基于贝叶斯推理的攻击路径预测研究

Research on predicting attack path based on Bayesian inference

针对攻击路径预测方法中存在的因路径冗余而导致的节点置信度错误计算问题,提出了攻击可行性概念,并改进了基于贝叶斯推理的似然加权法。通过对子攻击路径进行成本—收益分析,提出了基于攻击可行性的攻击路径生成算法,尽可能消除路径冗余;并在此基础上对似然加权法进行了改进,以支持上述算法,有效地避免计算节点置信度的错误计算。实验结果表明,该方法能有效地消除路径冗余,提高节点置信度的准确性,提升对攻击路径的有效预测。

Path redundancy in methods for predicting paths causes the problem that the computation of node belief is incor- rect. In order to solve it, this paper presented the concept of attack feasibility and improved the likelihood weighting algorithm based on the Bayesian inference. By analyzing cost-benefit on child attack paths, firstly it proposed an algorithm generating at- tack paths to eliminate path redundancy as far as possible; secondly, based c_n the above, it improved the likelihood weighting algorithm to support above algorithm so that it could effectively avoid the problem of incorrect computation and improve the ac- curacy of node belief. The finally experimental results show that the method can effectively eliminate the path redundancy and evidently improve the accuracy of node belief and consequently enhance the validity of prediction for attack paths.