%0 Journal Article %A Wang, Q. %A Zhang, L. %A Li, B. %A Zhu, Yi. %D 2019 %I Begell House %K MIMO relay, semi-blind receiver, asymmetric nested PARAFAC, Khatri-Rao, space-time-frequency coding %N 18 %P 1637-1649 %R 10.1615/TelecomRadEng.v78.i18.20 %T SEMI-BLIND RECEIVER FOR MIMO-OFDM RELAY SYSTEM BASED ON ASYMMETRIC NESTED PARAFAC DECOMPOSITION %U https://www.dl.begellhouse.com/journals/0632a9d54950b268,184c344e05a1a7b5,3af65c816c7d783b.html %V 78 %X In the MIMO-OFDM relay system, most semi-blind receivers based on tensor model adopt alternating least squares algorithm (ALS), which has high computational complexity. For this reason, a closed-form semi-blind receiver algorithm based on asymmetric nested PARAFAC decomposition (ANPD) is proposed for MIMO-OFDM relay system. This algorithm uses triple Khatri-Rao (KR) product to perform space-time-frequency (STF) coding on the symbol matrix, which formed a fifth-order tensor at the destination node that satisfies the proposed ANPD. Then based on this decomposition, Khatri-Rao factorization (KRF) is used to derive the semi-blind receiver for joint channel and symbol estimation. To verify the feasibility of the proposed receiver, the complexity cost and uniqueness issues of the receiver are also discussed. Compared with the existing one-way two-hop symmetry methods, the proposed scheme uses an asymmetric nested model to obtain additional coding diversity, which significantly improves the performance of the system. Simulation results strongly demonstrate that the proposed semi-blind receivers based on asymmetric scheme outperform the competitive receivers in parameter estimation accuracy. %8 2019-12-27