Pilot decontamination for multi-cell massive MIMO systems using asynchronous pilot design and data-aided channel estimation

In this work, we investigate the pilot contamination problem in multi-cell massive multi-input multi-output (MIMO) systems. The conventional approaches to address pilot contamination for multi-cell MIMO systems impose orthogonal pilots constraint across multiple users and cells, which leads to bandwidth inefficiency under large number of active users. We propose an uplink (UL) transmission protocol in which the pilot segments of users partially overlap across cells and even overlap with data from other cells, which provides more freedom in system designing. We also apply semi-blind algorithms at base station (BS) in order to separate the data streams of each user. Our analysis shows that semi-blind receiver removes pilot contamination effectively under large UL data length. The simulation results demonstrate that the performance of the proposed protocol with semi-blind receiver employing Constant Modulus Algorithm (CMA) is superior as compared to the existing schemes.