Research On Key Technologies Of Coordinated Communication Under Mobile Edge Computing

The rapid development of intelligent terminals has enriched people's life.However,delay sensitive and compute-intensive applications bring challenges to the computing capacity,energy consumption and computing delay of mobile terminals.In order to solve these problems,mobile edge computing has been proposed as a development of mobile cloud computing.Mobile edge access points(M-APs)are closer to users,which save processing energy and time.It has become a hot spot of current technology research and has a broad application prospect.Based on mobile edge computing and coordinated communication architecture,offloading strategies in three different mobile edge computing systems are studied in this thesis.The main work is as follows:Firstly,the architecture model,key technologies and application scenarios of mobile edge computing are introduced in details.The process of task offloading and the offloading strategy under different goals are emphatically characterized.Second,the vertical coordinated offloading computing under the mixed architecture of mobile cloud computing and mobile edge computing is studied.Three offloading computing modes are considered.Minimize the expected expenditure of the system by optimizing the computing ratio of each layer and the distribution density of M-APs.Variables of each layer's computing ratio are decoupled by the arithmetic geometric mean inequality.Then the problem is decomposed into two sub-problems.The Lagrangian duality and KKT conditions,the penalty function method and the multi-peak function adaptive evolution algorithm are used to iteratively solve the two subproblems.Simulation results show that the proposed optimization algorithm greatly improves the cost and delay compared with the local computing.Third,the joint optimization of transmit power and task offloading strategy under the horizontal coordinated communication and distributed computing is studied.In the optimization problem,variables are in coupling form.The original non-convex problem is decomposed into two subproblems.Then,Lagrangian duality,KKT condition and penalty function method are used to solve the problem iteratively.Simulation results show that,compared with the average system resource and task division,the proposed algorithm can effectively reduce the system expenditure.The different value of weight factor can be applied to the different scenarios.Fourth,the multi-user and multi-M-AP horizontal coordinated communication under mobile edge computing and distributed computing is studied.Considering fairness,subcarrier selection,transmit power and task offloading strategy are jointly optimized.The problem of minimizing the maximum offloading computing delay is studied.First,the integer variable is scaled.Then,based on fractional programming and auxiliary variables,the min-max problem is transformed into a min form.Finally,the problem became convex form by Taylor first order expansion.By simulation,the proposed algorithm can converge quickly.Compared with the system delay under non-optimal strategy,the performance of the offloading system is greatly improved by the proposed algorithm.