EVALUATING THE EFFECTIVENESS OF MACHINE LEARNING MODELS FOR PATH LOSS PREDICTION AT 3.5 GHz WITH FOCUS ON FEATURE PRIORITIZATION

Authors

  • F. E. Shaibu Department of Telecommunication Engineering, Federal University of Technology, Minna, Nigeria https://orcid.org/0000-0001-5470-3965
  • E. N. Onwuka Department of Telecommunication Engineering, Federal University of Technology, Minna, Nigeria
  • N. Salawu Department of Telecommunication Engineering, Federal University of Technology, Minna, Nigeria
  • S. S. Oyewobi Department of Telecommunication Engineering, Federal University of Technology, Minna, Nigeria

DOI:

https://doi.org/10.4314/njt.v43i4.15

Keywords:

Feature Selection, Path Loss, 5G Network, Machine Learning Model, Urban Environment

Abstract

Accurate path loss prediction is vital for efficient resource allocation, interference reduction, and overall network reliability in 5G networks, particularly in the widely deployed mid-band frequency spectrum (such as 3.5 GHz). This study evaluates the effectiveness of machine learning models for path loss prediction at 3.5 GHz with a focus on feature prioritization. A feature selection method, recursive feature elimination, was used to identify significant features from datasets obtained through measurement campaigns, weather stations, 3-D ray tracing, geographical data, and simulations. Out of eighteen features, eleven, including new environmental features, were identified as significant features contributing to path loss. These selected variables were then utilized to optimize and train four common machine learning models (ANN, XGBoost, RF, and k-NN) to evaluate their performance in predicting path loss in a specific urban area called an irregular urban environment. The performance of these models was assessed by comparing their predictions with the measured path loss. The Random Forest model closely matched the measured path loss over the entire path length in both LoS and NLoS scenarios, achieving the lowest MAE of 0.15 dB and RMSE of 0.57 dB in the LoS scenario and 0.62 dB and 1.42 dB in the NLoS scenario, with R2 scores of 0.999995437 and 0.999996828, respectively. This indicates its superior performance in predicting path loss in the urban environment.

Author Biographies

  • F. E. Shaibu, Department of Telecommunication Engineering, Federal University of Technology, Minna, Nigeria

    Farouq E. Shaibu received a B. Eng. in electrical engineering and an M. Eng. (Hons) degree in electronic and communication engineering from the University of Abuja, Abuja, Nigeria, in 2012 and 2020, respectively. In 2015, he joined the Federal Radio Cooperation of Nigeria (FRCN), as a broadcast engineer, where he is now a principal engineer. He is a PhD student studying communication engineering at the Federal University of Technology, Minna, Nigeria. His research interests include RF propagation, satellite communications, and machine learning applications in wireless communications.

  • E. N. Onwuka, Department of Telecommunication Engineering, Federal University of Technology, Minna, Nigeria

    Elizabeth N. Onwuka is a Professor of Telecommunications Engineering. She holds a PhD in Communications and Information Systems Engineering, from Tsinghua University, Beijing, People’s Republic of China; a Master of Engineering degree, in Telecommunications; and a Bachelor of Engineering degree from the Electrical and Computer Engineering Department, Federal University of Technology (FUT) Minna, Niger State, Nigeria. She is the head of the Green Wireless Networking (GWiN) Research Group, FUT Minna. Her research interest includes Mobile communications network, Resource Management in Wireless Communication Networks, Cognitive Radio, Green Wireless Networking, Internet of Things (IoT) Big Data Analytics, and 5G Technologies.

  • N. Salawu, Department of Telecommunication Engineering, Federal University of Technology, Minna, Nigeria

    Salawu Nathaniel is an Associate Professor. He received his B.Eng. in Electrical/Computer Engineering and M.Eng. in Communication Engineering degree from the Federal University of Technology, Minna, Niger State, Nigeria in 2002 and 2010 respectively. He obtained his Ph.D. degree in Electrical Engineering from Universiti Teknologi Malaysia (UTM) in 2018. Currently, he is a senior lecturer in the Department of Telecommunications Engineering, Federal University of Technology, Minna, Niger State, Nigeria, and his areas of research include but are not limited to radio resource management in wireless communication systems, wireless sensor networks, cognitive radio networks and Internet of Things (IoTs).

