Atmospheric Influence on the Path Loss at High Frequencies for Deployment of 5G Cellular Communication Networks

TL;DR

This study analyzes how atmospheric conditions like temperature, humidity, and rain affect 5G signal path loss at high frequencies in urban microcell environments, using regression and machine learning models for accurate prediction across seasons.

Contribution

It provides a comparative analysis of atmospheric effects on 5G path loss and evaluates machine learning models for reliable prediction in South Asian urban environments.

Findings

Machine learning models outperform regression techniques in path loss prediction.

Atmospheric factors significantly influence 5G signal attenuation.

Seasonal variations affect radio link quality and path loss estimates.

Abstract

Over the past few decades, the development of cellular communication technology has spanned several generations in order to add sophisticated features in the updated versions. Moreover, different high-frequency bands are considered for advanced cellular generations. The presence of updated generations like 4G and 5G is driven by the rising demand for a greater data rate and a better experience for end users. However, because 5G-NR operates at a high frequency and has significant propagation, atmospheric fluctuations like temperature, humidity, and rain rate might result in poorer signal reception, and higher path loss effects unlike the prior generation, which employed frequencies below 6 GHz. This paper makes an attempt to provide a comparative analysis about the influence of different relative atmospheric conditions on 5G cellular communication for various operating frequencies in any urban microcell (UMi) environment maintaining the real outdoor propagation conditions. In addition, the simulation dataset based on environmental factors has been validated by the prediction of path loss using multiple regression techniques. Consequently, this study also aims to address the performance analysis of regression techniques for stable estimations of path loss at high frequencies for different atmospheric conditions for 5G mobile generations due to various possible radio link quality issues and fluctuations in different seasons in South Asia. Furthermore, in comparison to contemporary studies, the Machine Learning models have outperformed in predicting the path loss for the four seasons in South Asian regions.