Loading…

A comprehensive survey of machine learning based mobile data traffic prediction models for 5G cellular networks

The advanced progression of mobile technology leads to a dramatic expansion in cellular data traffic (DT). Especially, the development of precise time series model in a 5G cellular network, becomes indispensable to increase the quality of services (QoS) and forecast cellular DT. The cellular DT pred...

Full description

Saved in:
Bibliographic Details
Main Authors: Mohan, R. Raj, Vijayalakshmi, K., Augustine, P. John, Venkatesh, R., Nayagam, M. Gomathy, Jegajohi, B.
Format: Conference Proceeding
Language:English
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The advanced progression of mobile technology leads to a dramatic expansion in cellular data traffic (DT). Especially, the development of precise time series model in a 5G cellular network, becomes indispensable to increase the quality of services (QoS) and forecast cellular DT. The cellular DT predictive models allow the operator in adapting to the traffic demand of the networks with user experience and better resource usage. Furthermore, owing to the characteristics of the higher heterogeneities amongst a number of base stations with traffic load, the cellular DT predictive model becomes challenging. Thus, the study develops a detailed analysis of artificial intelligence (AI) based techniques to predict mobile DT. An extensive analysis of existing ML model has been presented for the cellular DT prediction in 5G network. Additionally, a wide-ranging description of the advancement of mobile networks is provided. Likewise, the present approaches are studied in different ways including underlying methodologies, merits, major objectives, performance measures, and inferences. Also, wide-ranging experimental analyses of the surveyed methods have been taking place for recognizing the exclusive features of all approaches. Lastly, an overview of future directions and challenging issues are extensively discussed.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0177504