可信网络连接(TNC)是对可信平台应用的扩展,也是可信计算机制与网络接入控制机制的结合。针对当前TNC研究中缺乏对网络连接进行定量评估的问题,提出了一种网络连接可信度仿真评估方法。该方法根据网络连接属性与评估值之间存在的非线性映射关系,利用BP网络模型及LM优化算法对网络连接可信度进行评估。仿真结果表明该方法具有较好的预测评估准确率。
To cope with the shortage of quantitative evaluations in currently trusted network connection,this paper presented an evaluation method of network connection credibility with back propagation(BP) model and Levenberg-Marquardt(LM) optimization algorithm.It shows that the nonlinear relationship between the attributes and the credibility of network connections can be established by training neural-network using LM algorithm.Simulation results indicate this method an achieve high prediction and evaluation accuracy.