Optimal Training in Continuous Flat-Fading Massive MIMO Systems

摘要: Massive multiple-input multiple-output (MIMO) communication promises the full potential of MIMO realized through deployment hundreds antennas at each base station. Accurate channel state information (CSI) is crucial for making most massive MIMO. In time-varying channels, however, if not periodically updated, CSI obtained station becomes progressively outdated. this paper, we formulate a training optimization problem maximizing uplink sum-rates. The optimal period and update interval are characterized as solutions to fixed-point equations, which depend on normalized Doppler shift large-scale coefficients. It found that, sufficiently large number station, equal users per cell.