Beam Selection – Machine Learning Applied to the Physical Layer of Millimeter – Wave MIMO Systems
The talk will describe the problem “Machine Learning Applied to the Physical Layer of Millimeter-Wave MIMO Systems” (ML5G-PHY [beam selection]), in which participants are asked to design a machine learning model (e.g. a deep neural network) for selecting a set of beams given information about the environment obtained via LIDAR, cameras and GPS. The beam-selection problem will be briefly discussed. The presentation includes describing Raymobtime datasets as well as a hands-on introduction to baseline Python code. The goal is to help those interested get started on the challenging issue of using multimodal data to improve beam selection, an important problem related to the physical layer of modern communication networks.