Font Size: a A A

Research On Task-Split Parallel Transmission Offloading Network Based On NOMA-MEC

Posted on:2022-05-05Degree:MasterType:Thesis
Country:ChinaCandidate:M F WangFull Text:PDF
GTID:2518306605471644Subject:Master of Engineering
Abstract/Summary:PDF Full Text Request
Mobile edge computing provides users with a close range,high speed,and low delay by deploying servers at network access points and base stations.Extended computing services are a promising method of enhancing computing services.Non-orthogonal multiple access is considered to be the key technology of the next generation of wireless networks,which allows multiple users to be multiplexed with different powers on the same sub-channel.After successive interference cancellation,the superimposed signals of different users can be decoded correctly,thereby significantly improving spectrum efficiency.The combination of mobile edge computing and non-orthogonal multiple access technology can effectively reduce the delay and energy consumption of the system and achieve service requirements of low latency,low energy consumption,and high reliability in the fifth generation of mobile communications and even the future mobile communications.Current NOMA-MEC scenarios consider use uplink NOMA to offload computing tasks to MEC server,the use of downlink NOMA for offloading tasks is rarely considered.In addition,in large-scale NOMA-MEC communication scenarios,always has a large number of mobile users and MEC servers.In such communication scenarios,best matching strategies and computing task-split decisions can effectively improve the performance of system,but how to formulate the matching strategy between the user and the server,and computing task offloading decision is a question worthy of discussion.On this basis,this thesis proposes a NOMA-MEC task-split parallel transmission offloading network under multi-user and multi-server conditions.The main contents and contributions are as follows:1)This thesis assumes that the network has sufficient computing resources,that is,the number of MEC servers is much greater than the number of users,and each user has computing-intensive and delay-sensitive critical tasks that need to be offloaded to nearby MEC servers.This thesis considers that each user can only pair with two MEC servers,and each server can only pair with one user.According to the user's channel situation,the task is split into two parts,where one part is computed locally,and the other part is offloaded to the MEC server.For the task offloaded to the MEC server,the user also needs to perform a second split and use the downlink NOMA technology to offload to two MEC servers respectively.2)In view of the above research scenarios,this thesis proposes a resource optimization model for minimizing the total energy consumption of the system under the time constraint.Since the proposed optimization model is a mixed-integer non-convex optimization model,in order to reduce the computational complexity,it is decomposed into three sub-optimization problems: optimal offload delay and transmission power,optimal task-split ratio analysis,and user-server pairing,and then the solution of each sub-optimization problem is derived.Because the value of the task-split ratio will produce different results with the monotonicity of the sub-problem function,the solution of the task-split ratio is discussed in detail.3)Finally,this thesis proposes two optimization algorithms to solve the total energy consumption problem of the proposed multi-user multi-server NOMA-MEC task-split parallel transmission offloading network,and we further compares the performance differences between the two algorithms through simulation.In addition,this thesis compares the proposed system scheme with the OMA-MEC scheme under the same conditions,where all tasks are locally computed and all tasks are offloaded to the MEC server for computing.The simulation results show that the proposed scheme effectively reduces the energy consumption of the system compared with other schemes.
Keywords/Search Tags:NOMA-MEC, Task-split, User-pairing
PDF Full Text Request
Related items