Infinite Limits of Convolutional Neural Network for Urban Electromagnetic Field Exposure Reconstruction
Electromagnetic field exposure (EMF) has grown to be a critical concern as a consequence of the ongoing installation of fifth-generation cellular networks (5G). The lack of measurements makes it difficult to accurately assess the EMF in a specific urban area, as Spectrum cartography (SC) relies on a...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10478001/ |
| _version_ | 1797201348694900736 |
|---|---|
| author | Mohammed Mallik Benjamin Allaert Esteban Egea-Lopez Davy P. Gaillot Joe Wiart Laurent Clavier |
| author_facet | Mohammed Mallik Benjamin Allaert Esteban Egea-Lopez Davy P. Gaillot Joe Wiart Laurent Clavier |
| author_sort | Mohammed Mallik |
| collection | DOAJ |
| description | Electromagnetic field exposure (EMF) has grown to be a critical concern as a consequence of the ongoing installation of fifth-generation cellular networks (5G). The lack of measurements makes it difficult to accurately assess the EMF in a specific urban area, as Spectrum cartography (SC) relies on a set of measurements recorded by spatially distributed sensors for the generation of exposure maps. However, when the spatial sampling rate is limited, significant estimation errors occur. To overcome this issue, the exposure map estimation is addressed as a missing data imputation task. We compute a convolutional neural tangent kernel (CNTK) for an infinitely wide convolutional neural network whose training dynamics can be completely described by a closed-form formula. This CNTK is employed to impute the target matrix and estimate EMF exposure from few sensors sparsely located in an urban environment. Experimental results show that the kernel, even when only sparse sensor data are available, can produce accurate estimates. It is a promising solution for exposure map reconstruction that does not require large training sets. The proposed method is compared with other deep learning approaches and Gaussian Process regression. |
| first_indexed | 2024-04-24T07:46:07Z |
| format | Article |
| id | doaj.art-c008967fc8634dbc8ac19f7ab33fe8de |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-24T07:46:07Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-c008967fc8634dbc8ac19f7ab33fe8de2024-04-18T23:00:29ZengIEEEIEEE Access2169-35362024-01-0112494764948810.1109/ACCESS.2024.338083510478001Infinite Limits of Convolutional Neural Network for Urban Electromagnetic Field Exposure ReconstructionMohammed Mallik0https://orcid.org/0000-0003-1854-9483Benjamin Allaert1https://orcid.org/0000-0002-4291-9803Esteban Egea-Lopez2https://orcid.org/0000-0002-6926-4923Davy P. Gaillot3https://orcid.org/0000-0003-3455-5824Joe Wiart4https://orcid.org/0000-0002-8902-5778Laurent Clavier5https://orcid.org/0000-0002-3279-930XIMT Nord Europe, Lille, FranceIMT Nord Europe, Lille, FranceDepartment of Information Technologies and Communications, Universidad Politécnica de Cartagena (UPCT), Cartagena, SpainCNRS, UMR 8520-IEMN, Université de Lille, Lille, FranceChaire C2M, LTCI, Télécom Paris, Institut Polytechnique de Paris, Palaiseau, FranceIMT Nord Europe, Lille, FranceElectromagnetic field exposure (EMF) has grown to be a critical concern as a consequence of the ongoing installation of fifth-generation cellular networks (5G). The lack of measurements makes it difficult to accurately assess the EMF in a specific urban area, as Spectrum cartography (SC) relies on a set of measurements recorded by spatially distributed sensors for the generation of exposure maps. However, when the spatial sampling rate is limited, significant estimation errors occur. To overcome this issue, the exposure map estimation is addressed as a missing data imputation task. We compute a convolutional neural tangent kernel (CNTK) for an infinitely wide convolutional neural network whose training dynamics can be completely described by a closed-form formula. This CNTK is employed to impute the target matrix and estimate EMF exposure from few sensors sparsely located in an urban environment. Experimental results show that the kernel, even when only sparse sensor data are available, can produce accurate estimates. It is a promising solution for exposure map reconstruction that does not require large training sets. The proposed method is compared with other deep learning approaches and Gaussian Process regression.https://ieeexplore.ieee.org/document/10478001/5G EMF exposurekernel regressionneural tangent kernelinfinite width convolutional neural networksemi-supervised learning |
| spellingShingle | Mohammed Mallik Benjamin Allaert Esteban Egea-Lopez Davy P. Gaillot Joe Wiart Laurent Clavier Infinite Limits of Convolutional Neural Network for Urban Electromagnetic Field Exposure Reconstruction IEEE Access 5G EMF exposure kernel regression neural tangent kernel infinite width convolutional neural network semi-supervised learning |
| title | Infinite Limits of Convolutional Neural Network for Urban Electromagnetic Field Exposure Reconstruction |
| title_full | Infinite Limits of Convolutional Neural Network for Urban Electromagnetic Field Exposure Reconstruction |
| title_fullStr | Infinite Limits of Convolutional Neural Network for Urban Electromagnetic Field Exposure Reconstruction |
| title_full_unstemmed | Infinite Limits of Convolutional Neural Network for Urban Electromagnetic Field Exposure Reconstruction |
| title_short | Infinite Limits of Convolutional Neural Network for Urban Electromagnetic Field Exposure Reconstruction |
| title_sort | infinite limits of convolutional neural network for urban electromagnetic field exposure reconstruction |
| topic | 5G EMF exposure kernel regression neural tangent kernel infinite width convolutional neural network semi-supervised learning |
| url | https://ieeexplore.ieee.org/document/10478001/ |
| work_keys_str_mv | AT mohammedmallik infinitelimitsofconvolutionalneuralnetworkforurbanelectromagneticfieldexposurereconstruction AT benjaminallaert infinitelimitsofconvolutionalneuralnetworkforurbanelectromagneticfieldexposurereconstruction AT estebanegealopez infinitelimitsofconvolutionalneuralnetworkforurbanelectromagneticfieldexposurereconstruction AT davypgaillot infinitelimitsofconvolutionalneuralnetworkforurbanelectromagneticfieldexposurereconstruction AT joewiart infinitelimitsofconvolutionalneuralnetworkforurbanelectromagneticfieldexposurereconstruction AT laurentclavier infinitelimitsofconvolutionalneuralnetworkforurbanelectromagneticfieldexposurereconstruction |