Research On Efficient Wireless Offloading Technology For 6G

Wireless offloading in edge computing is an important method to realize efficient edge data processing.However,in the future,the number of access devices and task data generated by 6G network will surge,and most services have higher requirements in terms of delay and energy consumptions,which cannot be satisfied by the traditional offloading technology.At the same time,the new wireless technologies of 6G such as the space-air-ground integrated networks(SAGIN)and reconfigurable intelligent surface(RIS)bring opportunities to improve the efficiency of wireless offloading.How to apply 6G new wireless technology to achieve efficient wireless offloading is a key issue to be explored.To solve this problem,this thesis will study the SAGIN,RIS,non-orthogonal multiple access technology(NOMA)empowered wireless offloading with improved efficiency.The main work and contributions are as follows:(1)The SAGIN based wireless offloading with low-energy consumption.For the application scenario without ground network coverage,a hybrid wireless offloading scheme based on SAGIN is proposed.By deploying computing servers in Unmanned Aerial Vehicles(UAV)and satellite,users can achieve full coverage of wireless offloading with lowenergy consumption.The total energy consumption was minimized,under the constraint of the maximum time delay of tasks processing,and an optimization problem was proposed to optimize the computing resource allocation among UAVs,satellite and ground users.In order to solve this problem,the optimal solution of computing resource allocation is firstly derived under the given offloading objects.And a low complexity algorithm based on successive convex approximation(SCA)is proposed to optimize the offloading task proportion.The simulation results show that the proposed hybrid offloading scheme can reduce the energy consumption by 30.65%compared with the benchmark schemes.(2)Simultaneous transmitting and reflecting RIS and hybrid NOMA assisted wireless offloading with low time consumptionAiming at the problem that the efficiency of wireless offloading is reduced by wireless link blockage or limited communication coverage,a simultaneous transmitting and reflecting RIS(STAR-RIS)assisted multiaccess edge computing method is proposed.A hybrid NOMA(h-NOMA)technology is introduced to improve the offloading efficiency.The wireless offloading delay is minimized by jointly designing the offloading sequence and task division of edge nodes,the time allocation of each offloading stage of h-NOMA.and the RIS phase configuration.In order to solve this problem,the closed-form optimal solution of time allocation and task allocation is derived first,and then the complex problem is transformed into multiple simple subproblems.A low-complexity iterative optimization algorithm is proposed to solve the optimal solution.The simulation results show that compared with the benchmark schemes,the proposed scheme can reduce the offloading processing delay by up to 26.4%.(3)RIS assisted wireless offloading considering the energy consumption of RIS configurationThe energy cost of RIS configuration is further considered.An RISassisted wireless offloading scheme which considers the energy cost of phase configuration of RIS and wireless offloading is proposed.Under the constraint of limited computing resources,the aim is to reduce the total energy consumption of both RIS configuration and wireless offloading.An optimization problem is proposed to jointly optimize the RIS phase shift resolution,the RIS discrete phases,the base station computing resource allocation and the users’ transmission power.To solve this problem,the optimal closed solution of base station computing resource allocation is firstly derived,and the original problem is then decomposed into two subproblems:RIS element phase optimization problem and the user transmission power optimization problem.An alternative optimizing algorithm is finally provided.The simulation results show that the proposed scheme can reduce the energy consumption by 45%compared with the traditional scheme without considering the RIS configuration energy cost.