Blind Decoding of Massive MIMO Uplink Systems Based on the Higher Order Cumulants

In recent years, massive multiple input multiple output (MIMO) systems have received a particular focus on the contemporary research. This study investigates the pilot contamination problem in a massive MIMO uplink system. This problem has always been evident in the surrounding cells experiencing reuse of the piloting sequence on the massive MIMO system that leads to the deterioration of the decoding system. Unlike most of previous solutions to this problem, which assume the availability of perfect channel state information (CSI) or based on estimating the CSI using the pilot sequence, this article develops a novel blind decoding algorithm for MIMO uplink systems, that significantly mitigates the pilot contamination problem. The proposed method based on higher order cumulants utilizes independent statistical methods prior to the encoding without using pilot sequences. The proposed method estimates the channel matrix and separates the received signals from in-cell and neighbouring cells through a reduction in the cumulants dependent cost-effective functioning. Simulation results indicates that the presented blind decoding algorithm enhances the performance of the decoding process and outperforms the zero-forcing (ZF) and the minimum-mean square-error (MMSE) schemes under imperfect CSI. Moreover, the presented method approaches the MMSE deciphering performance and ZF interpreting with perfect CSI at high SNR.