Research On UAV Mobile Edge Computing Network Based On User Traffic Prediction

With the advent of 5g era,the number of mobile user devices is growing rapidly,and the data traffic is also growing exponentially,which brings great pressure to the base station in the cellular network.Although 5g communication base stations have been gradually popularized,5g communication network based on fixed base stations will still face challenges.The base station operating load in cellular network is limited.In the face of temporary explosive traffic,when the traffic scale exceeds the base station operating load,it will cause network congestion and affect the quality of user experience(Qo E).One of the effective ways to solve this problem is to support mobile edge computing of UAV.If we can predict the time and scale of potential explosive traffic,and then deploy UAV on demand,we can reduce the energy consumption and network operation cost of UAV.Based on this,this paper proposes an UAV mobile edge computing network scheme based on user traffic prediction,which is an effective way to solve the cellular network congestion.Firstly,particle swarm optimization(PSO)algorithm is used to improve BP neural network to get PSO-BP prediction model to predict the time and scale of potential explosive traffic.Secondly,in the mobile edge computing network,the problem of minimizing the energy consumption of ground user equipment and UAV in the process of task execution is studied by jointly optimizing the task unloading decision of ground user equipment,resource allocation mechanism and UAV flight trajectory.Finally,the energy consumption model of UAV mobile edge computing network is established based on the prediction model based on the prediction results of PSO BP model and the energy consumption optimization model of mobile edge computing network.The energy consumption model is simulated and analyzed,and the energy consumption is compared and analyzed under the same optimization algorithm,different prediction model and different optimization algorithm under the same prediction model.The simulation results show that the combination of PSO BP prediction model and mobile edge computing network energy consumption optimization model can reduce the number of UAVs required,optimize the energy consumption of UAVs and user equipment,and reduce the cost of operators in solving the problem of explosive flow.The research in this paper can well solve the impact of explosive traffic,effectively avoid network congestion,provide users with better quality of service(QoS),and reduce the cost of operators.