Agentic Wireless Digital Twin Construction and Calibration Using Real-World Measurements
Abstract
Wireless digital twins (WDTs) are promising enablers for developing and evaluating AI-native radio access networks, yet constructing a high-fidelity WDT typically requires substantial manual effort to integrate heterogeneous information and infer unknown propagation-related parameters. This paper proposes Agentic WDT (AWDT), an end-to-end agentic framework for autonomous WDT construction and calibration using readily available environmental information and measurements readily obtainable from commercial smartphones. AWDT comprises three agents: EnvAgent constructs the propagation environment, OpAgent infers BS and sector configurations, and MatAgent calibrates radio material properties. The agents iteratively refine the WDT using discrepancies between LTE/NR reference signal received power (RSRP) measurements and ray-tracing predictions. Real-world experiments show that AWDT reduces the RSRP prediction MAE from 11.46 to 5.25 dB, demonstrating a substantial improvement in ray-tracing fidelity. Evaluation with an independent measurement system further demonstrates cross-device transferability with lightweight device-specific bias adaptation, highlighting the potential of agentic AI for scalable WDT construction and calibration with limited prior knowledge.
Related
- Digital Twin-Guided Wireless Localization via Probabilistic Inverse Learning
- Large Language Models Meet Bayesian Optimization: Towards Self-evolving Wireless Digital Twins
- Tri-Band Channel Measurement-Enabled Multi-Layer Digital Twin for Terahertz Wireless Data Centers
- APOD: reasoning-guided agentic population ordinary differential equation discovery for pharmacological digital twins
- Bias-Aware Adjoint Calibration and Optimal Control of Stochastic Multiscale Bioprocess Digital Twins
BibTeX
@misc{lin2026agentic,
title = {{Agentic Wireless Digital Twin Construction and Calibration Using Real-World Measurements}},
author = {Fan-Hao Lin and Kun-Li Shih and Chu-Hsiang Huang and Hui Chen and Chao-Kai Wen and Henk Wymeersch},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.01437},
eprint = {2610.01437},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.01437},
note = {CC BY-NC-SA 4.0},
}
RIS
TY - UNPB
TI - Agentic Wireless Digital Twin Construction and Calibration Using Real-World Measurements
AU - Fan-Hao Lin
AU - Kun-Li Shih
AU - Chu-Hsiang Huang
AU - Hui Chen
AU - Chao-Kai Wen
AU - Henk Wymeersch
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.01437
UR - https://arxiv.org/abs/2610.01437
AB - Wireless digital twins (WDTs) are promising enablers for developing and evaluating AI-native radio access networks, yet constructing a high-fidelity WDT typically requires substantial manual effort to integrate heterogeneous information and infer unknown propagation-related parameters. This paper proposes Agentic WDT (AWDT), an end-to-end agentic framework for autonomous WDT construction and calibration using readily available environmental information and measurements readily obtainable from commercial smartphones. AWDT comprises three agents: EnvAgent constructs the propagation environment, OpAgent infers BS and sector configurations, and MatAgent calibrates radio material properties. The agents iteratively refine the WDT using discrepancies between LTE/NR reference signal received power (RSRP) measurements and ray-tracing predictions. Real-world experiments show that AWDT reduces the RSRP prediction MAE from 11.46 to 5.25 dB, demonstrating a substantial improvement in ray-tracing fidelity. Evaluation with an independent measurement system further demonstrates cross-device transferability with lightweight device-specific bias adaptation, highlighting the potential of agentic AI for scalable WDT construction and calibration with limited prior knowledge.
ER -