Open Wireless Digital Twin: End-to-End 5G Mobility Emulation with OpenAirInterface and Ray Tracing

TL;DR

This work introduces the Open Wireless Digital Twin (OWDT), an end-to-end 5G mobility emulation platform that combines OpenAirInterface (OAI) for the 5G NR stack with NVIDIA Sionna RT for deterministic ray-tracing propagation, all running on CPU-based hardware. By pre-computing time-evolving CIRs through ray tracing and applying them to baseband signals via CIR convolution, OWDT achieves real-time, high-fidelity emulation of urban mobility and enables KPI monitoring through O-RAN FlexRIC in near-RT. The authors demonstrate the platform in two Tokyo urban scenarios, Shin-nakano and Shibuya, showing realistic LoS dynamics, multipath evolution, and KPI trends (RSRP, MCS, BLER, throughput) that align with theoretical expectations. The results suggest OWDT’s potential to accelerate wireless system development, reduce experimental costs, and provide a flexible, fully OSS-based reference architecture for RAN/CN evaluation and network optimization.

Abstract

This study presents an end-to-end wireless digital twin platform constructed using open-source software and open data to enhance the evaluation of mobile communication systems. The proposed open wireless digital twin (OWDT) integrates OpenAirInterface (OAI) for Fifth-Generation New Radio (5G NR) protocol stack emulation and NVIDIA Sionna RT for high-resolution ray-tracing-based radio propagation modeling. This integration enables the realistic emulation of 5G wireless communication in mobility scenarios on a CPU-based Linux system, leveraging real-world building data to bridge the gap between theoretical simulations and real-world deployments. The platform also incorporates OAI FlexRIC, which is an implementation aligned with the O-RAN near-real-time RAN Intelligent Controller (near-RT RIC), to dynamically monitor key performance indicators (KPIs). Through extensive evaluation in urban environments, this study demonstrated the validity of the emulation framework, revealing its capability to replicate real-world communication dynamics with high fidelity. The results underscore the potential of the OWDT to accelerate wireless system development, reduce experimental costs, and optimize network configurations.

Figures (7)

• Figure 1: Workflow of generating CIRs for mobility scenario (left panel) and 5G NR system architecture including 5GCN, RAN, Near-RT RIC, and UE developed by the OAI project, along with the RF devices serving as the RU and COTS UE (right panel).
• Figure 2: Ray-tracing simulation for Shibuya scenario.
• Figure 3: Implementation of CIR convolution for the proposed OWDT in OAI.
• Figure 4: (a) Vehicle route on path gain coverage map for Shin-nakano scenario, (b) Power delay profile, (c) path gain, (d) phase delay profile, and (e) enlarged view of (d) for Shin-nakano scenario.
• Figure 5: (a) Vehicle route on path gain coverage map for Shibuya scenario, (b) Power delay profile, (c) path gain, (d) phase delay profile, and (e) enlarged view of (d) for Shibuya scenario.