  • S. S. Oyewobi, Department of Telecommunication Engineering, Federal University of Technology, Minna, Nigeria

    Stephen S. Oyewobi is a senior lecturer. He obtained his PhD from the University of Pretoria, South Africa in 2019. He currently teaches in the department of Telecommunication, at the Federal University of Technology Minna. His research interests include; the Internet of Things, Sensor networks, green technology, radio access technology, and resources management.

References

[1] NOKIA. "5G spectrum bands explained - low, mid and high band." NOKIA. https://www. nokia.com/thought-leadership/articles/spectru m-bands-5g-world/ (accessed 13 July, 2023).

[2] El-Moghazi, M. A., and Whalley, J. “The International Radio Regulations”. Springer Science and Business Media LLC, 2021.

[3] NCC. "Frequency Spectrum." Nigerian Communications Commission. https://ncc.gov .ng/technology/spectrum/frequency-allocation (accessed 7th July, 2023).

[4] Rouphael, T. J. "High-Level Requirements and Link Budget Analysis," in RF and Digital Signal Processing for Software-Defined Radio, T. J. Rouphael Ed.: Elsevier Ltd, 2009, ch. 4, pp. 87-122.

[5] Ray, K., Sharan, S. N., Rawat, S., Jain, S. K., Srivastava, S., and Bandyopadhyay, A. "Engineering Vibration, Communication and Information Processing," presented at the ICoEVCI, India, 2018.

[6] Shaibu, F. E., Onwuka, E. N., Salawu, N., Oyewobi, S. S., Djouani, K., and Abu-Mahfouz, A. M. "Performance of Path Loss Models Over Mid-Band and High-Band Channels for 5G Communication Networks: A Review," Future Internet, Review vol. 15, no. 11, pp. 1-32, 2023, doi: https://doi.org/10.3390 /fi15110362.

[7] Sun, Y., Peng, M., Zhou, Y., Huang, Y., and Mao, S. "Application of Machine Learning in Wireless Networks: Key Techniques and Open Issues," IEEE Communications Surveys and Tutorials vol. 21, no. 4, pp. 3072-3108, 2019, doi: 10.1109/COMST.2019.2924243.

[8] Chen, H., Ma, S., and Lee, H. “CNN-Based mmWave Path Loss Modelling for Fixed Wireless Access in Suburban Scenarios”, 05 August 2020, IEEE Antennas and Wireless Propagation Letters. 1694-1698.

[9] Sukemi, S., Oklilas, A. F., Fadli, M. W., and Alfaresi, B. "Path Loss Prediction Accuracy Based on Random Forest Algorithm in Palembang City Area," Jurnal Nasional Teknik Elektro, vol. 20, no. 1, pp. 1-7, 2023, doi: https://doi.org/10.25077/jnte.v12n1.1051.2023

[10] Barcellos, A. L. d. C., Duarte, J. C., and Mendes, A. C. "Radio Frequency Signal Levels Prediction Using Machine Learning Models," IEEE Latin America Transactions, vol. 21, no. 2, pp. 351-357, January 11 2023, doi: 10.1109/TLA.2023.10015229.

[11] Wu, L. et al., "Artificial Neural Network Based Path Loss Prediction for Wireless Communication Network," IEEE Access, vol. 8, no. 1, pp. 199523 - 199538, 2020, doi: 10.1109/ACCESS.2020.3035209.

[12] Kim, H., Jin, W., and Lee, H. "mmWave Path Loss Modeling for Urban Scenarios Based on 3D-Convolutional Neural Networks," presented at the International Conference on Information Networking (ICOIN), Jeju-Si, Korea, 12-15 January 2022, 2022.

[13] Kuno, N., Yamada, W., Inomata, M., and Sasaki, M. "Evaluation of Characteristics for NN and CNN in Path Loss Prediction," presented at the International Symposium on Antennas and Propagation (ISAP), Osaka, Japan, 2021.

