Lattice reduction aided belief propagation for massive MIMO detection

作     者：Senjie ZHANG Zhiqiang HE Kai NIU Shi JIN Hong CHENG 

作者机构：Key Laboratory of Universal Wireless Communications Ministry of EducationBeijing University of Posts and Telecommunications National Mobile Communications Research Laboratory Southeast University Intel Labs China 

出 版 物：《Science China(Information Sciences)》 (中国科学:信息科学(英文版))

年 卷 期：2019年第62卷第4期

页      面：127-139页

核心收录：

学科分类：080904[工学-电磁场与微波技术] 0810[工学-信息与通信工程] 0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学] 080402[工学-测试计量技术及仪器] 0804[工学-仪器科学与技术] 081001[工学-通信与信息系统] 

主　　题：massive MIMO MIMO detection belief propagation graph-based detection lattice reduction Markov random field 

摘      要：Efficient massive MIMO detection for practical deployment, which is with spatially correlated channel and high-order modulation, is a challenging topic for the fifth generation mobile communication(5G). In this paper, lattice reduction aided belief propagation(LRA-BP) is proposed for massive MIMO detection. LRA-BP applies the message updating rules of Markov random field based belief propagation(MRF-BP) in lattice reduced MIMO system. With the lattice reduced, well-conditioned MIMO channel,LRA-BP obtains better message updating and detection performance in spatially correlated channel than MRF-BP. Log-domain arithmetic is used in LRA-BP for computational complexity reduction. Simulation result shows that LRA-BP outperforms MRF-BP with 3–10 dB in terms of required SNR for 1% packet error rate in spatially correlated channel for 256-QAM. We also show that LRA-BP requires much lower complexity compared with MRF-BP.