RadioDiff-FS: Physics-Informed Manifold Alignment in Few-Shot Diffusion Models for High-Fidelity Radio Map Construction

2026-03-19

AI Summary

This work addresses the challenge of high-fidelity radio channel map construction, which is hindered by the high cost of electromagnetic simulations, poor generalization of data-driven methods, and scarcity of labeled measurements. The authors propose a few-shot diffusion model framework that innovatively decomposes channel maps into a dominant path component and a directionally sparse residual. A physics-guided manifold alignment mechanism adapts a pre-trained dominant-path generator to complex multipath environments using only a small number of real-world measurements. Directional consistency loss (DCL) enforces physical propagation constraints during the diffusion process, effectively modeling cross-domain shifts as bounded geometric translations. Experiments demonstrate significant improvements, achieving NMSE reductions of 59.5% and 74.0% in static and dynamic scenarios, respectively, along with an SSIM of 0.9752 and a PSNR of 36.37 dB.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Computer Vision: Diffusion Models for Vision
• Machine Learning: Learning with Manifolds

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Security and Privacy: Large-scale security measurements

Abstract

Radio maps (RMs) provide spatially continuous propagation characterizations essential for 6G network planning, but high-fidelity RM construction remains challenging. Rigorous electromagnetic solvers incur prohibitive computational latency, while data-driven models demand massive labeled datasets and generalize poorly from simplified simulations to complex multipath environments. This paper proposes RadioDiff-FS, a few-shot diffusion framework that adapts a pre-trained main-path generator to multipath-rich target domains with only a small number of high-fidelity samples. The adaptation is grounded in a theoretical decomposition of the multipath RM into a dominant main-path component and a directionally sparse residual. This decomposition shows that the cross-domain shift corresponds to a bounded and geometrically structured feature translation rather than an arbitrary distribution change. A Direction-Consistency Loss (DCL) is then introduced to constrain diffusion score updates along physically plausible propagation directions, suppressing phase-inconsistent artifacts that arise in the low-data regime. Experiments show that RadioDiff-FS reduces NMSE by 59.5% on static RMs and by 74.0% on dynamic RMs relative to the vanilla diffusion baseline, achieving an SSIM of 0.9752 and a PSNR of 36.37 dB under severely limited supervision.

Problem

• Radio Map Construction
• Few-Shot Learning
• Multipath Propagation
• High-Fidelity Modeling
• 6G Network Planning

Innovation

• few-shot diffusion
• physics-informed manifold alignment
• radio map construction
• direction-consistency loss
• multipath decomposition