REVIEW 5 cited by
CKMDiff: A Generative Diffusion Model for CKM Construction via Inverse Problems with Learned Priors
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Channel knowledge map (CKM) is a promising technology to enable environment-aware wireless communications and sensing with greatly enhanced performance, by offering location-specific channel prior information for future wireless networks. One fundamental problem for CKM-enabled wireless systems lies in how to construct high-quality and complete CKM for all locations of interest, based on only limited and noisy on-site channel knowledge data. This problem resembles the long-standing ill-posed inverse problem, which tries to infer from a set of limited and noisy observations the cause factors that produced them. By utilizing the recent advances of solving inverse problems with learned priors using generative artificial intelligence (AI), we propose CKMDiff, a conditional diffusion model that can be applied to perform various tasks for CKM constructions such as denoising, inpainting, and super-resolution, without having to know the physical environment maps or transceiver locations. Furthermore, we propose an environment-aware data augmentation mechanism to enhance the model's ability to learn implicit relations between electromagnetic propagation patterns and spatial-geometric features. Extensive numerical results are provided based on the CKMImageNet and RadioMapSeer datasets, which demonstrate that the proposed CKMDiff achieves state-of-the-art performance, outperforming various benchmark methods.
Forward citations
Cited by 5 Pith papers
-
Construction and Dynamic Update of Channel Gain Maps via 3D Gaussian Splatting
A 3D Gaussian-splatting model decomposes grid-averaged channel gain into direct and scattered paths, reconstructs static channel gain maps, and incrementally updates them from sparse new measurements.
-
BS-1-to-N: Diffusion-Based Environment-Aware Cross-BS Channel Knowledge Map Generation for Cell-Free Networks
BS-1-to-N uses a diffusion model with location-aware attention to generate a target base station's channel knowledge map from source base station maps and locations.
-
RadioMamba: Breaking the Accuracy-Efficiency Trade-off in Radio Map Construction via a Hybrid Mamba-UNet
On the RadioMapSeer benchmark, RadioMamba reports NMSE 0.0050 versus 0.0072 for RadioDiff, with 28 ms inference and 8.6M parameters.
-
Where to Perform Channel Measurements for CKM Construction: A Random Field Theory Analysis
Adaptive spatial discretization from Gaussian random-field theory plus greedy/SA selection of measurement sites reduces CGM reconstruction AMSE by roughly 20% versus uniform grids under known mean and covariance.
-
Learning-Driven Channel Representation for Wireless Localization: From Channel Observations to Location Inference
A survey organizes learning-driven wireless localization into observation, channel representation, and location inference, arguing representation quality is the decisive performance factor.
Discussion (0). Continue with ORCID to comment.