Low complexity MIMO detection using OSIC with conditional list generation

摘要

This paper proposes a low complexity MIMO detection using ordering SIC (OSIC) with conditional list generation. Combined with lattice-reduction (LR) technique, OSIC further gains the BER performance compared to SIC, since the more reliable signal is prior to being detected. However OSIC based on LR (LR-OSIC) has inferior performance to the maximum likelihood (ML) detection in large size MIMO systems. In this paper, we investigate the mean square error in LR domain to evaluate the condition of the channel. The proposed detection applies the conditional list detection to update the estimate of the LR-OSIC. The simulation results exhibited that the proposed detection achieved near-ML performance in the 8×8 MIMO and required almost same complexity of the LR-OSIC in the high E_(b)/N _(0) region.