Deep Learning Enabled Beam Tracking for Non-Line of Sight Millimeter Wave Communications

Abstract

To solve the complex beam alignment issue in non-line-of-sight (NLOS) millimeter wave communications, this paper presents a deep neural network (DNN) based procedure to predict the angle of arrival (AOA) and angle of departure (AOD) both in terms of azimuth and elevation, i.e., AAOA/AAOD and EAOA/EAOD. In order to evaluate the performance of the proposed procedure under practical assumptions, we employ a trajectory prediction method by considering dynamic window approach (DWA) to estimate the location information of the user equipment (UE), which is utilized as the input parameter of the trained DNN to generate the prediction of AAOA/AAOD and EAOA/EAOD. The robustness of the prediction procedure is analyzed in the presence of prediction errors, which proves that the proposed DNN is a promising tool to predict AOA and AOD in NLOS scenarios based on the estimated UE location. Simulation results shows that the prediction errors of the AOA and AOD can be maintained within an acceptable range of ±2 • .

Utilization of Beam Signatures Supporting High User Mobility With Extremely Low Feedback Overhead

“…A reinforcement learning (RL) based beam tracking strategy is investigated in [27], where the measured signal quality becomes the input to the Q-learning algorithm that decides the beam switching status. To predict the AoA/AoD both in terms of azimuth and elevation, a trajectory prediction method using DL is proposed to estimate the UE location information [28].…”

“…Although the fingerprint database is a position-based model which stores the potential pointing direction for given locations, modeling is not based on a continuous set of locations or trajectories. The prediction using the deep neural network (DNN) in [28], constructs the trajectory on-the-fly basis thus may lack accuracy compared to estimation based on the pre-determined typical trajectory set built on accumulated report data.…”