A Computationally Efficient Stack-Based Iterative Precoding for Multiuser MIMO Broadcast Channel

In this paper, we propose a computationally efficient stack-based iterative precoding (SBIP) algorithm that achieves a performance comparable to the sphere encoder for multi-user MIMO downlink systems. The proposed algorithm is basically based on stack algorithm aided lattice reduction in the closest lattice point search of vector precoding. The first full expansion is tried by selecting the first level node with a minimum metric, constituting a reference metric. To find the best sequence, it iteratively visits each node and terminates the expansion by comparing node metrics with the calculated reference metric. In addition, we derive the range of candidate set size by using a characteristic of vector precoding. All things considered, 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.