Beam Tracking for Distributed Millimeter-Wave Massive MIMO Systems Based on the Unscented Kalman Filter

In this letter, a beam tracking problem is studied in distributed millimeter-wave (mm-wave) massive multiple-input multiple-output (MIMO) systems. Firstly, a monopulse beam direction estimation method for a distributed mm-wave massive MIMO system is introduced. However, this method is very sensitive to noise and must be performed in every time slot in a single-user moving scene. To improve its accuracy, a monopulse beam tracking method based on the unscented Kalman filter (UKF) for a distributed mm-wave massive MIMO system is proposed. The simulation results show that the proposed beam tracking method has better tracking performance under different SNRs and different numbers of antennas of remote antenna units (RAUs), and its robustness and generalizability are quite well.