Research On Space-frequency Sparse Code Multiple Access Technology Based On Multi-antenna System

The fifth generation mobile communication(5G)is required to provide higher data rate,higher network capacity,massive communication links,ultra-low propagation latency and better quality of service.In mobile communication networks,multiple access technology is one of core technologies in physical layer of wireless communication,which is aimed at distinguishing users and providing access service.In an orthogonal multiple access network,the number of users is proportional to orthogonal physical resource elements,thus it cannot meet the requirements for 5G,such as the large capacity,massive links and low latency access.Non-orthogonal multiple access is becoming the research focus of 5G multiple access technology.Sparse code multiple access(SCMA)is one of code domain non-orthogonal multiple access technologies.We can utilize the sparsity of SCMA codebook to combine bit stream modulation mapping and spread spectrum together.In order to obtain shaping gain and increase the system capacity,the user bits are directly mapped to multi-dimensional codewords.The sparsity of the multi-dimensional codewords can also make the receiver conduct reliable data recovery with low complexity multi-user joint detection.In this paper,we carry out research work based on SCMA technology.The main research contents are as follows:(1)In order to combine the space diversity of multiple antenna technology with frequency diversity of SCMA technology,we apply SCMA technology to MIMO-OFDM multi-antenna system.In addition,we design a kind of multi-antenna space-frequency SCMA(SF-SCMA)system that is taking advantage of the characteristics of sparse factor graph.In this paper,the SF-SCMA system model is characterized.In addition,we elaborate the multi-user coding algorithm of SF-SCMA system and the idea and implementation process of the MPA joint detection algorithm.We compare the BER performance of the single antenna SCMA system with the SF-SCMA system in flat fading channel and frequency selective fading channel.(2)In order to reduce decoding complexity for SF-SCMA system,we propose a multi-user detection algorithm based on multi-input threshold.By the aid of good convergence property of symbol probability,the proposed algorithm can reduce the complexity of MPA decoding and improve the decoding efficiency with little error performance loss.We use EXIT chart to analyze the convergence of MPA decoding and its related factors.We elaborate the principle of the max-log MPA algorithm and rules of setting threshold.We compare the BER performance of the max-log MPA algorithm with the log MPA algorithm.In addition,we compare the error performance and the decoding delay of the algorithm with/without input threshold.In order to verify the effectiveness of the proposed algorithm,we design and test the FPGA hardware of max-log MPA algorithm based on fixed threshold.