Research On Resource Allocation And Computing Offloading Strategy In Mobile Edge Computing Scenario

With the rapid development of mobile Internet,intelligent mobile devices play a more and more important role in people's basic necessities of life,and the functions of various emerging applications are becoming increasingly powerful.Although the performance of mobile devices has been greatly improved compared to the past,they still have some problems such as insufficient computing resources and limited battery capacity,so they can not meet the needs of various applications in time.In view of this,a new network architecture,mobile edge computing(MEC),emerged as the times require,which can provide rich computing and storage resources for mobile devices.Based on the background of MEC,this paper studies the problems of computing offloading and resource allocation from single user and multi-user scenarios.The main research contents are summarized as follows:(1)In the single user scenario,there are several independent tasks waiting for being processed on the user,and there are multiple MEC servers around the user for offloading.Considering the difference in computing power between mobile devices and MEC servers,as well as the quality of the wireless channel between mobile devices and e Node B base stations,this paper proposes a binary vector representation of computing tasks based on onehot encoding and an alternate optimization offloading strategy based on discrete grey wolf optimization algorithm,which are used to select the most suitable server for the task to be offloaded and allocate the corresponding transmit power.The simulation results show that the proposed strategy can effectively reduce the total energy consumption of the system under the constraint conditions.(2)In a multi-user scenario,there is a MEC server and several user mobile devices,and each user has only one computing task waiting for being processed,either completely left in the local device or offloaded to the MEC server for computing.Although the MEC server has more computing resources than mobile devices,it is still limited for massive computing requests and cannot meet all the offloading requirements.Therefore,a discrete grey wolf optimization algorithm combining the cosine convergence factor and the crossover operator is proposed to jointly optimize the offloading strategy and the server computing resource allocation vector,which is used to allocate the most appropriate computing resources for the tasks that need to be offloaded.The simulation results show that compared with others,the algorithm can significantly reduce the total energy consumption of the system under the premise of ensuring that the server computing resource allocation does not overflow.