Adaptive stack-based iterative precoding for the multiuser MIMO downlink

The stack-based iterative precoding (SBIP) is an efficient algorithm that achieves a tradeoff between bit error rate (BER) performance and computational complexity in multi-user MIMO downlink systems. In this paper, we propose an adaptive SBIP algorithm with variable candidate set and statistically derived threshold for pruning. By analyzing the upper bound of accumulated metric, the range of candidate set is calculated. The probability distribution is also derived for statistical pruning of the tree search. Applying these ideas, the number of undesirable node visit is significantly reduced. Monte-Carlo simulations show that the proposed algorithm considerably reduces the computational complexity whereas the performance is close to that of sphere encoder (SE).