Deep Learning-based Predictive Beam Management for 5G mmWave Systems

Abstract

Periodic measurement reporting based beam management is not sufficiently agile for 5G New Radio (NR) and comes with significant overhead that scales with the number of beams and users. Furthermore, such an approach to beam selection is unlikely to be sufficient to avoid signal blocking in real world scenarios. We propose a method to accurately predict in advance the best serving beams and transmission points as users move through the network and thereby eliminate the need for frequent measurement reporting. Our prediction approach applies deep learning techniques similar to that used in Natural Language Processing (NLP) for translation/sentence completion tasks to the problem of predicting the best serving beams. The proposed solution enables the network to proactively switch users to new beams or cells to reduce blockage and handover related interruptions especially in high mobility scenarios. We evaluate our scheme in realistic scenarios using a new modeling technique where computer vision is used to obtain mobility traces of users from videos of live environments. We show significant benefits in terms of measurement report overhead reduction and signal-to-noise ratio enhancement through blockage prevention in several scenarios.