Research On Multiuser Superposition Transmission Technology Based On Factor Graph Structure

Nowadays,the explosive increment of the mobile Internet users and business has not only promoted the rapid development of 4G,but also put forward new requirements for 5G and post-5G mobile communication systems.It is difficult for traditional orthogonal multiple access(OMA)to meet the needs of massive data transmission in future mobile communications,and it is urgent to explore new multiple access technologies.Non-orthogonal multiple access(NOMA)allows multiple users to share physical resources,and has many advantages such as improving spectrum utilization,enabling large-scale connections and so on,having attracted widespread attention in the industry.Therefore,we will study the multiuser non-orthogonal transmission model based on the factor graph structure,and design and optimize the user superposition pattern.In this thesis,a multiuser superposition transmission system is introduced and a general factor graph model is refined.The specific research content is composed of three parts.Firstly,optimize the signal superposition protograph.This part focuses on the optimization problem of small-scale multiuser superposition transmission system,and proposes to combine genetic algorithm with optimization targets to search for the factor graph with the best superposition mode,so as to maximize the system transmission rate.The design principle for the optimal structure has been summarized.Secondly,extend and optimize the protograph.The optimized small-scale factor graph can be regarded as the protograph.It can be spatially coupled based on the two objective functions of average minimum girth and the girth distribution to further lengthen or eliminate short girths,improve system error performance,and help to achieve large-scale deployment.Thirdly,propose a cooperative NOMA scheme based on bilayer factor graph.The source to the destination constitutes the first layer and the relay to the destination constitutes the second.The design and optimizing method of the bilayer factor graph is introduced in detail,and the specific implementation scheme is given to maximize the total system transmission rate.Mutual information and bit error rate are mainly used to optimize and analyze the performance of the multiuser superposition transmission system.The simulation is performed under AWGN channel and Rayleigh channel,and the results verify the superiority of the various schemes proposed in this thesis.