Data-aided channel estimation in large antenna systems

This paper is concerned with the uplink in a multi-cell large antenna system. We study a channel estimation scheme where partially decoded data is used to estimate the channel. We show that there are two types of interference components in this scheme that do not vanish even when the number of antennas grows to infinity: cross-contamination and self-contamination. Cross contamination is in principle similar to pilot contamination in a conventional pilot-based channel estimation scheme, while self-contamination is unique for the data-aided scheme. The data-aided scheme can effectively suppress the contamination effect by increasing the data frame length without causing rate loss. This is confirmed by both analysis and simulation results.