Sparse Code Multiple Access System Based Optimization Design Of Code Modulation

For the application scenarios of large-scale mobile devices and IoT devices in 2020 and beyond,the 5th generation of mobile communication system(5G)needs to solve the massive connection problem urgently that the current mobile communication system cannot carry.Based on the current multiple access technology,the amount of users is limited by the number of orthogonal resources,and obviously,it cannot meet the requirements of 5G about large-scale connection.Sparse Code Multiple Access(SCMA)is a promising non-orthogonal multiple access technology that can achieve overload thanks to non-orthogonality,and consequently,it can support a larger number of terminals.Moreover,SCMA utilizes the sparseness of the codebook,which make it adopt the Message Passing Algorithm(MPA)to obtain the efficient detection performance.In addition,Low Density Parity Check Code(LDPC)is a type of linear block code which has significant advantages of low complexity decoding and superior coding performance due to the sparse structure of its check matrix.In conventional point-to-point transmission system,multi-level coding,which protects different information bits based on a plurality of component codes to approximate the theoretical channel capacity of the joint coding modulation,achieves the excellent performance.Under multi-user access,SCMA adopts a new high-dimensional codebook as the modulation method.The component code rates design of SCMA based on multilevel coding is not known.In order to solve this problem,this paper first studies the SCMA system based on Parallel Independent Decoding(PID)multi-level coding(MLC),and designs the optimal code rates of each level for each user.Firstly,the LDPC-coded SCMA multi-level coding transceiver model is given in this paper.Based on the MPA detection results,the bit LLR expressions of each level are derived.Secondly,based on mutual information and the LLR information derived above,this paper deduces the expressions of the total capacity and capacity of each level for each user.Through digital simulation,this paper presents the capacity comparison of each level for each user in SCMA system with 6 users sharing 4 resource blocks under modulation with order of 4and 8,and then obtains the optimal code rates combination of each level.Furthermore,in connection with the LDPC-coded SCMA system,this paper utilizes density evolution theory to analyze its coding performance and optimize the degree distribution of irregular LDPC codes.However,DE cannot be directly employed to analyze and optimize multi-user SCMA systems.First,because of the existence of asymmetric constellation modulation in the SCMA system,there is a conflict with the channel symmetry condition of the DE.In order to solve this problem,this paper designs a improved model based on the system model with individual detection and decoding.It is proposed to add a channel adaptor to guarantee the channel symmetry condition and add a LLR converter to ensure the consistency of data types between modules.Second,the traditional DE is a single-user algorithm and correspondingly it is not suitable for SCMA multi-user scenarios.For this issue,this paper advances a multi-user min-max design criterion,and devises a multi-user density evolution algorithm for regular codes and irregular codes based on min-max criterion.Thus,its analysis function under SCMA multi-user system has been implemented.Based on the above proposed methods,the performance threshold of SCMA system under different LDPC code structures is obtained by experimental simulation,and the optimal distribution of irregular LDPC codes under the corresponding code structure is given.