

Abstract

Next-generation (xG) wireless networks require accurate localization and synchronization for efficient resource management and emerging applications. Non-terrestrial networks (NTN) with low Earth orbit (LEO) satellites offer a promising alternative for positioning, navigation, and timing (PNT) by providing diversity and increasing the signal-to-noise ratio (SNR) over global navigation satellite systems (GNSS). However, the primary challenge in NTNbased localization with LEO satellites is the lack of precise clock synchronization, which introduces biases in time-of-arrival (TOA) measurements and limits localization accuracy. This paper introduces a joint cooperative localization and synchronization (JCLS) framework that addresses this challenge through spatiotemporal cooperation, soft information, and simultaneous synchronization. Furthermore, we propose a three-step algorithm for performing JCLS. The first step calculates a coarse position estimate using TOA measurements and the Gauss-Newton method. Then, this coarse estimate is updated using the LevenbergMarquardt method which performs joint localization and synchronization. Finally, we derive a soft information-based filter that is used to continuously refine the position and clock error estimates as new measurements are available. We characterize the fundamental performance limits of JCLS using Fisher information, which offers insight into its localization and synchronization accuracy bounds. Furthermore, simulation results based on TOA measurements of the 3rd Generation Partnership Project (3GPP) 5G New Radio positioning reference signal (PRS) demonstrate that the proposed algorithm for JCLS significantly improves localization and synchronization accuracy compared to non-cooperative methods.