一种新的基于神经网络的入侵容忍系统

An intrusion-tolerant system based on neural network

与预防攻击为主的安全方案不同,入侵容忍系统可以在受到攻击之后仍能提供正常连续的服务,作者从入侵容忍技术的原理入手,提出了一个入侵容忍模型,在该模型中采用了基于神经网络的入侵检测技术作为触发器,通过对该模型的工作流程和性能进行分析得出它具有很好的自适应性、安全性和弹性。

Different from security schemes mainly for prevention, the intrusion-tolerant system can provide normal and continuous service even after it has been attacked. The theories related to intrusion-tolerant is introduced, then a network intrusion-tolerant model is presented. In the model, it adopts an intrusion detection system based on neural network as its trigger, through the analysis of the workflow and performance, a conclusion is drawn that the model has good self-adaptability,security and flexibility, and can resist various of service attacks.