Graph-merged detection and decoding of polar-coded MIMO systems
Shusen Jing, Junmei Yang, Anlan Yu, Xiaohu You, Chuan Zhang · 2017
By adopting multiple-input multiple-output (MIMO) technique, the spectral efficiency and data rate of wireless communication systems can be highly improved. To fully take the advantages of MIMO, belief prorogation (BP) detection methods are considered as a way in balancing the hardware complexity and error performance. Also with BP methods, polar code, which has been adopted by 3GPP eMBB control channel, can achieve high reliability and throughput. In this paper, a graph merged detection and decoding (GMDD) method for polar-coded MIMO systems is proposed. When the factor graphs of MIMO detection and polar decoding are merged, the GMDD methods can help to achieve better error performance with nearly no additional cost of hardware and latency compared with the conventional separated detection and decoding (SDD) method, which has been verified by numerical results. The hardware architecture of the proposed GMDD method is given in this paper as well.