CDI Maps: Dynamic Estimation of the Radio Environment for Predictive Resource Allocation

Kulzer, Daniel Fabian; Stańczak, Sławomir; Botsov, Mladen

doi:10.1109/pimrc50174.2021.9569310

Cited by 7 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Predicting radio environment maps (REMs) that capture LSF could facilitate deployment planning and operation optimization of wireless networks. For instance, REM prediction can play an essential role in spectrum sharing [1], localization [2], path planning for unmanned aerial vehicles (UAVs) [3], finding coverage holes [4], optimizing resource allocation in dense networks [5], etc.…”

Section: Introductionmentioning

confidence: 99%

REM-U-Net: Deep Learning Based Agile REM Prediction With Energy-Efficient Cell-Free Use Case

Sallouha,

Sarkar,

Krijestorac

et al. 2024

IEEE Open J. Signal Process.

View full text Add to dashboard Cite

Radio environment maps (REMs) hold a central role in optimizing wireless network deployment, enhancing network performance, and ensuring effective spectrum management. Conventional REM prediction methods are either excessively time-consuming, e.g., ray tracing, or inaccurate, e.g., statistical models, limiting their adoption in modern inherently dynamic wireless networks. Deep learningbased REM prediction has recently attracted considerable attention as an appealing, accurate, and timeefficient alternative. However, existing works on REM prediction using deep learning are either confined to 2D maps or use a relatively small dataset. In this paper, we introduce a runtime-efficient REM prediction framework based on U-Nets, trained on a large-scale 3D maps dataset. In addition, data preprocessing steps are investigated to further refine the REM prediction accuracy. The proposed U-Net framework, along with preprocessing steps, are evaluated in the context of the 2023 IEEE ICASSP Signal Processing Grand Challenge, namely, the First Pathloss Radio Map Prediction Challenge. The evaluation results demonstrate that the proposed method achieves an average normalized root-mean-square error (RMSE) of 0.045 with an average of 14 milliseconds (ms) runtime. Finally, we position our achieved REM prediction accuracy in the context of a relevant cell-free massive multiple-input multiple-output (CF-mMIMO) use case. We demonstrate that one can obviate consuming energy on large-scale fading (LSF) measurements and rely on predicted REM instead to decide which sleep access points (APs) to switch on in a CF-mMIMO network that adopts a minimum propagation loss AP switch ON/OFF strategy.

show abstract

Section: Introductionmentioning

confidence: 99%

REM-U-Net: Deep Learning Based Agile REM Prediction With Energy-Efficient Cell-Free Use Case

Sallouha,

Sarkar,

Krijestorac

et al. 2024

IEEE Open J. Signal Process.

View full text Add to dashboard Cite

show abstract

“…Over time, several new approaches to wireless resource sharing, including dynamic spectrum access [1], licensed shared access [2] have been proposed. However, additional technological components, such as spectrum usage databases [3] and radio environment maps [4], had to be developed before being able to come closer to such sophisticated and dynamic spectrum usage approaches. To be able to correctly inform on spectrum usage, additional knowledge of the devices transmitting within the range of a wireless network is critical for future smart usage of the spectrum.…”

Section: Introductionmentioning

confidence: 99%

Self-supervised Learning for Clustering of Wireless Spectrum Activity

Ljupcho¹,

Cerar²,

Bertalanič³

et al. 2022

Preprint

View full text Add to dashboard Cite

In recent years, much work has been done on processing of wireless spectral data involving machine learning techniques in domain-related problems for cognitive radio networks, such as anomaly detection, modulation classification, technology classification and device fingerprinting. Most of the solutions are based on labeled data, created in a controlled manner and processed with supervised learning approaches. Labeling spectral data is a laborious and expensive process, being one of the main drawbacks of using supervised approaches. In this paper, we introduce self-supervised learning for exploring spectral activities using real-world, unlabeled data. We show that the proposed model achieves superior performance regarding the quality of extracted features and clustering performance. We achieve reduction of the feature vectors size by 2 orders of magnitude (from 3601 to 20), while improving performance by 2 to 2.5 times across the evaluation metrics, supported by visual assessment. Using 15 days of continuous narrowband spectrum sensing data, we found that 17% of the spectrogram slices contain no or very weak transmissions, 36% contain mostly IEEE 802.15.4, 26% contain coexisting IEEE 802.15.4 with LoRA and proprietary activity, 12% contain LoRA with variable background noise and 9% contain only dotted activity, representing LoRA and proprietary transmissions.

show abstract

True 3D Holography: A Communication Service of Tomorrow and Its Requirements for a New Converged Cloud and Network Architecture on the Path to 6G

Friese,

Galkow-Schneider,

Bassbouss

et al. 2023

2023 2nd International Conference on 6G Networking (6GNet)

View full text Add to dashboard Cite

CDI Maps: Dynamic Estimation of the Radio Environment for Predictive Resource Allocation

Cited by 7 publications

References 19 publications

REM-U-Net: Deep Learning Based Agile REM Prediction With Energy-Efficient Cell-Free Use Case

REM-U-Net: Deep Learning Based Agile REM Prediction With Energy-Efficient Cell-Free Use Case

Self-supervised Learning for Clustering of Wireless Spectrum Activity

True 3D Holography: A Communication Service of Tomorrow and Its Requirements for a New Converged Cloud and Network Architecture on the Path to 6G

Contact Info

Product

Resources

About