Research Article
Wavelet and Kalman Filter-Empowered Traffic Detection for Secure QUIC Network Communication
@INPROCEEDINGS{10.1007/978-3-031-80713-8_10, author={Liang Shan and Junyi Wu and Keyang Gu and Jinquan Nie and Riqing Xu}, title={Wavelet and Kalman Filter-Empowered Traffic Detection for Secure QUIC Network Communication}, proceedings={Data Information in Online Environments. 4th EAI International Conference, DIONE 2023, Nanchang, China, November 25--27, 2023, Proceedings}, proceedings_a={DIONE}, year={2025}, month={2}, keywords={QUIC LDDoS attack Wavelet transform Wavelet transform}, doi={10.1007/978-3-031-80713-8_10} }- Liang Shan
Junyi Wu
Keyang Gu
Jinquan Nie
Riqing Xu
Year: 2025
Wavelet and Kalman Filter-Empowered Traffic Detection for Secure QUIC Network Communication
DIONE
Springer
DOI: 10.1007/978-3-031-80713-8_10
Abstract
HTTP/3.0 is an application layer protocol built on top of QUIC, which utilizes its characteristics to provide faster and more reliable data transmission. The combi-nation of HTTP/3.0 and QUIC is considered an important development direction for the next generation of internet transmission protocols. QUIC (Quick UDP In-ternet Connection), as an efficient data transmission protocol, can provide better data transmission quality and user experience with its low latency and high data transmission speed characteristics. However, various new network attacks such as LDDoS attacks constantly threaten the transmission capability and robustness of QUIC transmission systems. To solve this problem, based on the Self-similarity and randomness of QUIC network traffic, a traffic detection model of QUIC network communication based on wavelet and Kalman filter is proposed. First, simulate LDDoS attack through NS3, then use wavelet transform to pre-process the jitter signal after the attack, and then use Kalman filter to denoise. At last, it is compared with the jitter signal which is processed only by Kalman filter. The experiment proves that this method can be used for attack detection and pre-vention of network nodes.
