Research On Sparse Code Division Multiple Access Technology Scheme

Sparse Code Multiple Access(SCMA)is one of the most studied NonOrthogonal Multiple Access(NOMA)techniques in fifth-generation communication(5G)systems.To achieve the Internet of Everything,a new type of multiple-access technology that is very powerful and flexible is required.However,the current multiple access technology is mainly affected by the number of orthogonal resources and cannot meet the massive 5G access requirements.Compared with the orthogonal multiple access technology,the non-orthogonal multiple access technology can meet the requirements of the same The transmission of more user data in time-frequency resources can meet one of the demand scenarios of 5G:large-scale Internet of Things business(mmtc).On this basis,sparse code division multiple access technology(SCMA)came into being.In order to meet the needs of 5G communication,it innovatively uses codebook technology at the transmitting end to combine modulation and spread spectrum technology in traditional communication.Combined,it increases user flexibility and resource utilization.However,the current SCMA technology still has no exact optimal solution in the codebook direction,and the multi-user detection algorithm used at the receiving end still has the problem of high complexity.Therefore,how to design a codebook with better performance and how to Reducing the complexity of the demodulation algorithm at the receiving end has become the two most difficult problems that need to be solved at present.Based on these two problems,this paper discusses step by step,popularizes and discusses the problems involved in SCMA technology,and proposes its own optimization scheme in the field of codebook design and the reduction of the complexity of detection algorithms.The specific research contents are as follows:First of all,this paper improves the codebook created by the user based on the traditional subset partitioning method,and proposes a new codebook cutting scheme.Aiming at the problem that the traditional PSK constellation as the mother codebook cannot fully utilize the entire constellation space,this The scheme uses the Lattic constellation as the mother codebook constellation.By dividing the mother codebook constellation,users can expand the constellation through constellation rotation and scaling,so that the codebook can use the constellation space more effectively.Algorithms on the side can perform better.Aiming at the problems of poor detection performance and high complexity of traditional message passing algorithm(MPA),an MPA optimization algorithm based on message pre-weighting is proposed,which optimizes the initial value of different codeword combinations by introducing weighting factors.Probability makes the detection faster,and the number of iterations can be reduced compared to the traditional MPA algorithm.For the multi-user detection algorithm MPA,a message passing algorithm based on dynamic user matching is proposed again.The algorithm uses the probability of each user’s codewords on different resources to match after iterative completion to see if the information on all resource blocks is are the same,if the codeword information produces roughly the same posterior probability,the iteration of the dynamic judgment for the user has been completed,the codeword has a high reliability,and the codeword information is set to "1" which is completely trusted.",in order to reduce the algorithm complexity of subsequent iterations.It is obtained through simulation that the detection algorithm is better than the traditional MPA algorithm in both complexity and BER performance.