A self-organizing map based solution to the automatic detection of meteor echoes in radio recordings

A novel automatic meteor detection approach, using a self-organizing map type of artificial neural network is presented in this paper. The proposed solution aims at analyzing a given dataset and extracting the meteor signals present in it. For this, a self-organizing map is trained using radio recordings. This neural network will then be used on a test dataset to check the performances of the proposed approach.