Cooperative Computation Offloading Technique For NOMA-MEC Systems

With the rapid development of wireless communication and the Internet,it is possible to realize the connection between each wireless terminal.Applications in the fields of life services,smart office,and information notifications running on wireless terminals provide the users with a wealth of services,but they also cause computing load and energy load of wireless terminals to become more and more serious.Limited by battery capacity and poor resources for computing and caching,the users often encounter unbearable delay when using these terminal applications actually.Even for some important services,such delay may cause incalculable security problems.Mobile Edge Computing(MEC)technology is considered to be a key technology to solve this problem.MEC provides computing and caching resources for wireless terminals by sinking cloud services to the edge of the wireless network.Wireless terminals can offload their computation-intensive and delay-sensitive tasks that cannot be processed locally in time to their nearby MEC servers.After MEC servers have processed these tasks,the results will be fed back to these wireless terminals.Only when MEC servers have adequate computing resources and are deployed close to the users,low-latency offloading can be achieved.However,the number of wireless terminals and the complexity of services have been increasing.The deployment and maintenance costs of MEC servers are very high,and there are some problems which need to be resolved in current MEC technology.For example,MEC mainly uses the usual Orthogonal Multiple Access(OMA)technology,which has low spectrum efficiency and limited maximum accessible capacity.Non-Orthogonal Multiple Access(NOMA)is an advanced multiple access technology.Compared with usual OMA technology,NOMA has better spectrum efficiency and network access capabilities.NOMA can effectively improve the data transmission efficiency of wireless networks.In terms of resource allocation,current MEC doesn't make full use of the resources of the wireless terminals.Terminals cooperation technology can unite wireless terminals in the network.Terminals cooperation technology mainly includes technologies such as cooperative relay,cooperative computation offloading,and cooperative caching,which can more efficiently utilize the resources of wireless terminal devices.This paper is dedicated to applying NOMA and terminals cooperation technology to MEC system to give full play to the respective advantages of NOMA and terminals cooperation technology.The goal of this paper is to design a MEC system that can meet the requirements of larger access scale,more computationally intensive,and lower delay.In order to achieve the research target,this paper mainly completed the following tasks:(1)When direct transmission channel between the MEC server and a wireless terminal(UE)who has computationally intensive and delay-sensitive tasks does not exist,this paper exploited the idle computing resources of the cooperative terminal on the basis of existing cooperative relay technology.The tasks will be processed by the MEC server and the cooperative terminal jointly.In order to improve the data transmission efficiency of this system,the hybrid NOMA transmission strategy was adopted for the data transmission between UE and the cooperative terminal.Finally,the system model of Device-enhanced NOMA-MEC system was established.(2)There may be multiple idle wireless terminals that can be selected as the cooperative terminal.Minimizing the offloading delay is the main optimization target in this paper.A selection strategy of the best cooperative terminal was designed in this paper.(3)In order to achieve the optimal resource allocation,this paper analyzed and obtained the closed-form solution of the probability of successful offloading.By analyzing the relationship between the successful offloading probability and the task splitting factor,the paper designed a task split factor search algorithm which can ensure that the successful offloading probability of the system is maximized.The simulation results proved the correctness of the theoretical analysis results,and showed that the strategy proposed in this paper possesses lower average offloading delay and successful offloading probability than several other technical schemes.