Beam alignment is required in millimeter-wave communication to ensure high data rate transmission. However, with narrow beamwidth in massive MIMO, beam alignment could be computationally intensive due to the large number of beam pairs to be measured. To target this problem, we’ve studied the location-aware beam alignment framework by exploiting the location information of the user equipment (UE) and potential reflecting points [1]. Both UE and BS perform a coordinated beam search from a small set of beams within the error boundary of the location information. The selected beams are then used to guide the search for future beams. To further reduce the number of beams to be searched, we propose an intelligent search scheme within a small window of beams to determine the direction of the actual beam. The BS and UE coordinate the search iteratively and the search is stopped when the target rate is achieved.

In mmWave mobile systems, the beam tracking is required to maintain the beam alignment. Two common tracking methods for mmWave beamforming are codebook-based method and perturbation method. However, the codebook method is limited by the resolution of the codebook whereas perturbation method requires lengthy beam training which incurs large overhead as well as delay, rendering these approaches unsuitable for mobile environment. To solve this problem, we’ve studied the filter-based beam tracking which estimates the channel parameters (channel gain and angle) and error compensation method which control the beamwidth adaptively. In [2], the BS tracks the UE based on particle filter and, the adequate beamwidth is set via the partial activation of the antenna array. In [3], the channel is tracked with the extended Kalman filter (EKF) at the BS and the beamforming weight is updated with a robust minimum mean squared error beamformer bounded by the array vector error which is fed from the error variance estimated by the EKF.