{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T05:03:17Z","timestamp":1789707797710,"version":"4.0.1"},"reference-count":51,"publisher":"MDPI AG","issue":"18","license":[{"start":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T00:00:00Z","timestamp":1789084800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>In this paper, the solar irradiance is investigated in a part of central Serbia, where a low-cost IoT sensor network was deployed at three locations: a mountain slope, an open field near a village, and an obstructed position in the city center. All three locations are in the same NASA POWER grid cell. After multi-stage quality control, 26,145 valid daytime records were compared to the satellite reference. The satellite assigns identical values to all three positions but the measured mean daytime irradiances are 267.7, 351.5 and 19.6 W\/m2, respectively. The rural station is in the best agreement with the reference (R2 = 0.567); the mountain station suffers from a persistent shading bias (MBE = \u2212138.9 W\/m2); the signal at the urban station is attenuated by surrounding buildings and vegetation by a factor of ~12. Also, 25 regression models (gradient boosting, recurrent, convolutional, fully connected and graph-based) were trained on the 25-year monthly NASA POWER record for the same cell. XGBoost obtained R2 = 0.998, the hybrid TCN-GNN R2 = 0.968 and a plain ReLU network R2 = 0.912, and a seasonal model with calendar features was used to reconstruct the satellite reference for two months of 2026 not yet available in the archive. The deployed hardware, network and energy behavior are documented quantitatively: per-site link delivery ratios of 96.8%, 83.5% and 58.2%, the photovoltaic harvesting record of the nodes, and the absence of any energy-aware transmission scheduling. Two of the trained models were compiled for the ESP32 nodes and measured on the deployment hardware: a depth-limited gradient-boosted corrector runs in 46.3 \u00b5s using 11.1 kB of flash, and an INT8 fully connected network in 138.1 \u00b5s using 3.0 kB, while the graph hybrid cannot be converted for microcontroller execution at all. This defines an edge\u2013cloud partition in which local inference performs bias correction and fault detection while the cloud tier retains the heavy models and periodic retraining. To the best of the authors\u2019 knowledge, this is the first multi-site ground-based IoT irradiance record for central Serbia with such different terrain types.<\/jats:p>","DOI":"10.3390\/s26185781","type":"journal-article","created":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T13:29:29Z","timestamp":1789133369000},"page":"5781","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Regional Ground-Based IoT Solar Irradiance Monitoring: A Multi-Site Study Across Mountain, Rural, and Urban Environments"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7532-6185","authenticated-orcid":false,"given":"Dejan","family":"Vuji\u010di\u0107","sequence":"first","affiliation":[{"name":"Faculty of Technical Sciences \u010ca\u010dak, University of Kragujevac, 32102 \u010ca\u010dak, Serbia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7270-6702","authenticated-orcid":false,"given":"Du\u0161an","family":"Markovi\u0107","sequence":"additional","affiliation":[{"name":"Faculty of Agronomy \u010ca\u010dak, University of Kragujevac, 32102 \u010ca\u010dak, Serbia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pranay","family":"Obla Anandbabu","sequence":"additional","affiliation":[{"name":"Viterbi School of Engineering, University of Southern California, Los Angeles, CA 90089, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-4260-3523","authenticated-orcid":false,"given":"Shrihari Rajeev","family":"Kulkarni","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology, Vellore 632014, India"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6078-413X","authenticated-orcid":false,"given":"Zoran","family":"Stamenkovi\u0107","sequence":"additional","affiliation":[{"name":"Synopsys Armenia Educational Department, National Polytechnic University of Armenia, Yerevan 0026, Armenia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7676-4942","authenticated-orcid":false,"given":"Sini\u0161a","family":"Ran\u0111i\u0107","sequence":"additional","affiliation":[{"name":"Faculty of Technical Sciences \u010ca\u010dak, University of Kragujevac, 32102 \u010ca\u010dak, Serbia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,11]]},"reference":[{"key":"ref_1","unstructured":"Stackhouse, P.W., Zhang, T., Westberg, D., Barnett, A.J., Bristow, T., Macpherson, B., and Hoell, J.M. (2026, September 01). POWER Release 8 (with GIS Applications) Methodology (Data Parameters, Sources, & Validation), Data Version 8.0.1, Available online: https:\/\/power.larc.nasa.gov\/."