电信科学 ›› 2017, Vol. 33 ›› Issue (9): 50-57.doi: 10.11959/j.issn.1000-0801.2017226

• 研究与开发 • 上一篇    下一篇

基于变量节点串行策略的SCMA低复杂度译码研究

吉明明,郑建宏   

  1. 重庆邮电大学通信与信息工程学院,重庆 400065
  • 修回日期:2017-07-16 出版日期:2017-09-01 发布日期:2017-09-11
  • 作者简介:吉明明(1992-),男,重庆邮电大学通信与信息工程学院硕士生,主要研究方向为5G关键技术——新型多址技术SCMA。|郑建宏(1961-),男,重庆邮电大学通信与信息工程学院教授,主要研究方向为新一代宽带移动通信协议及系统应用。
  • 基金资助:
    国家科技重大专项基金资助项目(2016ZX03002010-003)

Research on low complexity decoding for SCMA based on serial strategies of variable node

Mingming JI,Jianhong ZHENG   

  1. School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China
  • Revised:2017-07-16 Online:2017-09-01 Published:2017-09-11
  • Supported by:
    The National Science and Technology Major Project of China(2016ZX03002010-003)

摘要:

稀疏码多址接入(SCMA)技术作为5G无线通信网络的一个竞争性的非正交多址方案,具有广阔的应用前景。但目前基于SCMA的上行链路均采用泛洪方案的消息传递算法(MPA)进行译码,无论是检测的复杂度还是收敛速度都不甚理想。提出一种改变信息更新策略的串行消息传递算法——S-MPA(serial MPA),按照变量节点的次序进行消息处理和传递,每一个变量节点同时进行校验消息的接收和变量消息的发送。理论和仿真结果表明,该算法不仅能保持良好的性能,而且也有较低的解码复杂度。

关键词: 稀疏码多址接入, 多用户检测, 变量节点, 串行消息传递

Abstract:

Sparse code multiple access technology,as a competitive non orthogonal multiple access scheme for the fifth generation wireless communication network,has a broad application prospect.However,current SCMA uplink use the flooding scheme of message passing algorithm for decoding,both the detection complexity and convergence properties are not ideal.A serial message passing algorithm which changed the information updating strategy was proposed,which performed message processing and deliveried according to the order of the variable nodes.Each of variable node simultaneously performed the reception of the verification message and the transmission of the variable message.The theoretical and simulation results show that the proposed algorithm not only maintains good performance,but also has low decoding complexity.

Key words: sparse code multiple access, multiuser detection, variable node, serial message passing

中图分类号: 

No Suggested Reading articles found!