Wireless Propagation Modeling Through Bayesian Networks
Open Access
- 2 December 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 181632-181643
Abstract
Over the last three decades, cellular network planning has evolved as a discipline in response to the ever-increasing complexity of mobile telephony. One of the key inputs to coverage planning, a central stage of the cellular network planning process, is the channel propagation model. This encompasses the main sources of degradation experienced by a signal in its way from transmitter to receiver. Currently, various well-known empirical models are used to estimate propagation losses in different environments. However, these mathematical models have been obtained by inference from measurements taken at specific locations, and generally their predictions do not match actual measurements obtained in other locations. Even worse, typically such predictions are not mutually consistent. To manage this uncertainty in terms of probability estimates rather than predictions, in this paper we propose a machine learning approach that is based on training a Bayesian network model from virtual data generated by sampling several empirical models. As a first step towards this initiative, we start by developing a Bayesian network driven by three well-known empirical propagation models, in their versions for urban environments: Okumura-Hata, COST 231 Walfish-Ikegami and Lee, which allow us to compare them in terms of probability estimations.Keywords
Funding Information
- Spanish National Plan for Scientific and Technical Research and Innovation (PID2022-139248NB-I00)
- European Regional Development Fund
This publication has 36 references indexed in Scilit:
- Channel measurement and characterisation for 5G multi‐scenarios at 26 and 38 GHzIET Microwaves, Antennas & Propagation, 2023
- Large scale survey for radio propagation in developing machine learning model for path losses in communication systemsScientific African, 2023
- Propagation path loss prediction modelling in enclosed environments for 5G networks: A reviewHeliyon, 2022
- Bayesian networks for interpretable machine learning and optimizationNeurocomputing, 2021
- Bayesian network modeling: A case study of an epidemiologic system analysis of cardiovascular riskComputer Methods and Programs in Biomedicine, 2016
- A Bayesian network to discover relationships between negative features in sport: a case study of teen playersQuality & Quantity, 2013
- New insights into the classification and nomenclature of cortical GABAergic interneuronsNature Reviews Neuroscience, 2013
- Learning Bayesian networks: approaches and issuesThe Knowledge Engineering Review, 2011
- Probabilistic graphical models in artificial intelligenceApplied Soft Computing, 2011
- A simple graphical approach for understanding probabilistic inference in Bayesian networksInformation Sciences, 2009