基于贝叶斯决策的自组网推荐信任度修正模型

在分析了信任评估过程中攻击手段及其相互间关系的基础上,提出了基于贝叶斯决策理论的根据推荐偏差度修正对推荐的信任度方法.使用贝塔分布描述推荐偏差度,依据最小损失原则修正对推荐的信任度,并将具备推荐信任修正机制的信任模型运用在自组网的路由协议中,以便优化路由选择.MATLAB下的仿真结果表明,该方法能够有效抵御一些针对信任管理的威胁并提升信任管理的正确率,进而提高自组网环境下检测恶意节点的效率.

Bayesian Decision-Making Based Recommendation Trust Revision Model in Ad Hoc Networks

In this paper, new attacks against trust recommendation are identified and relationships between these attacks are analyzed. Then a trust model is presented with recommendation trust revision support based on Bayesian Decision-Making theory and the minimum- loss principle. The proposed trust model with recommendation trust revision (TMRTR) is employed in ad hoc networks for optimizing ad hoc routing. Simulation results in MATLAB environment show the method proposed is helpful to reduce the impact of some new threats to trust management and improves the effect of trust management. Synchronously, the proposed system can effectively detect malicious nodes in ad hoc networks.