上行SCMA系统的动态子图检测算法

面对未来网络需求量的爆炸性增长，稀疏码多址接入（Sparse Code Multiple Access，SCMA）作为一种基于码本的非正交多址技术，在用户连接数、频谱效率和低时延都有很大的提升。由于码字的稀疏性，SCMA采用消息传递算法（Message Passing Algorithm，MPA）实现多用户检测，但是复杂度还是相对较高，以至于很难应用于实际系统。为了解决上述问题，提出了一种动态子图消息传递算法（Dynamic Sub-graph Message Passing Algorithm，DS-MPA）进行信号检测。DS-MPA利用剩余值来确定下一次迭代的子图，减少了资源节点的更新用户数量，达到了降低复杂度的目的。仿真结果显示算法性能几乎无损并且降低了1/3左右的复杂度。

A dynamic sub-graph detection algorithm for uplink SCMA systems

In the face of the explosive growth of network demand in the future,sparse code multiple access(SCMA),as a codebook based non-orthogonal multiple access technology,can greatly improve the number of user connections,spectrum efficiency and low delay.Because of the sparsity of the codeword,SCMA uses message passing algorithm(MPA) to implement multiuser detection,but the complexity is still relatively high so that it is difficulty to be applied in the actual system.In order to solve the above problems,a dynamic sub-graph message passing algorithm(DS-MPA) is proposed for signal detection.DS-MPA uses residual values to determine the next iteration sub-graph,thus reducing the number of users,updating the resource nodes,and achieving the goal of reducing complexity.The simulation results show that the performance of the proposed algorithm is almost lossless and the complexity is reduced about 1/3.