A Downlink Max-SINR Precoding for Massive MIMO System

To acquire the maximal array gain and mitigate the impact of pilot contamination, the downlink precoding algorithm on Max-SINR criterion was investigated and improved for Massive MIMO system. The objective function assured to maximize the utilization rate of the transmission power under the condition that SINR is not lower than the desired threshold. The Lagrangian function was deduced according to the objective function. Further SINR computation counted the channel estimation error caused by pilot contamination and its computational complexity was reduced by utilizing the character of Massive MIMO channel. At last the optimal downlink precoding matrix was obtained by KKT conditions. When base station configured 100 antennas and the pilot contamination was severe, numerical results proved that the spectrum efficiency of the proposed algorithm was 1.9 times higher than the popular Massive MIMO precoding --MF algorithm. The proposed algorithm can effectively mitigate the impact of pilot contamination and achieve high throughput with maximizing gain, and is valuable to Massive MIMO system.