Traffic Prediction System for Heterogeneous green Cellular Networks (Hetnets)

Abstract

The development of cellular technology results in a rapid increase in cellular network traffic. For networks to improve their quality of service (QoS), time-series models which are used to forecast the cellular traffic have become crucial.To maximize the use of available resources, cellular network loading modelling and forecasting are necessary for the allocation of bandwidth provisioning while maintaining the maximum network utilisation. It is required to improve the network's performance and quality even if many users are present in the network. This can be achieved by enhancing the network performance with reduced energy consumption, which can then be used to simplify and ease the lives of consumers by appropriately serving their demands. The novelty introduced in this work is to create a model that can assist with accurately forecasting load traffic in cellular networks. This paper discuss about a regression model with different algorithms to predict the cellular traffic. The intelligent model predicts the traffic with the help of real time traffic data set obtained from Kaggle dataset. The comparison results of the traffic prediction model for the three regression algorithms are presented. The suggested solution performed better than expected when it came to predict cellular network traffic. The implementation of prediction in heterogeneous cellular networks provides a pathway for the energy efficient green cellular networks.