Research On SCMA Uplink Multiuser Detection Algorithm

Most of the existing communication systems communicate with orthogonal resources in which the number of devices is limitted with a certain maximum.5G research gradually focus on non-orthogonal multiple access technology,owing to huge traffic,more mobile terminals,and explosive growth of throughput capacity in recent years.Sparse code multiple access(SCMA)is a multi-dimensional codebook-based non-orthogonal multiplexing technique proposed to address the above requirements.This paper introduces the basic technology of SCMA,firstly.The uplink communication model is simulated and analyzed.This paper studies the constellation mapping and sparse spread spectrum technology of SCMA transmitter,and introduces the design principle of multiuser nonorthogonal codebook.Since the number of users is greater than the number of available resources,users who occupy non-orthogonal resources should use a codebook that maximizes the minimum Euclidean distance;users who occupy resources should use the same codebook.Secondly,the message passing algorithm(MPA)of the receiver end is studied.The paper proposes two multiuser detection algorithms with low complexity in uplink.In this paper,a fast convergence MPA algorithm is obtained by changing the iterative order of the internal soft message values,and the improved fast convergence MPA algorithm is obtained by changing the user iteration order.An incomplete iterative MPA detection algorithm is proposed for users with high signal-to-noise ratio.These two algorithms can reduce the detection complexity and delay at the expense of less performance.Finally,the interference elimination in MPA algorithm is studied in this paper.By separating the MPA internal iterative structure,this paper proposes quasi-parallel interference cancellation MPA detection algorithm.In this algorithm the users who receive the high noise ratio signal output the codeword information in advance.After decoding is correct,the user data is re-encoded and remapped,and then the user channel fading coefficient is added to subtract these users' reconstruction signal from the received signal and also the factor graph matrix is updated;the remaining users' detection is based on the updated received signal and the factor graph matrix so that the low received signal to noise ratio user obtains the performance gain.The algorithm improves the performance of the system under the premise of sacrificing complexity and delay.