Millimeter Wave Phased Array Antenna Synthesis Using a Machine Learning Technique for Different 5G Applications

摘要

A machine learning (ML) technique has been used to synthesis a linear millimetre wave (mmWave) phased array antenna by considering the phase-only synthesis approach. For the first time, gradient boosting tree (GBT) is applied to estimate the phase values of a 16-element array antenna to generate different far-field radiation patterns. GBT predicts phases while the amplitude values have been equally set to generate different beam patterns for various 5G mmWave transmission scenarios such as multicast, unicast, broadcast and unmanned aerial vehicle (UAV) applications.