A Dynamic Access Method for Cross-domain Circulation of Data Based on Differential Algebraic Neural Networks

A Dynamic Access Method for Cross-domain Circulation of Data Based on Differential Algebraic Neural Networks

연구 분야: Analysis

학회: 2024 9th International Symposium on Computer and Information Processing Technology (ISCIPT)

The cross domain data flow access method relies on decentralized processing and storage modes, which poses a risk of single point of failure access and affects the security of data access. Therefore, a cross domain data cyclic dynamic access method based on differential algebraic neural networks was designed. We have identified the dynamic access objects for cross domain data flow and analyzed the requirements of dynamic access for cross domain data flow, which facilitates subsequent data processing and flow. A dynamic access model for cross domain data circulation was constructed based on differential algebraic neural networks. By utilizing the basic structure and functions of differential algebraic neural networks, cross domain cyclic data was processed and transmitted to adapt to changes in dynamic data access. Define dynamic access rules for cross domain circulation, analyze which users can access which data, specify the time, location, method, and permissions of access, and ensure the security of dynamic access for cross domain circulation. The experimental results show that this method has good access performance and can be applied in real life.