Game Theoretic Approach to QoS Oriented Machine Learning Model Development Toward 5G Network Migration Planning

With the increasing number of connected devices, the demand for mobile data has grown exponentially. The existing legacy 4G/LTE network has been unable to meet the demands and keep up with the expectations in terms of speed, latency, number of connected devices and quality of service. The situation...

Full description

Saved in:
Bibliographic Details
Main Authors: Arjun Ray, Manish Kr. Yadav, Babu R. Dawadi, Krishna R. Bhandari
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10971944/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849722230068477952
author Arjun Ray
Manish Kr. Yadav
Babu R. Dawadi
Krishna R. Bhandari
author_facet Arjun Ray
Manish Kr. Yadav
Babu R. Dawadi
Krishna R. Bhandari
author_sort Arjun Ray
collection DOAJ
description With the increasing number of connected devices, the demand for mobile data has grown exponentially. The existing legacy 4G/LTE network has been unable to meet the demands and keep up with the expectations in terms of speed, latency, number of connected devices and quality of service. The situation demands a transition to 5G network; however, the transition comes with significant challenges for telecom operators, requiring substantial initial investments in infrastructure, technology, and human resources. This paper presents a novel two-phased approach to facilitate efficient 5G migration using machine learning and evolutionary game theory. This study primarily focuses on the development of a simulation-based analysis and a mathematical framework. In the first phase, a machine learning model is trained on Long Term Evolution (LTE) network simulation data to predict upgradability scores for existing 4G base stations, enabling a data-driven approach based on quality of service to prioritize Radio Access Network (RAN) migration. The second phase employs evolutionary game theory to observe the migration patterns for both core and RAN components of interconnected telecom operators in three distinct scenarios over a span of five years. The simulation considers critical factors such as revenue, customer retention/acquisition, traffic volume, coverage area, 5G based service demand, and human resources requirements, providing a framework for future network migration. Our research addresses the growing strain on legacy 4G networks caused by exponential growth in mobile data traffic and connected devices. By combining machine learning and game theoretic modeling, we offer telecom operators an approach to make informed decision regarding optimal migration strategies. This study contributes to the body of knowledge on network migration strategies and provides practical insights for telecom operators navigating the complex landscape of 5G deployment and migration.
format Article
id doaj-art-221c4b811cf447cbb1d7aa250e3dfdf3
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-221c4b811cf447cbb1d7aa250e3dfdf32025-08-20T03:11:25ZengIEEEIEEE Access2169-35362025-01-0113739497397410.1109/ACCESS.2025.356309610971944Game Theoretic Approach to QoS Oriented Machine Learning Model Development Toward 5G Network Migration PlanningArjun Ray0https://orcid.org/0009-0002-7144-9860Manish Kr. Yadav1https://orcid.org/0009-0006-9730-2778Babu R. Dawadi2https://orcid.org/0000-0001-6449-399XKrishna R. Bhandari3https://orcid.org/0000-0003-4064-1905Department of Electronics and Computer Engineering, Institute of Engineering, Tribhuvan University, Pulchowk Campus, Patan, Bagmati, NepalDepartment of Electronics and Computer Engineering, Institute of Engineering, Tribhuvan University, Pulchowk Campus, Patan, Bagmati, NepalDepartment of Electronics and Computer Engineering, Institute of Engineering, Tribhuvan University, Pulchowk Campus, Patan, Bagmati, NepalDepartment of Economics and Management, University of Helsinki, Helsinki, FinlandWith the increasing number of connected devices, the demand for mobile data has grown exponentially. The existing legacy 4G/LTE network has been unable to meet the demands and keep up with the expectations in terms of speed, latency, number of connected devices and quality of service. The situation demands a transition to 5G network; however, the transition comes with significant challenges for telecom operators, requiring substantial initial investments in infrastructure, technology, and human resources. This paper presents a novel two-phased approach to facilitate efficient 5G migration using machine learning and evolutionary game theory. This study primarily focuses on the development of a simulation-based analysis and a mathematical framework. In the first phase, a machine learning model is trained on Long Term Evolution (LTE) network simulation data to predict upgradability scores for existing 4G base stations, enabling a data-driven approach based on quality of service to prioritize Radio Access Network (RAN) migration. The second phase employs evolutionary game theory to observe the migration patterns for both core and RAN components of interconnected telecom operators in three distinct scenarios over a span of five years. The simulation considers critical factors such as revenue, customer retention/acquisition, traffic volume, coverage area, 5G based service demand, and human resources requirements, providing a framework for future network migration. Our research addresses the growing strain on legacy 4G networks caused by exponential growth in mobile data traffic and connected devices. By combining machine learning and game theoretic modeling, we offer telecom operators an approach to make informed decision regarding optimal migration strategies. This study contributes to the body of knowledge on network migration strategies and provides practical insights for telecom operators navigating the complex landscape of 5G deployment and migration.https://ieeexplore.ieee.org/document/10971944/Evolutionary game theoryintelligent approachmachine learningnetwork migration5G network deployment
spellingShingle Arjun Ray
Manish Kr. Yadav
Babu R. Dawadi
Krishna R. Bhandari
Game Theoretic Approach to QoS Oriented Machine Learning Model Development Toward 5G Network Migration Planning
IEEE Access
Evolutionary game theory
intelligent approach
machine learning
network migration
5G network deployment
title Game Theoretic Approach to QoS Oriented Machine Learning Model Development Toward 5G Network Migration Planning
title_full Game Theoretic Approach to QoS Oriented Machine Learning Model Development Toward 5G Network Migration Planning
title_fullStr Game Theoretic Approach to QoS Oriented Machine Learning Model Development Toward 5G Network Migration Planning
title_full_unstemmed Game Theoretic Approach to QoS Oriented Machine Learning Model Development Toward 5G Network Migration Planning
title_short Game Theoretic Approach to QoS Oriented Machine Learning Model Development Toward 5G Network Migration Planning
title_sort game theoretic approach to qos oriented machine learning model development toward 5g network migration planning
topic Evolutionary game theory
intelligent approach
machine learning
network migration
5G network deployment
url https://ieeexplore.ieee.org/document/10971944/
work_keys_str_mv AT arjunray gametheoreticapproachtoqosorientedmachinelearningmodeldevelopmenttoward5gnetworkmigrationplanning
AT manishkryadav gametheoreticapproachtoqosorientedmachinelearningmodeldevelopmenttoward5gnetworkmigrationplanning
AT baburdawadi gametheoreticapproachtoqosorientedmachinelearningmodeldevelopmenttoward5gnetworkmigrationplanning
AT krishnarbhandari gametheoreticapproachtoqosorientedmachinelearningmodeldevelopmenttoward5gnetworkmigrationplanning