Name: Kai Shi
Title: Associate Professor
Supervisor Type: PhD Supervisor, Master Supervisor
Email: shikai0229@tjut.edu.cn
Office Address & Phone: 7-409
Network and Information Security, Artificial Intelligence
1. Key Research and Development Program of the Ministry of Science and Technology, 2021YFC3300402, Research on Key Technologies for Secure Collaborative Sharing of Municipal Social Governance Data Based on Blockchain, 2021/12–2024/11, ¥1.3 million, Completed, Principal Investigator
2. National Natural Science Foundation of China (Youth Program), 61202381, Research on Transmission Control Mechanism in Dynamic Spectrum Resource Sharing Environment, 2013/01–2015/12, ¥240,000, Completed, Principal Investigator
3. Tianjin Natural Science Foundation, 15JCQNJC00500, Research on Transmission Control Mechanism in Vehicular Ad Hoc Networks, 2015/04–2018/03, ¥60,000, Completed, Principal Investigator
4. National Key Research and Development Program, 2016YFB0800805, Test Operation Control and Demonstration Verification, 2016/07–2019/06, ¥2.975 million, Completed, Second Participant
5. National Natural Science Foundation of China (General Program), 61272450, Research on Key Technologies of DDoS Defense for Next-Generation Internet, 2013/01–2016/12, ¥820,000, Completed, Third Participant
(1) Kai Shi, Penghao Ding, Jinsong Wang: A few-shot detection method for new types of network traffic attacks based on meta-learning with cross-attention. *Journal of Supercomputing* 81(9): 1062 (2025)
(2) Shi, K., Xue, T., Wang, J., Wu, Y., & Guo, H. (2026). BDFL: A blockchain-based decentralized federated learning security framework. *Multimedia Systems* (2026).
(3) K. Shi, H. Zhou and J. Wang, "FedDPSA: A Robust Semi-Asynchronous Differentially Private Federated Learning Framework in Heterogeneous Environments," 2025 IEEE International Conference on Systems, Man, and Cybernetics (SMC), Vienna, Austria, 2025, pp. 2455–2462.
(4) Zongpu Wei, Jinsong Wang, Zening Zhao, Zhao Zhao, Kai Shi: UpsFed-IDS: U-shaped split federated intrusion detection system for securing UAV communication in dynamic networks. *Ad Hoc Networks* 183: 104133 (2026)
(5) Zongpu Wei, Jinsong Wang, Zening Zhao, Kai Shi: Toward data efficient anomaly detection in heterogeneous edge-cloud environments using clustered federated learning. *Future Generation Computer Systems* 164: 107559 (2025)
Candidate of the 2nd/3rd Levels of Tianjin 131 Talent Training Program
Young and Middle-Aged Academic Backbone of Tianjin Universities
Third Prize of Tianjin Science and Technology Progress Award (Second Participant)
Second Prize of Tianjin Teaching Achievement Award (Fourth Participant)
