Research On Sparse Code Multiple Access Technology For 5G Large-Scale Random Access

With the rapid development of Internet of Things and mobile Internet technology,the contradiction between the growing demand for random access and the scarcity of spectrum resources has brought severe challenges to the fifth-generation mobile communication system.Sparse code multiple access(SCMA)technology combines multidimensional modulation and sparse spread spectrum technology to significantly improve the number of users and the spectral efficiency of the system.SCMA system uses message passing algorithm(MPA)to detect multiple users at the receiving end.Its computational complexity increases exponentially with the increase of the number of users,which brings great challenges to the hardware design and implementation of the system.Therefore,the purpose of this paper is to design a low complexity and high performance multi-user detection MPA algorithm.Based on the research of SCMA technology and the principle of MPA algorithm,this paper proposes a low complexity multiuser detection scheme.This paper first studies the basic principles and key technologies of SCMA system,including the design principle of uplink transmitter and receiver,factor matrix design principle and multi-user detection algorithm.In the aspect of multi-user detection,this paper mainly studies the maximum a posteriori algorithm with the best detection performance,the widely used MPA algorithm and the typical design principle,algorithm flow and computational complexity analysis of several typical MPA optimization algorithms.Based on the above research,this paper proposes a low complexity message passing algorithm based on dynamic threshold and convergence stability judgment(DTS-MPA)by reducing the number of iterations and simplifying the factor graph.DTS-MPA algorithm combines channel state information to set dynamic confidence threshold,which improves the situation that fixed threshold can not adapt to the change of channel conditions.At the same time,the dynamic threshold decision is combined with the convergence stability decision to avoid the system decoding error caused by the non convergence of some messages caused by the simplification factor graph.In the iterative process,DTS-MPA algorithm decodes the user nodes that meet the decision conditions in advance to reduce the computational complexity of the algorithm.Finally,through theoretical analysis and simulation,it is proved that the proposed DTS-MPA algorithm can achieve the tradeoff between system BER performance and detection complexity.