Geometry-Aware Cross-Height Channel Knowledge Map Prediction for UAV-Assisted Communications With Uncertainty-Guided 3D Sensing

2026-07-01

AI Summary

This work addresses the challenge of reconstructing full-height channel knowledge maps from sparse, multi-altitude observations in geometrically complex urban environments using low-altitude UAVs. The authors propose the first geometry-aware conditional prediction framework that integrates 3D scene priors, sparse measurements, and target altitude descriptors within an FPN-Transformer architecture to enable zero-shot and few-shot generation of cross-altitude channel knowledge maps. An uncertainty quantification module is incorporated to facilitate safety-constrained active sensing. Experimental results demonstrate a zero-shot RMSE of 5.347 dB, which improves to 3.518 dB with only 10 labeled samples. Moreover, uncertainty-guided active perception reduces the reconstruction error from 6.94 dB to 4.79 dB under a budget of 40 sampling queries, significantly outperforming baseline methods.

Abstract

Low-altitude Unmanned Aerial Vehicles (UAVs) often need to infer channel knowledge across a range of heights from only sparse observations collected at a few altitude layers. To address this challenge, this paper studies height-conditioned cross-height channel knowledge map (CKM) prediction for UAV-assisted communications in geometry-rich urban environments. We develop a geometry-aware conditional prediction framework that combines urban scene priors, sparse multi-altitude observations, and target-height descriptors to reconstruct dense CKMs at unobserved target heights. An uncertainty head is further introduced to characterize prediction confidence and to support cost-aware online UAV sensing under motion and safety constraints. Experiments on a layered aerial CKM benchmark show that the proposed Feature Pyramid Network (FPN)-Transformer achieves the best overall performance under both unseen-scene zero-shot and legacy patch-random protocols, reducing the Root Mean Square Error (RMSE) to 5.347dB and 1.111dB, respectively, compared with 6.937dB and 1.221dB for the strongest baseline 3D-RadioDiff. Moreover, after applying our unseen-scene few-shot adaptation, the RMSE further decreases from 5.347dB in zero-shot prediction to 3.518dB with 10-shot two-height support, while the uncertainty-guided cost-aware sensing policy improves active reconstruction from 6.94dB at initialization to 4.79dB at sensing budget 40, outperforming uncertainty-only sensing at 5.08dB and random aerial sampling at 5.84dB.

Problem

• UAV-assisted communications
• cross-height channel prediction
• channel knowledge map
• 3D sensing
• urban environments

Innovation

• geometry-aware prediction
• cross-height channel knowledge map
• uncertainty-guided sensing
• FPN-Transformer
• UAV-assisted communications