Research On Resource Allocation Optimization Of NOMA-MEC Wireless Networks

In future wireless networks,mobile edge computing(MEC)is considered to be a key technology to reduce network pressure,which can improve user's quality of service(Qo S)by offloading the computing-intensive or delay-sensitive applications to the edge of the wireless network.However,with the rapid increase in the number of mobile equipment,the limitation of radio resources will cause congestion in the offloading of computing tasks,and this congestion will severely prolong the transmission delay of computing tasks and the overall delay of completing computation tasks,which has become the main limiting factor for MEC.Non-orthogonal multiple access(NOMA)has been regarded as a key enabling technology in next-generation wireless communication network due to its superior spectrum efficiency,and the combination of MEC and NOMA can effectively reduce the execution time and energy consumption of computing tasks.Therefore,this thesis focuses on the problem of delay and energy consumption of computation offloading in the NOMA-MEC network.The specific research results are as follows:Firstly,the NOMA-MEC network without inter-cell interference is considered,in which multiple users are allowed to offloading computation tasks to the same base station via NOMA protocol,and each base station is assigned an orthogonal subchannel.The problem of joint offloading decision,transmission power and computing resource allocation is studied in order to maximizing the users' task offloading gains,which is measured by the weighted sum of reductions in execution delay and energy consumption of offloading computation tasks to the MEC server compared with local computing.In order to deal with the mixed integer nonlinear programming(MINLP)problem,it is proposed to decompose the original problem into tractable computing resource allocation(CRA),transmission power(TP)and offloading decision(OD).For the CRA problem,the convex optimization technology is used to obtain the optimal solution,and an improved whale optimization algorithm is proposed to solve the TP and OD problems.The simulation results show that the performance of the proposed algorithm is close to the optimal solution corresponding to the exhaustive method,and the running time of algorithm is much shorter,which verifies the correctness and effectiveness of the proposed algorithm.Also,the performance of the proposed algorithm is better than the standard whale optimization algorithm and particle swarm optimization algorithm.Furthermore,the proposed NOMA offloading scheme has better performance than the conventional offloading schemes.Secondly,the NOMA-MEC network with inter-cell interference is considered,in which the NOMA technology is used to offloading computation tasks to the BS equipped with a server in each cell,and each BS shares multiple same subchannels.In order to maximize the users' task offloading gains,the problem of joint task offloading decision,uplink transmission power of users and computing resource allocation at the MEC servers are studied.Due to the combinatorial nature of this problem,solving for optimal solution is difficult and impractical for large-scale networks.In order to solve this problem,the original problem is decoupled into a resource allocation problem with fixed task offloading decision and a task offloading problem that optimizes the optimal-value function corresponding to the resource allocation problem.Convex optimization technology and improved whale optimization algorithm are used to solve the problem of computing resource allocation and transmission power respectively,and the improved simulated annealing algorithm is proposed to find the approximate optimal solution for the problem of task offloading decision.The simulation results show that the performance of the proposed algorithm is close to the optimal solution but with lower complexity.Compared with the traditional OFDMA and greedy offloading schemes,the proposed NOMA offloading scheme significantly improves the offloading utility of users.This thesis studies the problem of resource allocation optimization of NOMA-MEC wireless network under two different system models,and jointly optimization of task offloading decision and resource allocation to maximize the offloading utility of each user.The obtained results can provide some useful insights into the design and optimiz ation of future NOMA-MEC networks.