An Adaptive Honeypot Deployment Algorithm Based on Learning Automata
Yan Zhang, Chong Di, Zhuoran Han, Yichen Li, Shenghong Li · 2017
The honeypot is a kind of proactive defense technology against malicious attacks in the field of information security. Successful and timely detection of network attacks highly depends on efficient honeypot deployment. This paper proposes an adaptive honeypot deployment algorithm based on learning automata (LA) called LA-AHD for improving the network security. The whole network suffers from external attacks in an attack-defense scenario based on a mathematical model. The proposed algorithm considers the entirety of candidate nodes in the network as a LA and models each candidate node as an action of the LA. Through the interactions with attacks and the evolutions of the LA, a certain number of honeypots can be adaptively deployed at the proper place in the network. The computational experiments verify the effectiveness and efficiency of the proposed LA-AHD algorithm. The results show the algorithm can improve the speed of selecting the honeypots and increase the honeypot capture rate substantially.