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1803","DOI":"10.1016\/j.solener.2012.03.006","article-title":"A New Solar Radiation Database for Estimating PV Performance in Europe and Africa","volume":"86","author":"Huld","year":"2012","journal-title":"Sol. Energy"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1016\/j.rser.2018.03.003","article-title":"The National Solar Radiation Data Base (NSRDB)","volume":"89","author":"Sengupta","year":"2018","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"339","DOI":"10.1016\/j.solener.2018.02.059","article-title":"Evaluation of Global Horizontal Irradiance Estimates from ERA5 and COSMO-REA6 Reanalyses Using Ground and Satellite-Based Data","volume":"164","author":"Urraca","year":"2018","journal-title":"Sol. Energy"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1024","DOI":"10.1016\/j.rser.2014.07.117","article-title":"A Review of Validation Methodologies and Statistical Performance Indicators for Modeled Solar Radiation Data: Towards a Better Bankability of Solar Projects","volume":"39","author":"Gueymard","year":"2014","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"569","DOI":"10.1016\/j.renene.2016.12.095","article-title":"Machine Learning Methods for Solar Radiation Forecasting: A Review","volume":"105","author":"Voyant","year":"2017","journal-title":"Renew. Energy"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"487","DOI":"10.1016\/j.rser.2019.02.006","article-title":"Automatic Hourly Solar Forecasting Using Machine Learning Models","volume":"105","author":"Yagli","year":"2019","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"e039676","DOI":"10.1136\/bmjopen-2020-039676","article-title":"Comparison of ARIMA Model and XGBoost Model for Prediction of Human Brucellosis in Mainland China: A Time-Series Study","volume":"10","author":"Alim","year":"2020","journal-title":"BMJ Open"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Chen, T., and Guestrin, C. (2016, January 13\u201317). XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.","DOI":"10.1145\/2939672.2939785"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long Short-Term Memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Cho, K., van Merri\u00ebnboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., and Bengio, Y. (2014, January 25\u201329). Learning Phrase Representations Using RNN Encoder\u2013Decoder for Statistical Machine Translation. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, Qatar.","DOI":"10.3115\/v1\/D14-1179"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"102946","DOI":"10.1016\/j.mex.2024.102946","article-title":"A Critical Review of RNN and LSTM Variants in Hydrological Time Series Predictions","volume":"13","author":"Waqas","year":"2024","journal-title":"MethodsX"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"1084","DOI":"10.1016\/j.egyr.2022.07.139","article-title":"Multivariate Time Series Prediction by RNN Architectures for Energy Consumption Forecasting","volume":"8","author":"Amalou","year":"2022","journal-title":"Energy Rep."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"2991","DOI":"10.1038\/s41598-024-53460-y","article-title":"A Novel Extreme Adaptive GRU for Multivariate Time Series Forecasting","volume":"14","author":"Zhang","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"6775","DOI":"10.1007\/s00500-023-09531-9","article-title":"A Bi-GRU-Based Encoder\u2013Decoder Framework for Multivariate Time Series Forecasting","volume":"28","author":"Balti","year":"2024","journal-title":"Soft Comput."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Shahid, S.M., Ko, S., and Kwon, S. (2022, January 12\u201315). Performance Comparison of 1D and 2D Convolutional Neural Networks for Real-Time Classification of Time Series Sensor Data. Proceedings of the 2022 International Conference on Information Networking (ICOIN), Jeju-si, Republic of Korea.","DOI":"10.1109\/ICOIN53446.2022.9687284"},{"key":"ref_17","unstructured":"Bai, S., Kolter, J.Z., and Koltun, V. (2018). An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. arXiv."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Lara-Ben\u00edtez, P., Carranza-Garc\u00eda, M., Luna-Romera, J.M., and Riquelme, J.C. (2020). Temporal Convolutional Networks Applied to Energy-Related Time Series Forecasting. Appl. Sci., 10.","DOI":"10.20944\/preprints202003.0096.v1"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"042050","DOI":"10.1088\/1742-6596\/1213\/4\/042050","article-title":"Temporal Convolutional Networks for Anomaly Detection in Time Series","volume":"1213","author":"He","year":"2019","journal-title":"J. Phys. Conf. Ser."