MLT Wind Estimation using Machine Learning Meteor Selection Algorithm

This study explores the application of machine learning methods, specifically Convolutional Neural Networks (CNNs), for the selection of meteor data to derive wind fields. By implementing a CNN-based meteor selection process, we aim to out-perform traditional meteor selection criteria. We present wind fields derived from our CNN-selected meteor data and compare these wind fields to those obtained using traditional meteor selection methods. This comparison seeks to highlight the potential of CNNs in specular meteor radar data analysis and possibly showcase improvement over existing methods, which have been the status quo for more than two decades. Our findings will provide insights into the effectiveness and reliability of CNNs in this context, offering a new perspective on wind field measurement techniques that can be applied to both old and new specular meteor radars.