Find in Library
Search millions of books, articles, and more
Indexed Open Access Databases
Belief Driven Attack and Defense Policy Optimization Mechanism in Honeypot Game
oleh: JIANG Yang-yang, SONG Li-hua, XING Chang-you, ZHANG Guo-min, ZENG Qing-wei
Format: | Article |
---|---|
Diterbitkan: | Editorial office of Computer Science 2022-09-01 |
Deskripsi
As a typical deception defense means,honeypot technology is of great significance in actively trapping attackers.The existing design methods mainly optimize the trapping decision of honeypot through the game model,ignoring the impact of the attacker's belief on the game decision of both sides.There are some shortcomings,such as weak adaptive optimization decision-making ability,easy to be seen through and used by the attacker and so on.Therefore,a belief based honeypot game mechanism(BHGM) is proposed.Based on the multi round game process of attacker completing the task,BHGM focuses on the impact of honeypot action on attacker's belief and the impact of belief on whether the attacker continues to attack.At the same time,a belief driven algorithm for solving the optimal attack and defense strategy is designed based on the upper confidence bound apply to tree(UCT).Simulation results show that the belief driven attacker strategy can choose to continue the attack or stop the loss in time based on the current belief to obtain the maximum profit,while the belief driven honeypot strategy can reduce attacker's suspicion as much as possible to lure him to continue the attack and obtain greater profit.