Construction of IPv6 Network Multi-Layer Collaborative Defense Model Under Unbounded DoS Attack
Due to the high concealment and complexity of unbounded DoS attack,single-layer defense can not solve the intrusion problem.Therefore,a multi-layer collaborative defense model of IPv6 network under unbounded DoS attack is constructed,which is based on the concept of collaborative defense.Module traffic is collected and DNN model is used to detect abnormal IPv6 network traffic situations.The collaborative HCF self-learning algorithm is introduced to realize IPv6 network defense information sharing and abnormal network traffic filtering.IPv6 network defense is carried out by using the reinforcement learning algorithm of attack damage correlation.The experimental results show that the model can recover the CPU in 2 ms.IPv6 network throughput after different types of DoS at-tacks is consistently higher than 200 KiB/s.This shows that the model can effectively resist different types of DoS attacks,react quickly and have good defense effect.
DoS attackIPv6 networkcoordinated defensedeep neural networkattack source tracing
国家科技期刊平台
NSTL
万方数据
