中文摘要：

基于非合作演化博弈理论，提出了在攻防双方信息不对称情况下具有学习机制的攻防演化博弈模型。结合攻防效用函数，对非合作演化博弈攻防过程中的纳什均衡点的存在性和唯一性进行论证。用系统动力学建立演化博弈模型进行仿真，仿真结果表明引入第三方动态惩罚策略的演化博弈模型存在纳什均衡，指出在网络安全技术进步的同时，发展攻击者追踪技术，增强网络攻击行为可审查性，实现动态惩罚，是解决网络安全问题的重要途径。

英文摘要：

An offense-defense game model with learning mechanism in the case of asymmetric information was proposed based on non-cooperation evolution game theory. Combined with utility function, the existence and uniqueness of Nash equilibrium in the offense-defense process were proved. Simulation by system dynamics shows that there is Nash equi- librium in evolutionary game model after introducing the dynamic penalty strategy of the third party. Therefore, when improving all kinds of security technology, promoting attacker tracing technology, enhancing the censorship of network attack behaviors and dynamic penalty are fundamental ways to information security.