Autors: Singh H., Kumar V., Bonev, B. G., Petkov, P. J., Saxena K., Manolova, A. H.
Title: Hyperparameter-Tuned Machine Learning Models for Predicting Radio Wave Attenuation Due to Rain
Keywords: Communication System Reliability, Machine Learning Models, Radio Wave Attenuation, Rain Fade Prediction

Abstract: Accurate prediction of radio wave attenuation due to rain becomes essential to maintain communication systems' reliability because high-frequency terrestrial and satellite links require this knowledge. The current models from ITU and Brazil and RAL produce different outcomes while requiring a new standardized predictive method. The study developed a predictive model based on machine learning ideas and produced a combined attenuation target by combining three standard models with essential characteristics like frequency patterns and rain rate measurements and empirical variables. Multiple regression models received training through Hyper parameter tuning which allowed for their performance evaluation with MAE and R2 metrics. The results showed XGBoost performed best among all applied models because it produced an MAE value of 0.3478 and an R2 value of 0.9997 which represented strong accuracy in identifying signal attenuation patterns.

References

  1. Kumar, V., Singh, H., Saxena, K., Bonev, B., & Prasad, R. (2023). A Deep Learning Based Model for Prediction of RF Wave Attenuation Due to Rain. Wireless Personal Communications, 131(2), 1437-1460.
  2. Singh, H., Kumar, V., Saxena, K., & Bonev, B. (2020, February). An Intelligent model for prediction of attenuation caused by rain based on machine learning techniques. In 2020 international conference on contemporary computing and applications (IC3A) (pp. 92-97). IEEE.
  3. Livieratos, S. N., & Cottis, P. G. (2019). Rain attenuation along terrestrial millimeter wave links: A new prediction method based on supervised machine learning. IEEE Access, 7, 138745-138756.
  4. Sodunke, M. A., Ojo, J. S., Lawal, Y. B., Ojo, O. L., Owolabi, G. A., & Olateju, A. I. (2025). Application of Machine Learning Models for Rainfall Prediction and Estimation of Rain-induced Attenuation for Satellite Communication in a Tropical Region. Journal of Atmospheric and Solar-Terrestrial Physics, 106443.
  5. Samad, M. A., & Choi, D. Y. (2020). Learning-assisted rain attenuation prediction models. Applied Sciences, 10(17), 6017.
  6. Pu, K., Liu, X., Xian, M., & Gao, T. (2020). Machine learning classification of rainfall types based on the differential attenuation of multiple frequency microwave links. IEEE Transactions on Geoscience and Remote Sensing, 58(10), 6888-6899.
  7. Cornejo, A., Landeros-Ayala, S., Matias, J. M., Ortiz-Gomez, F., Martinez, R., & Salas-Natera, M. (2022). Method of rain attenuation prediction based on long-short term memory network. Neural Processing Letters, 54(4), 2959-2995.

Issue

60th International Scientific Conference on Information, Communication and Energy Systems and Technologies, ICEST 2025 - Proceedings, 2025, Macedonia, https://doi.org/10.1109/ICEST66328.2025.11098257

Copyright The Institute of Electrical and Electronics Engineers

Вид: публикация в международен форум, публикация в реферирано издание, индексирана в Scopus