Constructing 4D Radio Map in LEO Satellite Networks with Limited Samples

2025-01-06

AI Summary

To address the challenge of constructing high-fidelity 4D radio maps (3D spatial + spectral dimensions) under sparse sensor deployment and limited sampling rates in low-Earth-orbit (LEO) satellite–terrestrial spectrum sharing, this paper proposes DeepRM—a novel unsupervised deep learning framework. DeepRM uniquely integrates neural compressed sensing with tensor decomposition, enabling wideband spectrum reconstruction and accurate 3D spectral mapping from sub-Nyquist measurements without labeled training data. Experimental results demonstrate that, at extremely low sampling rates, DeepRM reduces power spectral reconstruction error by over 30% compared to state-of-the-art methods, significantly enhancing RF environment awareness. Moreover, it substantially alleviates reliance on high-speed analog-to-digital converters (ADCs) and densely deployed sensor arrays. By offering a scalable, low-overhead sensing paradigm, DeepRM advances practical, efficient spectrum coexistence between satellite and terrestrial networks.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Matrix & Tensor Methods
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• 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
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Recently, Low Earth Orbit (LEO) satellite networks (i.e., non-terrestrial network (NTN)), such as Starlink, have been successfully deployed to provide broader coverage than terrestrial networks (TN). Due to limited spectrum resources, TN and NTN may soon share the same spectrum. Therefore, fine-grained spectrum monitoring is crucial for spectrum sharing and interference avoidance. To this end, constructing a 4D radio map (RM) including three spatial dimensions and signal spectra is important. However, this requires the large deployment of sensors, and high-speed analog-to-digital converters for extensive spatial signal collection and wide power spectrum acquisition, respectively. To address these challenges, we propose a deep unsupervised learning framework without ground truths labeling requirement, DeepRM, comprised of neural compressive sensing (CS) and tensor decomposition (TD) algorithms. Firstly, we map the CS process into the optimization of a neural networksassociated loss function, and design a sparsity-performance balance training algorithm to reconstruct a wide power spectrum under limited sub-Nquist samples. Secondly, according to the output of neural CS algorithm, we also utilize neural networks to perform TD, and construct the 3D RM for each frequency, even under very sparse sensor deployment. Extensive evaluations show that DeepRM achieves lower error than its corresponding state-of-the-art baselines, especially with limited samples.

Problem

• Low Earth Orbit (LEO) Satellite Networks
• 4D Radio Map
• Spectrum Sharing

Innovation

• DeepRM
• Neural Compressive Sensing
• Tensor Decomposition