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"8016505","DOI":"10.1109\/LGRS.2021.3095505","article-title":"Channel Attention-Based Temporal Convolutional Network for Satellite Image Time Series Classification","volume":"19","author":"Tang","year":"2022","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"17351","DOI":"10.1007\/s00521-020-04867-x","article-title":"A CNN\u2013LSTM Model for Gold Price Time-Series Forecasting","volume":"32","author":"Livieris","year":"2020","journal-title":"Neural Comput. Appl."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"9903518","DOI":"10.1155\/2021\/9903518","article-title":"Multivariate CNN-LSTM Model for Multiple Parallel Financial Time-Series Prediction","volume":"1","author":"Widiputra","year":"2021","journal-title":"Complexity"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"124889","DOI":"10.1016\/j.energy.2022.124889","article-title":"Forecasting Monthly Gas Field Production Based on the CNN-LSTM Model","volume":"260","author":"Zha","year":"2022","journal-title":"Energy"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"212528","DOI":"10.1016\/j.geoen.2023.212528","article-title":"Time Series Forecasting of Oil Production in Enhanced Oil Recovery System Based on a Novel CNN-GRU Neural Network","volume":"233","author":"Chen","year":"2024","journal-title":"Geoenergy Sci. Eng."},{"key":"ref_25","unstructured":"Kipf, T.N., and Welling, M. (2017, January 24\u201326). Semi-Supervised Classification with Graph Convolutional Networks. Proceedings of the 5th International Conference on Learning Representations (ICLR), Toulon, France."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Wu, Z., Pan, S., Long, G., Jiang, J., and Zhang, C. (2019, January 10\u201316). Graph WaveNet for Deep Spatial-Temporal Graph Modeling. Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI), Macao, China.","DOI":"10.24963\/ijcai.2019\/264"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"103094","DOI":"10.1016\/j.cose.2023.103094","article-title":"GRU-Based Interpretable Multivariate Time Series Anomaly Detection in Industrial Control System","volume":"127","author":"Tang","year":"2023","journal-title":"Comput. Secur."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"2875","DOI":"10.1109\/LSP.2024.3479917","article-title":"A Recurrent Spatio-Temporal Graph Neural Network Based on Latent Time Graph for Multi-Channel Time Series Forecasting","volume":"31","author":"Li","year":"2024","journal-title":"IEEE Signal Process. Lett."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Nava-Pintor, J.A., Alcal\u00e1-Rodr\u00edguez, U.E., Guerrero-Osuna, H.A., Mata-Romero, M.E., Lopez-Neri, E., Garc\u00eda-V\u00e1zquez, F., Sol\u00eds-S\u00e1nchez, L.O., Carrasco-Navarro, R., and Luque-Vega, L.F. (2025). Development and Evaluation of Solar Radiation Sensor Using Cost-Effective Light Sensors and Machine Learning Techniques. Technologies, 13.","DOI":"10.3390\/technologies13050182"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"153","DOI":"10.21595\/jme.2020.21667","article-title":"A Conversion Guide: Solar Irradiance and Lux Illuminance","volume":"8","author":"Michael","year":"2020","journal-title":"J. Meas. Eng."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"e170101421779","DOI":"10.33448\/rsd-v10i14.21779","article-title":"Performance Evaluation of the BH1750FVI Sensor (Low Cost) in the Measurement of Global Solar Radiation","volume":"10","author":"Silveira","year":"2021","journal-title":"Res. Soc. Dev."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"681","DOI":"10.1111\/j.1467-9671.2012.01337.x","article-title":"A New 3-D Solar Radiation Model for 3-D City Models","volume":"16","author":"Hofierka","year":"2012","journal-title":"Trans. GIS"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"3891","DOI":"10.1016\/j.rser.2012.03.042","article-title":"A Review of Concentrating Solar Power Plants in the World and Their Potential Use in Serbia","volume":"16","year":"2012","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_34","unstructured":"Republic Hydrometeorological Service of Serbia (RHMZ) (2024). Meteorological Yearbook, RHMZ."},{"key":"ref_35","unstructured":"ROHM Semiconductor (2019). BH1750FVI Ambient Light Sensor IC, ROHM Semiconductor. Datasheet BH1750FVI-E, Rev. D."},{"key":"ref_36","unstructured":"Bosch Sensortec (2022). BME280 Combined Humidity and Pressure Sensor, Bosch Sensortec. Datasheet BST-BME280-DS002, Rev. 1.6."},{"key":"ref_37","unstructured":"Espressif Systems (2023). ESP32 Series Datasheet, Espressif Systems. Rev. 4.1."