Digital Network Twin-Enabled Synchronization and Localization

doi:10.1109/JSAC.2025.3610461

2025-09-16

PRE

DOI：10.1109/JSAC.2025.3610461

原文链接

原文求助

摘要

En 中文

This paper addresses the challenge of achieving simultaneous synchronization and localization of all the active terminals within a cellular network from only one Base Station (BS). We propose a novel approach leveraging Digital Network Twins (DNT), which integrates these two critical tasks within a unified framework. We begin by analyzing User Equipment (UE)-to-network time synchronization, both theoretically and through experimental validation using a 5th generation (5G) testbed, identifying it as the primary obstacle to accurate localization. Then, to address this challenge, we introduce a DNT-based framework that leverages high-fidelity ray-tracing simulations on a 3D digital replica of the environment. This enables precise UE-to-network alignment, dynamic environmental mapping, and accurate real-time localization starting from one Next Generation Node Base (gNB). The proposed method integrates Angle Delay Channel Power Matrix (ADCPM) characterization and Time of Flight (ToF) data with the DNT prior knowledge of the environment, eliminating the need for network cooperation or prior on-field channel measurements for precise localization. We first validate the proposed approach through an outdoor measurement campaign and then demonstrate its effectiveness via numerical simulations, compared to existing localization techniques in scenarios where only a single gNB is available. The method achieves on average a positioning accuracy of less than $6{\,}$ m in the static case and $8{\,}$ m in the dynamic scenario, using a ray-tracing granularity that is not excessively fine ( $4 \times 4{\,}$ m), even under worst-case synchronization and Non-Line of Sight (NLoS) conditions.

Keyword:

Localization

synchronization

DNT