Mobile Edge Computing Offloading Optimization Based On Machine Learning And Game Theory

In the era of Big Data,the amount of data generated by computing tasks has increased steeply,and the computing efficiency of traditional cloud computing can no longer meet the demand,so Mobile Edge Computing(MEC)has emerged.However,the over-reliance on MEC servers leads to excessive computational pressure,so it is necessary to accurately predict the load of MEC servers for resource allocation.In this paper,the mobile edge computing offloading algorithm is optimized based on the prediction of MEC server load resources,and the specific work is as follows.In order to solve the MEC server computing resource allocation problem,it is especially critical to accurately predict its load resource data.Considering the time series characteristic of load resource prediction data,a time series prediction model based on convolutional neural network and long and short-term memory neural network is proposed to predict and analyze its CPU utilization,which has a large impact.Combining with the characteristics of MEC server and the large scale of the prediction data,the convolutional neural network is used to extract features for dimensionality reduction of the time series data,and the long and short-term memory neural network is used for time series prediction,and the Attention mechanism is introduced to focus the important information of the series to finally obtain the prediction results.Finally,the root mean square error and the average absolute value error are used as the judging indexes,and the proposed C-LSTM prediction model has higher prediction accuracy compared with the baseline model of other traditional time series prediction models.In the mobile edge computing scenario,end devices are biased to offload to MEC for computation,while the load resources of MEC servers are limited.By introducing vehicles and UAVs to assist the MEC in task computation,the server task computation is mathematically modeled,however,without a reasonable computational task offloading strategy,the computational task transfer becomes disordered,resulting in a large computational cost.By using game theory for computational offloading solution,the computational offloading decision problem is modeled as a computational offloading game,which is proved to be a potential game and there is a Nash equilibrium through the study.Considering the effects of computational delay and energy consumption,the proposed computational offloading game algorithm effectively reduces computational delay and energy consumption and achieves cost minimization compared with other methods.