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"363","DOI":"10.1007\/s00704-013-1000-3","article-title":"Transmissivity of Solar Radiation through Crowns of Single Urban Trees\u2014Application for Outdoor Thermal Comfort Modelling","volume":"117","author":"Konarska","year":"2014","journal-title":"Theor. Appl. Climatol."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"32","DOI":"10.5937\/GeoPan1601032T","article-title":"Solar Permeability of Different Tree Species in Szeged, Hungary","volume":"20","author":"Kiss","year":"2016","journal-title":"Geogr. Pannon."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"102506","DOI":"10.1016\/j.uclim.2025.102506","article-title":"Transmissivity of Solar Radiation and the Influence of Diffuse Radiation in Different Trees inside a Tropical Urban Park","volume":"62","author":"Acero","year":"2025","journal-title":"Urban Clim."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"102522","DOI":"10.1016\/j.uclim.2025.102522","article-title":"Evaluating the Effectiveness of Tree Canopy and Building Shade in Urban Heat Mitigation Using Solar Radiation Transmittance","volume":"62","author":"Wu","year":"2025","journal-title":"Urban Clim."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1016\/S0960-1481(03)00171-X","article-title":"Building Cluster and Shading in Urban Canyon for Hot Dry Climate","volume":"29","author":"Bourbia","year":"2004","journal-title":"Renew. Energy"},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"113904","DOI":"10.1016\/j.enbuild.2024.113904","article-title":"A Comprehensive Study of How Urban Morphological Parameters Impact the Solar Potential, Energy Consumption and Daylight Autonomy in Canyons and Buildings","volume":"305","author":"Rostami","year":"2024","journal-title":"Energy Build."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"120801","DOI":"10.1016\/j.energy.2021.120801","article-title":"Experimental Investigation about the Adoption of High Reflectance Materials on the Envelope Cladding on a Scaled Street Canyon","volume":"230","author":"Battista","year":"2021","journal-title":"Energy"},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Iha, T., Kawamitsu, I., Ohshiro, A., and Nakamura, M. (2021, January 27\u201330). An LSTM-Based Multivariate Time Series Predicting Method for Number of Restaurant Customers in Tourism Resorts. Proceedings of the 36th International Technical Conference on Circuits\/Systems, Computers and Communications (ITC-CSCC), Jeju, Republic of Korea.","DOI":"10.1109\/ITC-CSCC52171.2021.9501432"},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Elsayed, N., Maida, A.S., and Bayoumi, M. (2019, January 14\u201317). Gated Recurrent Neural Networks Empirical Utilization for Time Series Classification. Proceedings of the 2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData), Atlanta, GA, USA.","DOI":"10.1109\/iThings\/GreenCom\/CPSCom\/SmartData.2019.00202"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"015003","DOI":"10.1088\/2631-8695\/ac4834","article-title":"Research on Fault Diagnosis of Automobile Engines Based on the Deep Learning 1D-CNN Method","volume":"4","author":"Du","year":"2022","journal-title":"Eng. Res. Express"},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"Jain, D., and Upadhyaya, S. (2024, January 2\u20135). Enhancing Solar Energy Forecasting: 1D CNN with Explainable AI Techniques. Proceedings of the 2024 IEEE India Geoscience and Remote Sensing Symposium (InGARSS), Goa, India.","DOI":"10.1109\/InGARSS61818.2024.10984181"},{"key":"ref_49","doi-asserted-by":"crossref","unstructured":"Contreras, J., Ceberio, M., and Kreinovich, V. (2021). Why Dilated Convolutional Neural Networks: A Proof of Their Optimality. Entropy, 23.","DOI":"10.20944\/preprints202104.0501.v1"},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"19342","DOI":"10.1038\/s41598-022-24036-5","article-title":"Novel Deep Learning Hybrid Models (CNN-GRU and DLDL-RF) for the Susceptibility Classification of Dust Sources in the Middle East: A Global Source","volume":"12","author":"Gholami","year":"2022","journal-title":"Sci. Rep."},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Raja Sekaran, S., Pang, Y.H., You, L.Z., and Yin, O.S. (2024). A Hybrid TCN-GRU Model for Classifying Human Activities Using Smartphone Inertial Signals. PLoS ONE, 19.","DOI":"10.1371\/journal.pone.0304655"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/18\/5781\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T04:13:35Z","timestamp":1789704815000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/18\/5781"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,11]]},"references-count":51,"journal-issue":{"issue":"18","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["s26185781"],"URL":"https:\/\/doi.org\/10.3390\/s26185781","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,11]]}}}