Research On Caching Optimization And User Mobilitv Prediction For Mobile Edge Computing

In recent years,the rapid growth of mobile devices requires better architecture of mobile network and larger network capacity,the emergence of Mobile Edge Computing(MEC)decreasing the pressure on mobile core network by deploying the source of computing,storage and processing functions at the edge of the network.Based on the framework of Mobile Edge Computing,this paper studies the optimization of cache and the prediction of user mobility.In this paper,we first introduce the concept of Mobile Edge Computing,emphasize the characteristics of proximity and context-based information,explain the applications in several common scenarios,and discuss the future research directions.Based on the multi-layer cache architecture on MEC,this paper proposes a Cooperative Multicast-Aware Caching(CMAC)strategy,by using the technology of multicast and cooperative mechanism between base stations,to reduce the time-delay of user requests and improve the cache efficiency.And use the greedy algorithm to verify the effectiveness of the CMAC by comparing with the performance of the Multicast-Aware Caching(MAC).Compared with the MAC,the average latency by using CMAC is reduced by 13%.The paper also proposes the framework of Mobility Prediction as Service(MPaS)based on MEC,utilizing the model of Long Short-Term Memory(LSTM)to predict the user's mobility.The LSTM is trained and simulated on public dataset,and comparing the Artificial Neural Network(ANN)and LSTM prediction performance,to verify the effectiveness of LSTM user mobility prediction.