Improving Traffic Forecasting Accuracy in Mobile Networks Through Machine Learning-Based Classification Techniques
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Network traffic forecasting is one of the activities that employs machine learning techniques for efficient network administration.This activity tries to manage user demands, minimize service disruptions, monitor user behavior in applications, and detect any issues before they arise [1].A base station uses traffic predictions for intelligent power use.This issue is examined using base station sleeping, mobile edge computing, network slicing, and extra power during periods of high demand.Both short-term and long-term forecasts are used for traffic, although occasionally medium-term forecasting is also required.Forecasting the amount of internet traffic is one of the most crucial duties in the proactive administration of contemporary telecommunications networks.Enhancing traffic demand forecasting's precision and effectiveness can aid ISPs in creating sensible business plans and boosting the sector's financial gains.Additionally, very accurate forecasting findings can be useful for advanced network design, route planning, both immediate and long-term business capacity planning, and improved resource management.To put it another way, an accurate traffic prediction system may help ISPs manage their networks in advance and guarantee improved network quality of experience (QoS) [2].Regardless of their location or time, mobile networks enable users to stay connected at all times, particularly while they are outside.The availability of mobile connection services is still lacking, despite the advancements made in hardware and software development [3].The antennas in certain places are too short to offer complete signal coverage, and the cellular phone companies' network has not yet been fully extended.Furthermore, future communication networks need to manage the restricted spectrum in order to accommodate the massive rise in heterogeneous wireless devices.In this regard, efforts are being undertaken to address the spectrum's coexistence, increase understanding of it, strengthen scheme authentication to enhance spectrum management and monitoring, and enable secure communications, among other things.Mobile network traces offer a multitude of information on the network and user activity.However, modeling and analyzing network traces is a difficult procedure.Particularly in the case of extremely congested networks, the unique insights gleaned from the network traffic analysis are extremely beneficial to both network operators and network customers [4].The network operators can split the network coverage area into several subregions if the traffic patterns in cellular networks are effectively predicted and projected.Consequently, traffic patterns may be used to assign network resources to sub-regions.Mobile customers would benefit from such traffic modeling in terms of Journal of Contemporary Education Theory & Artificial Intelligence
Fields of Study & Subjects
Traffic Prediction and Management Techniques
Software-Defined Networks and 5G
Internet Traffic Analysis and Secure E-voting
Publication details
- Publication date
- 30 March 2026
- Language
- EN
- Document type
- Dataset