Deep Learning-Based Fast Security Detection Method for Cross-Domain Access Behaviors in IPv6 Networks

With the popularity of IPv6 networks and the increase in cross-domain access behavior, network security issues are becoming increasingly prominent. This paper proposes a fast cross-domain access behavior security detection method for IPv6 networks based on deep learning, aiming to achieve real-time monitoring of cross-domain access behavior and accurate identification of potential security threats through deep learning technology. To verify the effectiveness of our research method, three sets of comparative experiments were designed to compare the false positive rate, detection delay and recall rate with three existing methods: SD IoT, Eth PSD and PCA-DNN. The experimental results show that our research method performs excellently in detecting latency, maintaining a low level of latency with minimal fluctuations. In terms of recall rate, as the number of iterations increases, the recall rate gradually increases and tends to stabilize and is higher than the other three methods in most iterations. The core of this research method is to use deep learning models to conduct in-depth analysis of cross-domain access behavior in IPv6 networks. By extracting key features and constructing efficient classification models, accurate identification of potential security threats can be achieved. The experimental results show that this method not only improves detection accuracy and efficiency, but also enhances the stability and adaptability of the system.