Energy Optimization Algorithm Based On Sequential Resource Discovery In Mobile Edge Computing

Mobile edge computing(MEC)has become one of the key technologies for 5G networks,providing a powerful platform to address the issues such as latency,energy consumption,and capacity of future networks.Energy consumption is the problem that cannot be ignored.It is particularly important to solve the problem of high energy consumption and improve the energy efficiency of MEC systems through the study of key issues such as resource discovery,resource allocation,power optimization and computation offloading.However,current research on these problems focuses on separate optimization,or partial joint optimization,as part of the entire MEC system,it is necessary to jointly optimize these problems.In addition,in relay-assisted MEC system,the task is only divided into two parts for execution,and current research work on energy consumption all assumes that the task is offloaded to the first available node,resulting in the advantages of node diversity being difficult to be fully utilized.Therefore,this paper proposes an algorithm based on sequential resource discovery to study the energy consumption problem of MEC system.The specific research work is as follows:1)Energy optimization algorithm based on sequential resource discovery in mobile edge computing: This section proposes a sequential Resource Discovery-Allocation-TransmissionOffloading(DATO)algorithm to jointly optimize the design of resource allocation,task offloading and transmission power to minimize the energy consumption of the MEC system.The resource discovery policy indicates when resource discovery should be stopped and offload the task to the selected server.The resource allocation policy indicates which server the task should be offloaded to.The transmission policy determines the optimal transmission power that should be used when transmitting data to the edge server.The offloading policy shows the optimal ratio of data offloaded to the edge server.In addition,by further optimizing the resource discovery sequence to determine which resource should be detected at each step,the sequence optimization has a significant impact on the speed of resource discovery,which in turn affects the system energy consumption.The above problem can be formulated as a random sequential decision-making problem and the optimal strategy can be obtained using dynamic programming.By parameterizing the above problem,the expressions of optimal transmission power and offloading ratio can be further deduced.The experimental results show that the proposed optimization algorithm based on sequential resource discovery can reduce the energy consumption of the system.2)Energy optimization algorithm based on sequential resource discovery in relay-assisted MEC: On the basis of the first problem,in order to further explore the diversity of relay nodes and remote servers,a relay-remote selection and computation offloading energy optimization algorithm based on sequential resource discovery is further proposed.Under the comprehensive consideration of different communication and computation conditions between the user and the relay nodes,the relay nodes and the remote servers,the user needs to first perform sequential resource discovery when selecting the relay node,and each time after finding an available node,it must decide whether to start computation offloading or continue to discover the next node.The offloading policy indicates the optimal ratio of tasks to be executed locally,on relay nodes and remote servers.Similar to the first work,the problem can be formulated as a random sequential decision-making problem and dynamic programming can be used to obtain the optimal strategy.Simulation results show that the strategy proposed in this paper is better than the random relay-remote selection and shortest-distance server selection strategies in terms of reducing energy consumption.In addition,according to the experimental results,the optimization of the offloading ratio has a more significant impact on the overall energy consumption.