Predictive Analytics for 5G Network Performance Using Bayesian Networks
Abstract
This research aims at analyzing the application of Bayesian Network in the context of Predictive Analytics as a tool in F5 Networks for managing 5G network. With the help the probabilistic modeling features provided by the Bayesian Networks, it is possible to control prospective network performance problems as well as allocate the appropriate resources timely and efficiently. The approach that has been incorporated in the methodology includes the use of historical as well as real-time data, model building and calibration phase using the Bayesian Network model, and the incorporation of the trained model into the actual network management system. The findings show high prediction performance, cost savings benefits, and potential decreases in the network’s cessation rate and malfunctions. From these results, one can conclude that the application of Bayesian Networks in the 5G networks can significantly improve the predictive ability and performance of the new generation networks and become useful for network operators.