摘要
IoT security is very crucial to IoT applications,and security situational awareness can assess the overall security status of the IoT.Traditional situational awareness methods only consider the unilateral impact of attack or defense,but lackconsideration of joint actions by both parties.Applying gametheory to security situational awareness can measure the impact of the opposition and interdependence of the offensive and defensive parties.This paper proposes an IoT security situational awareness method based on Q-Learning and Bayesian game.Through Q-Learning update,the long-term benefits of action strategies in specific states were obtained,and static Bayesian game methods were used to solve the Bayesian Nash Equilibrium of participants of different types.The proposed method comprehensively considers offensive and defensive actions,obtains optimal defense decisions in multi-state and multi-type situations,and evaluates security situation.Experimental results prove the effectiveness of this method.
出处
《国际计算机前沿大会会议论文集》
2021年第2期190-203,共14页
International Conference of Pioneering Computer Scientists, Engineers and Educators(ICPCSEE)
基金
the National Key Research and Development Program of China(No.2017YFB1400700).