基于递归神经网络的原始训练数据防泄漏密码生成系统设计

Design of original training data leakage prevention password generation system based on recurrent neural network

为加强原始训练数据的传输安全性,实现对信息应用文件的及时性编码,设计基于递归神经网络的原始训练数据防泄漏密码生成系统。以递归神经网络框架作为数据信息的基层传输依据条件,借助训练数据收发器与信息防泄漏加密模块,实现密码生成系统的硬件执行环境搭建。在此基础上,通过建立源码文件、译码文件的方式,确定与原始训练数据相关的防泄漏信息编码原则,完成密码生成系统的软件执行环境搭建,联合相关硬件设备结构体,实现原始训练数据防泄漏密码生成系统的顺利应用。对比实验结果表明,所设计系统可同时处理的原始训练数据总量更大,但所需的译码等待时间却相对更短,可大幅增强训练主机对于信息应用文件的编码及时性。

In order to strengthen the transmission security of the original training data and realize the timeliness coding of information application files,a recurrent neural network-based original training data anti-leakage password generation system is designed.The recurrent neural network framework is used as the basis for the basic transmission of data and information,and the hardware execution environment of the password generation system is built with the help of training data transceivers and information leakage prevention encryption modules.On this basis,by establishing source code files and decoding files,determine the principle of anti-leakage information coding related to the original training data,complete the construction of the software execution environment of the password generation system,and combine the related hardware equipment structure to realize the original training data Smooth application of antileak password generation system.The results of comparative experiments show that the total amount of original training data that the designed system can process at the same time is larger,but the required decoding waiting time is relatively shorter,which can greatly enhance the timeliness of the training host for coding information application files.