[14] Shaibu, F. E., Onwuka, E. N., Salawu, N., and Oyewobi, S. S. "Performance Analysis of Path Loss Models for Wireless Communications at 3.5 GHz and 23 GHz in a Regular Urban Environment," presented at the 2nd International Conference on Multidisciplinary Engineering and Applied Science (ICMEAS), Abuja, Nigeria, 2023.

[15] Rafie, I. F. M., Lim, S. Y., and Chung, M. J. H. "Path Loss Prediction in Urban Areas: A Machine Learning Approach," IEEE Antennas and Wireless Propagation Letters, vol. 22, no. 4, pp. 809-813, 01 December 2022 2022, doi: 10.1109/LAWP.2022.3225792.

[16] Ahmad, K., and Hussain, S. "Machine Learning Approaches for Radio Propagation Modeling in Urban Vehicular Channels," IEEE Access, vol. 10, pp. 113690-113698, 2022, doi: 10.1109/AC CESS.2022.3218622.

[17] Zhang, Y. P., and Hwang, Y. "Measurements of the characteristics of indoor penetration loss," in Proceedings of IEEE Vehicular Technology Conference (VTC), Stockholm, Sweden, 08-10, June 1994, vol. 03: IEEE, pp. 1741-1744, doi: 10.1109/VETEC.1994.345395

[18] Rodriguez, I., Nguyen, H. C., Jorgensen, N. T. K., Sorensen, T. B., and Mogensen, P. "Radio Propagation into Modern Buildings: Attenua-tion Measurements in the Range from 800 MHz to 18 GHz," presented at the 2014 IEEE 80th Vehicular Technology Conference (VTC2014-Fall), Vancouver, BC, Canada, 14-17, September, 2014.

[19] Laxmi, P. S., Venkatesh, K., Deepthi, K., and Rani, G. U. "Recursive Feature Elimination Method Hybrid Machine Learning Based Intrusion Detection System," International Journal of Creative Research Thoughts (IJCRT), vol. 6, no. 1, pp. 397-402, January 2018 2018.

[20] Awad, M., and Fraihat, S. "Recursive Feature Elimination with Cross-Validation with Decision Tree: Feature Selection Method for Machine Learning-Based Intrusion Detection Systems.," Journal of Sensor and Actuator Networks, vol. 12, no. 67, pp. 1-23, September 2023 2023, doi: 10.3390/jsan12050067.

[21] Adedeji, B. P. "Energy parameter modeling in plug-in hybrid electric vehicles using supervised machine learning approaches " e-Prime - Advances in Electrical Engineering, Electronics and Energy vol. 8, no. 100584, pp. 1-25, 2024, doi: https://doi.org/10.1016/j.prim e.2024.100584.

[22] Idogho, J., and George, G. "Path Loss Prediction Based on Machine Learning Techniques: Support Vector Machine, Artificial Neural Network, and Multilinear Regression Model," Open Journal of Physical Sciences (OJPS), vol. 3, no. 2, pp. 1-20, 2022, doi: 0.52417/ojps.v3i2.393.

[23] Aldossari, S. A. "Predicting Path Loss of an Indoor Environment Using Artificial Intelligence in the 28-GHz Band," MDPI - Electronics, vol. 12, no. 3, p. 497, 2023, doi: https://doi.org/10.3390/electronics12030497.

[24] Han, J. Y., Jo, O., and Kim, J. "Exploitation of Channel-Learning for Enhancing 5G Blind Beam Index Detection," IEEE Transactions on Vehicular Technology, vol. 71, no. 3, pp. 2925-2938, 07 January 2022 2022, doi: 10.1109/TVT.2021.3140019.

Downloads

Published

2025-01-08

Issue

Section

Computer, Telecommunications, Software, Electrical & Electronics Engineering

How to Cite

EVALUATING THE EFFECTIVENESS OF MACHINE LEARNING MODELS FOR PATH LOSS PREDICTION AT 3.5 GHz WITH FOCUS ON FEATURE PRIORITIZATION. (2025). Nigerian Journal of Technology, 43(4), 754 – 762. https://doi.org/10.4314/njt.v43i4.15