Mobile Edge Computing Resource Optimization Based On Multi-dimensional User Difference

With the further development of mobile communication networks and the advent of the era of the Internet of things,the number of network edge devices is increasing rapidly,and the data traffic shows an explosive growth.The proposal of mobile edge computing has become the most effective way to alleviate the link pressure.At the same time,relying on the centralized processing of cloud computing center,mobile edge computing processes and caches part of the user data in the edge devices,greatly increasing the real-time and security of user data,therefore,mobile edge computing in the new generation of mobile network has attracted the attention of researchers and network providers.There are the following defects in the research of mobile edge computing:(1)the subjective factors of users are not considered;(2)the idea of joint optimization is lacking;(3)the network topology and user mobility are unreasonable.In this paper,a three-layer network model of mobile edge computing is adopted,and the simulation platform is implemented by computer code.Edge devices are divided into two layers:micro base station and macro base station.Because the macro base station covers a wide range,has a large cache capacity,and is relatively far away from the user,the effect of caching files in the macro base station and the micro base station is very different.This paper considers video,picture and text as three kinds of services,and based on multi-dimensional user differences such as file characteristics,device performance and user characteristics,puts forward a calculation method of user satisfaction.The algorithm clearly defines video transmission continuity and revenue delay.In this paper,user satisfaction is the decisive factor to measure network performance.This index is based on objective factors and considers users' subjective factors as well.Based on the three-layer network model and the introduction of video service,an algorithm that can store and test most of the solution space,and consider the transmission continuity and revenue delay of video,is needed to solve the edge cache problem.In this paper,the feasibility of a deep Q network is demonstrated.At the same time,the interface strategies between environments and agents are designed,and a compromise strategy of convergence time and performance is selected.Based on user satisfaction and a deep Q network,this paper proposes an edge cache update algorithm.Compared with the traditional algorithms,this algorithm has better performance,and in the aspect of video services,the performance improvement is more obvious.Under the scenario of distributed access mode,this paper demonstrates the feasibility of the resource allocation method based on belief propagation.A user resource joint optimization algorithm based on belief propagation is proposed,which takes the number of users reaching the lowest transmission rate as the optimization target and the user ownership and bandwidth allocation as the optimization object.The optimization target is mainly for the video service,considering the impact of equipment performance and user characteristics on the video service.After the user attribution is determined,each base station needs to be used as the bandwidth allocation unit to distribute the bandwidth to the users for which the base station provides services.This paper proposes a bandwidth allocation algorithm suitable for this scenario.Then,this paper combined edge caching and resource allocation to optimize,and the effect was that the overall user satisfaction decreased,while the user satisfaction for video services increased.Finally,user mobility is added to the network model to make the simulation scene closer to the reality.The simulation results show that the performance of the proposed joint optimization algorithm is worse than that of the traditional algorithm in terms of user satisfaction of all services,but it still has great advantages in terms of user satisfaction of video services.