基于条件概率选择更新的SCMA系统译码方案

稀疏码分多址(sparse code multiple access, SCMA)是针对第5代移动通信系统提出的一种新型非正交多址方案,其接收端采用的消息传递算法(message passing algorithm, MPA)在迭代更新过程中存在冗余计算,因而复杂度过高。针对这一问题,首先利用蒙特卡罗法对接收信号的条件概率密度函数值进行统计分析,发现随着信噪比的变化,条件概率密度函数值呈现一定的统计规律。然后,根据统计结果,提出基于条件概率选择更新的MPA(conditional probability selection update MPA, CPSU-MPA)。该算法选取一定比重的条件概率密度函数值进行更新,同时对未参与更新的外部信息进行舍弃补偿。仿真结果表明,所提算法能有效地降低译码复杂度,且误码率性能牺牲较小。

Decoding scheme of SCMA system based on conditional probability selection update

Sparse code multiple access (SCMA) is a new non-orthogonal multiple access scheme for the fifth generation mobile communication system. The message passing algorithm (MPA) adopted by the receiver has redundant calculations during the iterative update process, so the complexity is too high. To solve this problem, the Monte Carlo method is first used to statistically analyze the conditional probability density function value of the received signal, and it is found that with the change of the signal to noise ratio, the conditional probability density function value shows a certain statistical phenomenon. Then based on the statistical results, a conditional probability selection update MPA (CPSU-MPA) is proposed. The algorithm selects a conditional probability density function value with a certain proportion to update, and at the same time discards and compensates external information that does not participate in the update. Simulation results show that the proposed algorithm can effectively reduce the decoding complexity, and the sacrifice of bit error rate performance is small.