Investigation of Traffic Classification Applied to an Astronomical Data Transmission Network of the XAO Using Deep Learning

• Published in: Research in astronomy and astrophysics 2023-03, Vol.23 (3), p.35003
• Main Authors: Wang, Jie, Zhang, Hai-Long, Wang, Na, Ye, Xin-Chen, Wang, Wan-Qiong, Li, Jia, Zhang, Meng, Zhang, Ya-Zhou, Du, Xu
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Astronomical data; astronomical databases: miscellaneous; Classification; Communications traffic; Data transmission; Deep learning; Image classification; Machine learning; Network analysis; Neural networks; Remote control; Remote observing; Star maps; surveys; Traffic; Traffic information; Transmission efficiency; virtual observatory tools
• ISSN: 1674-4527; 2397-6209
• DOI: 10.1088/1674-4527/acafc5

A telecommunication network used for the transmission of astronomical observation data, telescope remote control and other astronomical research purposes is a critical infrastructure. The monitoring and analysis of network traffic, which help improve the network performance and the utilization of network resources, are a challenging task. The accurate identification of the astronomical data traffic will effectively improve transmission efficiency. In this paper, a classification method applied to types of traffic containing astronomical data using deep learning is proposed. The advantages of a convolutional neural network model in image classification are exploited to classify types of traffic containing astronomical data. The objective is to identify the mixed traffic in the network and accurately identify types of traffic containing astronomical data. The effectiveness of the model in improving classification accuracy is also discussed. Actual traffic data captured by Tcpdump and Wireshark are tested, and the experimental results indicate that the proposed method can accurately classify types of traffic containing astronomical data.