Mobile Edge Caching And Computing In 5G Network

In order to meet the needs of large-scale mobile devices accesses and rapid growth of communication capacity,small cell will achieve ultra dense deployment in the next gener-ation mobile communication system(5G),and storage and computing resources of small cell can be utilized to provide ubiquitous computing support for mobile applications,such as augmented reality game.However,this scheme will increase the load of the system back-haul link,and will bring huge energy consumption.In order to solve these problems,many researchers put forward the scheme of mobile edge caching and computing.However,the existing schemes about mobile edge caching and computing have the fol-lowing problems:First of al,most of the edge caching schemes are based on fixed network topology,while ignore the user mobility.Secondly,in order to solve the problem of high energy consumption of 5G network,it is feasible to use renewable energy.However,the randomness of renewable energy leads to the dynamic computing ability of the edge cloud server.Thus,the existing computing offloading strategies powered by main grid are diffi-cult to be applied in this scenario.Finally,user mobility results D2D(Device-to-Device)based edge computing(mobile cloudlets)is dynamic,which can lead to the failure of the computational task offloading.Facing with the above problems and challenges,this thesis is conducted from the following four aspects:1.User mobility in edge caching is studied.Through analyzing the effect of user mobility on caching of the small cell and the user device,this thesis puts forward the op-timization mobility-aware caching strategy,which is proved that the problem is NP-hard.Based on the submodular optimization,the greedy algorithm is used to solve the problem.The experimental results show that,compared with traditional cache strategy,this strategy has significantly improved cache hit ratio.2.Randomness of contact time between the user,and the user and small cell is studied.Based on the code cache,this thesis sets up the content placement model with maximization cache hit ratio and minimization transmission energy consumption.Through the solution of the model,this thesis proposes a green mobility-aware code cache strategy.The experi-mental results show that the proposed scheme has the highest cache hit ratio and the lowest transmission energy consumption compared with other cache strategies.3.Mobile edge cloud computing powered by renewable energy is studied.Based on the analysis of the renewable energy,a model for minimization the delay of user computing task and energy consumption of main grid is built.The model is decomposed into two sub-problems of computational resource allocation and task placement by using the alternating optimization method.By solving the two sub-problems,the computing task offloading strat-egy under the renewable energy supply is obtained.The experimental results show that the proposed algorithm can reduce at least 20%task delay and 30%energy consumption com-pared with the random computing offloading scheme and the uniform computing offloading scheme.4.Connection unreliability of mobile edge computing is studied.This thesis breaks the traditional reliance on mobile cloudlet connection of D2D,and propose the opportunistic mobile ad hoc cloudlet service mode(OCS).At the same time,the task delay and energy consumption are analyzed,which result in optimal offloading strategy.Finally,the selection algorithm of computing tasks in the remote cloud,mobile cloudlet and this mode is given.The experimental results show that this model is superior to the other two models in terms of delay and energy consumption when the ratio of data size after task execution over original data size associated with the task is smaller than 1((i.e.r<1))and the contact rate of two mobile devices A is larger than 0.00014.In summary,mobile edge caching and calculation proposed by this thesis can make full use of computing resources of edge cloud,users mobility and the dynamic renewable energy,which can provide cache and computing services for users,and improve the quality of the user experience.