RadioDiff-Inverse: Diffusion Enhanced Bayesian Inverse Estimation for ISAC Radio Map Construction

TL;DR

This paper tackles RM construction under coarse environmental knowledge and noisy sparse measurements by casting it as a Bayesian inverse problem. It introduces RadioDiff-Inverse, a training-free framework that leverages a pre-trained unconditional diffusion model to learn the RM prior and perform posterior-based reconstruction, enabling environmental perception such as building outlines and BS localization via pathloss within ISAC. The approach blends Bayesian filtering with diffusion-based priors, providing robust RM reconstruction even with limited data and no precise BS locations. Empirical results on a large-scale RM dataset show state-of-the-art accuracy in RM reconstruction, environmental inference, and resilience to noise and sparse sampling, outperforming traditional interpolation and GAN-based baselines. This work offers a scalable, cross-domain solution for environment-aware wireless networks and ISAC applications when detailed environmental maps are unavailable or privacy-sensitive.

Abstract

Radio maps (RMs) are essential for environment-aware communication and sensing, providing location-specific wireless channel information. Existing RM construction methods often rely on precise environmental data and base station (BS) locations, which are not always available in dynamic or privacy-sensitive environments. While sparse measurement techniques reduce data collection, the impact of noise in sparse data on RM accuracy is not well understood. This paper addresses these challenges by formulating RM construction as a Bayesian inverse problem under coarse environmental knowledge and noisy sparse measurements. Although maximum a posteriori (MAP) filtering offers an optimal solution, it requires a precise prior distribution of the RM, which is typically unavailable. To solve this, we propose RadioDiff-Inverse, a diffusion-enhanced Bayesian inverse estimation framework that uses an unconditional generative diffusion model to learn the RM prior. This approach not only reconstructs the spatial distribution of wireless channel features but also enables environmental structure perception, such as building outlines, and location of BS just relay on pathloss, through integrated sensing and communication (ISAC). Remarkably, RadioDiff-Inverse is training-free, leveraging a pre-trained model from Imagenet without task-specific fine-tuning, which significantly reduces the training cost of using generative large model in wireless networks. Experimental results demonstrate that RadioDiff-Inverse achieves state-of-the-art performance in accuracy of RM construction and environmental reconstruction, and robustness against noisy sparse sampling.

Figures (7)

• Figure 1: Comparison of different scenarios. Row 1: Input measurements; Row 2: Output RMs; Row 3: Ground Truth .
• Figure 2: Illustration of RM construction under different information.
• Figure 3: The illustration of inferencing sequence.
• Figure 4: The illustration of denoising procedure of RadioDiff-Inverse.
• Figure 5: The comparision of constructed RM between different methods

Theorems & Definitions (1)

• Remark 1