Adaptive Decision Feedback Detection with Parallel Interference Cancellation and Constellation Constraints for Multi-Antenna Systems

Abstract: In this paper, a novel low-complexity adaptive decision feedback detection with parallel decision feedback and constellation constraints (P-DFCC) is proposed for multiuser MIMO systems. We propose a constrained constellation map which introduces a number of selected points served as the feedback candidates for interference cancellation. By introducing a reliability checking, a higher degree of freedom is introduced to refine the unreliable estimates. The P-DFCC is followed by an adaptive receive filter to estimate the transmitted symbol. In order to reduce the complexity of computing the filters with time-varying MIMO channels, an adaptive recursive least squares (RLS) algorithm is employed in the proposed P-DFCC scheme. An iterative detection and decoding (Turbo) scheme is considered with the proposed P-DFCC algorithm. Simulations show that the proposed technique has a complexity comparable to the conventional parallel decision feedback detector while it obtains a performance close to the maximum likelihood detector at a low to medium SNR range.