基于阈值和区间筛选的SCMA多用户检测算法

SCMA multiuser detection algorithm based on threshold and interval filtering

在5G通信用户信息高吞吐量和大规模连接应用场景下,稀疏多址接入(sparse code multiple access,SCMA)可用于改善5G智慧场景通信质量。为了降低SCMA系统检测端算法复杂度,基于Max-log-MPA算法,提出了一种基于单阈值门限和区间范围的多参数组合设定的消息传递算法。在用户信息值迭代运算前利用门限阈值进行判断,当满足设定条件后方可参与算法的迭代过程,迭代计算过程中再次设定区间参数进行过滤筛选,使得通信场景所需的用户节点概率值参与计算,用户信息得到译码,2部分参数均可一定程度筛选用户信息值,从而达到降低算法复杂度的目的。仿真结果表明,提出的算法收敛速度和算法复杂度明显优于消息传递算法(message passing algorithm,MPA),误比特率(bit error ratio,BER)性能损失也可得到保证,在误码率性能和算法复杂度权衡之间有灵活的参数组合方案,具有很强的适应性。

In the scenario of high-throughput and large-scale connection applications for 5G communication users,sparse code mul-tiple access(SCMA)can be used to improve the communication quality of 5G smart scenes.In order to reduce the complexity of the detection algorithm in the SCMA system,a message passing algorithm based on a single threshold and interval range was proposed,which was set through multiple parameter combinations based on the Max-log-MPA algorithm.The threshold value was used to judge whether to participate in the algorithm iteration before the iteration calculation of the user information value.When the set con-dition was satisfied,the user information value could participate in the algorithm iteration process.In the iteration calculation pro-cess,the interval parameter was set again for filtering and selection,so that the user node value required for the communication sce-nario could participate in the calculation,and the user information could be decoded.Both parts of the parameters could filter the user information values to some extent,thereby achieving the goal of reducing the algorithm complexity.Simulation results show that the proposed algorithm is superior to the message passing algorithm(MPA)algorithm in terms of convergence speed and algorithm com-plexity.The bit error ratio(BER)performance loss can also be guaranteed.The algorithm has a flexible parameter combination scheme between BER performance and algorithm complexity,which has strong adaptability.