Research On Multi-priority Task Scheduling Algorithms For Mobile Edge Computing

As one of the core technologies of 5G,Mobile Edge Computing(MEC)plays an increasingly important role with the promotion and popularization of 5G network.MEC has important applications in the field of telemedicine due to its proximity,high bandwidth and low delay.There are many kinds of telemedicine industry.Typical medical scenarios include remote first aid and remote monitoring.The requirements for task delay and energy consumption are also different.In some scenarios,users may be in a mobile state.Therefore,this thesis aims at three typical scenarios of telemedicine,considering practical problems such as task priority and user mobility,designs corresponding task migration strategies in different application scenarios,and deeply discusses the application of MEC in telemedicine.Firstly,consider the application scenario of MEC in remote first aid.By adding task priority mechanism to MEC task migration model,the average processing delay of mobile device task is optimized.According to the number of current users,task attributes,network environment and MEC server resources,the task is dynamically scheduled to MEC server or mobile device for processing.Through modeling analysis,it is found that the model is a 0-1 integer programming problem,and it is difficult to get the optimal solution directly.In this thesis,a reverse auction algorithm based on task priority is designed to allocate channel resources.After a single round of auction,task scheduling decisions are obtained.Compared with other task migration algorithms,the introduction of task priority mechanism greatly guarantees the real-time and high quality of remote first aid tasks.Secondly,considering the application scenario of MEC in remote monitoring,the user's biased factor of task processing delay and mobile device energy consumption is introduced on the basis of task priority,so that the algorithm can flexibly schedule tasks according to the current task's delay sensitivity and device's power consumption,reduce the average processing cost of tasks,and realize the collaborative optimization of delay and energy consumption.Through modeling and analysis,it is found that the problem belongs to the dynamic non-cooperative game with limited steps and complete information.A time delay energy consumption optimization algorithm based on game theory is designed,and task scheduling is carried out on the basis of demonstrating the existence of Nash equilibrium in the model.Compared with the existing algorithm,this algorithm can meet the user's demand for delay and energy consumption flexibly,and guarantee the quality of remote monitoring service through task priority mechanism.Thirdly,we study the mobility of users in some telemedicine scenarios.Under the constraints of task priority,user mobility and MEC server resources,the base station sequence of user mobility is obtained through the mobile trajectory prediction technology,and the task is scheduled to the appropriate server for processing to optimize the average processing delay of the task.Through modeling analysis,it is found that the problem is a 0-1 integer programming problem,and a heuristic optimization algorithm is designed to schedule the task.Compared with other algorithms,this algorithm reduces the average processing delay of tasks,improves the resource utilization rate,and ensures the processing efficiency and quality of telemedicine services.