Suppression of pilot-contamination in massive MIMO systems

Abstract

We propose a novel channel estimation method for suppression of interference during uplink training in massive MIMO systems. The method makes use of the received uplink data to improve upon the estimation of solely training based estimators. It is based on the maximum a-posteriori criterion and outperforms state-of-the-art methods in terms of estimation accuracy and robustness. We further propose a simplified semi-blind method.