基于NN的空天网络行为信任模型

空天网络所具有的动态性、开放性、复杂性特点,使得空天网络需要更高的安全性。为解决空天网络实体间的信任问题,并充分利用先验知识,基于神经网络理论对空天网络实体的行为信任进行了建模。通过建立相识社区,利用相识者的直接信任,完成了推荐信任的计算,并采用BP神经网络的惩罚项理论解决了恶意推荐问题。为空天网络环境中的行为信任研究提供了新的思路。

A Behavior Trust Model in Aerospace Network Based on Neural Network Theory

In view of the characteristics of dynamic, openness and complexity, this makes aerospace network high in security .To make full use of prior knowledge, the neural network theory should be used to build up a model of the trust. Through the establishment of acquaintance community, recommending trust value is obtained. Lastly, the problem of malice recommendation is solved by BP neural network with penalty. The model proposed in this paper provides a new way for studying behavior trust in Aerospace Network.