Agentic Wireless Digital Twin Construction and Calibration Using Real-World Measurements

2026-10-01

AI Summary

This study addresses the reliance on extensive manual integration of heterogeneous information and inference of unknown parameters in wireless digital twin (WDT) construction by proposing AWDT, a multi-agent framework. AWDT introduces a pioneering end-to-end multi-agent collaboration mechanism in which EnvAgent, OpAgent, and MatAgent work synergistically to fuse environmental data, ray tracing, and mobile LTE/NR reference signal power measurements. Through iterative closed-loop calibration among the three agents, autonomous WDT construction is achieved. Real-world measurements demonstrate that, with only limited prior knowledge, the mean absolute error of reference signal received power (RSRP) prediction decreases from 11.46 dB to 5.25 dB. These results indicate significant improvements in simulation fidelity, cross-device transferability, and system scalability.

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.

Problem

• Wireless Digital Twin
• Autonomous Construction
• Calibration
• Ray-tracing Fidelity

Innovation

• Wireless Digital Twin
• Agentic AI
• Ray Tracing Calibration
• Multi-Agent Framework
• Cross-Device Transferability