Non-orthogonal Multiple Access Technology Based On Joint Constellation Design

In recent years,with the rapid development of Internet and Internet of Things technology,device access volume and data traffic show explosive growth,which brings a series of severe challenges to the future mobile communication technology.As a key technology of future mobile communication technology,non-orthogonal multiple access technology(NOMA)is a promising technology,which can meet the demand of largescale access of devices and improve resource utilization.In this paper,NOMA in power domain and coding domain are studied.Then the resource allocation in power domain NOMA and the design of joint constellation in coding domain NOMA are explored.In this paper,the principle and characteristics of NOMA are introduced respectively from the perspective of power domain and coding domain,then the two NOMA technologies are further studied.The research of NOMA in power domain is carried out in a downlink with perfect channel estimation,and the allocation of resources was mainly studied.This paper fixed the power allocation method in the process of resource optimization,designed the allocation method of sub-channel resources,and proposed a method based on multi-label classification to design the subchannel assignment.The results show that the proposed method can achieve the average throughput which is very close to the optimal solution under the condition of perfect channel estimation.However,resource allocation in power domain NOMA depends on instantaneous channel state information,so it is not applicable in the scenario where accurate channel state information is difficult to obtain.In this case,NOMA in the coding domain can be considered.This paper applied the coding domain NOMA in cluster-based vehicular communications,and a system model was built.In order to make the joint constellation obtain better shaping gain,a serially concatenated scheme is proposed and the superposition matrix of multi-user transmission over resources is optimized.In order to further improve the performance and reduce the use of resources,it is considered to apply constellation rearrangement technology in the above model to optimize the mapping mode of constellation points.In the optimization process of the constellation,genetic algorithm is used as the optimization method,and the average pair error probability and cutoff rate are used as the performance metric to measure the system performance.The results show that superposition matrix optimized by serially concatenated scheme has better performance than other schemes,and the performance can be further improved by using constellation rearrangement.