Bridging the Sampling Distribution Shift in Radio Map Estimation: A Trajectory-Aware Paradigm

2026-05-27

AI Summary

This work addresses the distribution shift in drone-based wireless sensing caused by the mismatch between trajectory-based sampling and the i.i.d. assumption commonly adopted in model training. To this end, the authors propose a trajectory-aware training paradigm, ST-TBS, which is the first to statistically characterize how trajectory sampling reduces spatial diversity and introduces information redundancy. ST-TBS mitigates this issue through a randomly triggered trajectory sampling strategy that preserves trajectory continuity while enhancing sampling diversity, thereby effectively aligning the training and test distributions. Integrated with a spatial field reconstruction model and a trajectory-aware data generation scheme, ST-TBS significantly improves generalization in real-world deployment scenarios, reducing the RMSE under trajectory observation on the SpectrumNet dataset from 0.2632 to 0.0571.

Abstract

Learning-based radio map estimation (RME) plays a critical role in UAV-assisted wireless sensing, enabling tasks such as coverage prediction and network optimization. Most current methods assume an independently and identically distributed (i.i.d.) training and testing setting based on random sampling. However, practical UAV measurements are collected sequentially along feasible trajectories, resulting in highly structured and spatially correlated patterns. This mismatch introduces a sampling distribution shift that increases the intrinsic difficulty of spatial field recovery and compromises the generalization of models trained under i.i.d. assumptions. To mitigate this issue, we propose a trajectory-aware training paradigm based on Stochastic-Triggered Trajectory-Based Sampling (ST-TBS), which preserves trajectory continuity while introducing sampling variability. Moreover, from a statistical perspective, we show that trajectory-based sampling reduces spatial diversity and increases information redundancy compared to random sampling. Extensive experiments on the RadioMapSeer and SpectrumNet datasets demonstrate that models trained with random sampling suffer significant performance degradation under trajectory-based observations, with RMSE increasing from 0.0391 to 0.2632 on SpectrumNet. Conversely, our proposed ST-TBS method effectively reduces the RMSE to 0.0571. These results highlight the necessity of aligning training and deployment sampling distributions for reliable RME.

Problem

• radio map estimation
• sampling distribution shift
• trajectory-based sampling
• spatial correlation
• UAV-assisted wireless sensing

Innovation

• trajectory-aware learning
• sampling distribution shift
• radio map estimation
• spatial field recovery
• Stochastic-Triggered Trajectory-Based Sampling