Research On Key Technologies Based On SCMA System

By 2020,mobile communication will enter the 5G era.The transformation from mobile Internet to Internet of Things is the biggest change that 5G brings compared with 4G.However,in the context of the current shortage of time and frequency resources,it is a huge challenge to achieve large-scale access to the Internet of Things.For this reason,5G large-scale access is the key to the successful application of 5G vertical industry in the future.Sparse Code Multiple Access(SCMA)is a new non-orthogonal multiple access scheme for 5G communication.This scheme supports multiple mobile terminals to reuse the same empty port resources to achieve resource sharing,which greatly increases the number of system visits in the Internet of Things scenario,and truly solves the problem of mass access.At the same time,because of the sparse structure of codewords,SCMA can use message passing algorithms(MPA)with higher decoding efficiency to decode,which has good practical prospects.Firstly,this paper introduces the principle of SCMA and compares it with Orthogonal Frequency Division Multiplexing(OFDM)to illustrate the superiority of SCMA technology,and explains the codebook design method.At the same time,a simple codebook design method is given,which lays a foundation for the research and Simulation of decoding algorithm.Secondly,aiming at the problem of low decoding efficiency of SCMA multi-user detection algorithm,a MPA detection algorithm based on partial codeword spherical decoding is proposed.The algorithm reduces the number of user codewords participating in iteration by setting spherical radius.According to the non-orthogonal characteristics of SCMA,a new decision mechanism of channel quality is introduced to further reduce the computational complexity of MPA algorithm.Meanwhile,the initial probability equalization strategy of partial codewords is adopted to effectively improve the convergence rate of the algorithm.The experimental results show that the PCSD-MPA algorithm reduces the complexity of MPA almost without affecting the system performance by setting a reasonable spherical radius,and the computational complexity of the algorithm decreases withthe increase of SNR.Then,in order to further reduce the computational complexity of SCMA detection algorithm,a dynamic factor graph method is proposed to eliminate redundant operations in log-MPA,and good results are achieved.Finally,inspired by forward error correction coding,considering the poor bit error rate performance of log-MPA algorithm,a log-MPA detection algorithm(ECPC-log-MPA)based on error compensation sphere decoding of partial code-words(ECPC)is proposed,which introduces error compensation mechanism and sparseness adaptive matching pursuit(S-MPA)on the basis of spherical decoding.Parsity adaptive matching pursuit(SAMP)algorithm estimates the error vector and compensates the test results.The simulation results show that the algorithm can effectively reduce the computational complexity while reducing the bit error rate.At the same time,its convergence speed is also improved,and it has good practicability.