Prediction method of 5G high-load cellular based on BP neural network

Abstract

This paper used Back Propagation (BP) Neural Network algorithm to evaluate the network capacity of the 5G cellular, which was based on the daily network traffic of the it. Then the plan can solve the problem of limited 5G capacity in advance, reduce network delay, ensure the network performance and improve users’ perception. We firstly analyze the correlation of KPI indicators that involved in 5G high load cellular to determine strong correlation indicators. Then the BP Neural Network is used to train the KPI sample data and output the simulation results. Finally, according to the result, the cellular network capacity will be evaluated by the result which will also determine whether the cellular has risk of high load.