云计算环境下异常信息优化检测仿真

Simulation of Anomaly Detection in Cloud Computing Environment

对云计算环境下异常信息的优化检测,能够有效解决云计算环境中信息存储以及传输过程的安全问题。对异常信息的检测,需要定义异常信息行为的属性值,计算出异常信息行为分类器的输出最大目标概率值,完成异常信息的优化检测。传统方法先给出异常信息的调用流图特征,从而给出异常信息的"行为"特性,但忽略了计算出异常信息的最大目标概率值,导致检测精度偏低。提出基于朴素贝叶斯的云计算环境下异常信息优化检测方法。上述方法利用字符矩方法提取出异常信息的样本特征,得到未知异常信息中的每个分块空间的关系特征向量,给出异常信息行为模式在朴素贝叶斯中的表述方程,定义异常信息行为的属性值,计算出异常信息的输出最大目标概率值,并完成对云计算环境下异常信息优化检测。实验结果表明,所提方法检测精度较高,为保障云计算环境的安全稳定运行奠定了基础。

This research focuses on optimization detection method for abnormal information in cloud computing en- vironment. The research used character moment method to extract sample feature of abnormal information and ob- tained vector of relationship feature of each block space in unknown abnormal information. Expression function of be- havior pattern of abnormal information in naive Bayes was provided and attribute value of the abnormal information be- havior was defined. The research also worked out maximum target probability value of input of abnormal information. Experimental results show that the method has higher detection precision. It lays a foundation for ensuring safe and stable operation of cloud computing environment.