Machine-learning Approach to Detecting and Analyzing Meteor Echoes (MADAME)

We present a Machine-learning Approach to Detecting and Analyzing Meteor Echoes (MADAME), a radar data processing workflow featuring advanced machine-learning techniques using both supervised and unsupervised learning. Our results show that a Convolutional Neural Network (CNN) based one-stage object detection model, called YOLOv4, performs remarkably well in detecting and identifying meteor head and trail echoes in processed radar signals. The detector can identify more than 80 echoes per minute with 0.99 precision, 0.94 recall on head echoes, and 0.98 precision, 0.89 recall on trail echoes in the testing data obtained from the Jicamarca High Power Large Aperture radar. MADAME is also capable of autonomously processing data in interferometer mode and returning the radiant source and vector velocity of the target. In the testing data, the Eta Aquarids meteor shower can be clearly identified from the meteor radiant source distribution analyzed automatically by MADAME, demonstrating the proposed algorithm's functionality. In addition, MADAME found that about 50 percent of the meteors are traveling in inclined near-inclined circular orbits. Furthermore, meteor head echoes with a trail are more likely to originate from the shower meteor source and are potentially associated with higher mass particles. Our results highlight the capability of advanced machine learning techniques on radar signal processing, providing an efficient and powerful tool to facilitate future and new meteor research.

Citation

Li, Yanlin (2023). Machine-learning Approach to Detecting and Analyzing Meteor Echoes (MADAME) [Data set]. Scholarsphere.

Files

Metadata

Work Title Machine-learning Approach to Detecting and Analyzing Meteor Echoes (MADAME)
Access
Open Access
Creators
  1. Yanlin Li
License In Copyright (Rights Reserved)
Work Type Dataset
Publication Date March 23, 2023
Deposited March 23, 2023

Versions

Analytics

Collections

This resource is currently not in any collection.

Work History

Version 1
published

  • Created
  • Updated
  • Added Creator Yanlin Li
  • Added main18_interferometer_p2p_cluster3.m
  • Added yolo640v2.mat
  • Added interpret_bbox10.m
  • Added inference.zip
  • Updated Publication Date, License Show Changes
    Publication Date
    • 2023-03-23
    License
    • https://rightsstatements.org/page/InC/1.0/
  • Published
  • Updated

Version 2
published

  • Created
  • Added 2023_05_03_yolo_scholarshpere.zip
  • Published
  • Updated

Version 3
published

  • Created
  • Deleted interpret_bbox10.m
  • Deleted main18_interferometer_p2p_cluster3.m
  • Deleted yolo640v2.mat
  • Published
  • Updated