CKM Beyond Channel Gain: Spatial Correlation Map Construction with Deep Learning

2026-04-22

AI Summary

This work addresses the challenge of reconstructing high-dimensional spatial channel maps (SCMs) from sparse measurements by proposing a deep learning approach that integrates environmental prior knowledge. The method decouples the SCM into a path gain map and a path angle map, and introduces the E-SRResNet architecture, which combines multi-head attention mechanisms with multi-scale feature fusion. Crucially, it incorporates priors such as line-of-sight (LoS) information, building layouts, and base station locations to enable efficient and accurate reconstruction of the full SCM from limited observations. To the best of our knowledge, this is the first study to apply deep learning to high-dimensional SCM modeling, overcoming the limitation of conventional methods that only reconstruct gain maps. Experiments on the CKMImageNet dataset demonstrate that the proposed method significantly outperforms existing baselines, achieving cosine similarities above 0.8 with ground-truth SCMs across most regions.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Multimodal Learning
• Intelligent Robots: State Estimation

Application Category

• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Channel knowledge map (CKM) is a promising technique to achieve environment-aware wireless communication and sensing. Constructing the complete CKM based on channel knowledge observations at sparse locations is a fundamental problem for CKM-enabled wireless networks. However, most existing works on CKM construction only consider the special type of CKM, i.e., the channel gain map (CGM), which only records the channel gain value for each location. In this paper, we consider the channel spatial correlation map (SCM) construction, which signifies the location-specific spatial correlation matrix for multi-antenna systems. Unlike CGM construction, constructing SCM poses significant challenges due to its extremely high-dimensional structure. To address this issue, we first decompose the high-dimensional SCM into lower-dimensional path gain map (PGM) and path angle map (PAM). Then we propose a deep learning model termed E-SRResNet for constructing high-quality SCM from sparse samples, which incorporates multi-head attention (MHA) mechanisms and multi-scale feature fusion (MSFF) to accurately model both local and global spatial relationships of channel parameters and complex nonlinear mappings. Furthermore, we preprocess the dataset to provide priors including line-of-sight (LoS) map, binary building map and base station (BS) map for the model to reconstruct SCM more accurately. Simulations conducted on the CKMImageNet dataset demonstrate that the proposed E-SRResNet achieves significant performance improvements over baseline methods. Moreover, the cosine similarity between the constructed SCM and the ground truth exceeds 0.8 in most regions, validating the effectiveness of the proposed construction method.

Problem

• spatial correlation map
• channel knowledge map
• multi-antenna systems
• high-dimensional structure
• sparse samples

Innovation

• spatial correlation map
• deep learning
• multi-head attention
• multi-scale feature fusion
• channel knowledge map