ABSTRACT
Life is nearly impossible today without a cellular device, but the exponential increase in users over the last decade has vastly increased mobile data traffic, which slows the user experience. Network mobile traffic prediction is a feature of the new fifth-generation mobile network (5G) to perform smart and efficient infrastructure planning and quality of service management. In order to improve resource usage and user experience, mobile traffic predicts models enable the users experience for demand of traffic. Regression models from machine learning algorithms have been introduced for predicting mobile data traffic. Our work depends upon a comparative study of traffic prediction systems and congestion prediction including decision tree, logistic regression, and neural networks, and our method reduced and controlled traffic congestion. Moreover, performance of the three prediction models is verified in relation to various view such as key objectives, implicit method, benefit, inferences, and performance measurements. Our method satisfies all the constraints of the problem.
