The Intelligent Networking and Security Team at the School of Artificial Intelligence and Software Engineering has recently achieved a significant milestone with their research paper titled “BDTM: Bidirectional Detection and Traceability Mitigation of LDoS Attacks in SDN” being accepted by IEEE Transactions on Information Forensics and Security (TIFS), a premier international journal in cybersecurity. Professor Ma Xiaopu served as the first author, with his master’s student Li Xiancong as the corresponding author. Nanyang Normal University is listed as the sole completing institution.

TIFS is one of the most influential academic journals in the field of international cybersecurity. It is ranked as a Class A recommended journal by both the China Computer Federation (CCF) and the Chinese Association for Cryptologic Research (CACR), and is also classified as a SCI Zone 1 TOP journal by the Chinese Academy of Sciences. This achievement marks the university’s first breakthrough in publishing in TIFS (a CCF Class A and SCI Zone 1 TOP journal) as an independent completing institution.  

Although the flexible architecture of Software-Defined Networking (SDN) provides strong support for the development of intelligent networks, Low-rate Denial of Service (LDoS) attacks remain a serious threat to network security. In particular, sophisticated LDoS attacks—which integrate methods such as IP spoofing and distributed attacks—can bypass existing detection and defense systems and even reverse-exploit mitigation strategies to launch cross-plane attacks.  

To tackle this challenge, the research team proposed a bidirectional detection and traceable mitigation solution that excels in detection accuracy, real-time performance, and interpretability. The solution innovatively designs a multidimensional, bidirectional feature matching mechanism to accurately identify attack traffic, effectively addressing vulnerabilities in existing mitigation strategies. Meanwhile, by continuously tracking and predicting attacker behavior, it achieves more efficient and thorough attack defense. This solution outperforms current mainstream methods in detection accuracy, response speed, and resource consumption, demonstrating broad application prospects and promotional value.  

Written by: Li Xiancong, Guo Kangda & Zhang Jun  

Source: NYNU Academic Activities (Chinese)  

https://www.nynu.edu.cn/info/1048/30114.htm