Delay-Optimization Strategies For Non-Orthogonal Multiple Access Assisted Edge Computing

With the continuous development of mobile Internet technology,the demand of mobile applications for computation resources is increasingly high,which has brought heavy computation burdens to mobile users.Edge computing and non-orthogonal multiple access have been considered as technical solutions that can effectively solve this problem.Edge computing enables MUs to offload complex computation tasks to edge servers,thereby improving computational efficiency.NOMA allows a group of wireless terminal users to transmit data in the same channel at the same time,and migitates co-channel interference through the successive interference cancellation mechanism,thereby improving the spectrum efficiency and energy efficiency in the network.This thesis combines edge computing technology with NOMA technology,and designs a NOMA based edge computing delay optimization management strategy for multi-access networks and multi-user networks.The main contributions are as follows:1.In a multi-access network,a delay optimization scheme for multi-access edge computing based on NOMA is designed,and the joint optimization of computation offloading and time allocation under the multi-access edge computing framework is realized.The formulated problem is a strict non-convex optimization problem.Therefore,we carry out a hierarchical transformation of the problem and decompose it into a bottom layer problem and a top layer problem: the bottom layer problem is about the optimization of the computation tasks allocation,and we turn it into a convex optimization problem for solving;The top layer problem is about the optimization of transmission delay,which is transformed into a unimodal function optimization problem for solving.Numerical simulation results show that the average relative error of the results obtained by our algorithm is less than 0.1% compared with the results obtained by Linear-Search,but the solution speed is improved by at least 90%.Therefore,the accuracy and effectiveness of our algorithm have been fully verified.2.In a multi-user network,a multi-user edge computing optimization scheme for joint computation task scheduling and uplink and downlink transmission time allocation based on NOMA is designed.The purpose of this scheme is to minimize the overall delay of the system while ensuring the MUs' computation requirements.Since the joint optimization problem is a non-convex optimization problem,it is very difficult to solve.This thesis explores the inherent convex optimization characteristics of the joint problem and proposes a corresponding layered algorithm,which can accurately obtain the optimal solution of the original joint problem.Numerical simulation results verify the accuracy and effectiveness of our proposed algorithm.Compared with LINGO,it can save more than 95% of the algorithm running time.At the same time,the performance advantage of the optimization scheme compared to other offloading schemes is also verified.