Channel Knowledge Map Construction via Guided Flow Matching

TL;DR

This paper introduces a fast, deterministic framework called LT-GFM for constructing accurate channel knowledge maps in wireless networks, overcoming the slow stochastic sampling of diffusion models.

Contribution

The paper proposes a novel linear transport guided flow matching framework that reduces inference steps and applies to multiple CKM types, integrating environmental semantics for physics-informed results.

Findings

Achieves higher distributional fidelity with lower FID scores.

Speeds up inference by a factor of 25 over diffusion models.

Applicable to both channel gain and spatial correlation map construction.

Abstract

The efficient construction of accurate channel knowledge maps (CKMs) is crucial for unleashing the full potential of environment-aware wireless networks, yet it remains a difficult ill-posed problem due to the sparsity of available location-specific channel knowledge data. Although diffusion-based methods such as denoising diffusion probabilistic models (DDPMs) have been exploited for CKM construction, they rely on iterative stochastic sampling, rendering them too slow for real-time wireless applications. To bridge the gap between high fidelity and efficient CKM construction, this letter introduces a novel framework based on linear transport guided flow matching (LT-GFM). Deviating from the noise-removal paradigm of diffusion models, our approach models the CKM generation process as a deterministic ordinary differential equation (ODE) that follows linear optimal transport paths, thereby drastically reducing the number of required inference steps. We propose a unified architecture that is applicable to not only the conventional channel gain map (CGM) construction, but also the more challenging spatial correlation map (SCM) construction. To achieve physics-informed CKM constructions, we integrate environmental semantics (e.g., building masks) for edge recovery and enforce Hermitian symmetry for property of the SCM. Simulation results verify that LT-GFM achieves superior distributional fidelity with significantly lower Fr\'echet Inception Distance (FID) and accelerates inference speed by a factor of 25 compared to DDPMs.