云计算恶意行为轨迹检测算法研究

发布时间：2018-06-17 01:26

本文选题：云计算 + 恶意行为　；参考：《计算机仿真》2016年02期

【摘要】：在对云计算恶意行为轨迹进行检测的过程中,由于恶意行为轨迹具有不同的特征,使得行为轨迹之间存在关联性较低。传统的挖掘方法,在进行挖掘时,低关联性会给恶意行为轨迹的选择带来干扰,无法有效实现云计算恶意行为轨迹的挖掘。提出采用贝叶斯博弈的云计算恶意行为轨迹挖掘算法,用户端每隔一段时间向服务端传输一次位置数据信息,统一设定一个时间段,完成分段操作。通过随机投影转换将高维空间中的数据点映射至低维空间中。对恶意行为攻击信号的状态空间进行更新迭代,进行对恶意行为攻击信号的平滑操作和降噪滤波处理移动对象在云计算环境中的重要性。将时间间隔TI看作是一个博弈周期,依据上一周期的检测结果,对恶意行为移动对象存在概率进行调整。仿真结果表明,所提方法具有很高的检测精度。
[Abstract]:In the process of detecting the malicious behavior trajectory of cloud computing, due to the different characteristics of malicious behavior trajectory, there is a low correlation between the behavior trajectory. Traditional mining methods, when mining, low correlation will bring interference to the selection of malicious behavior trajectory, can not effectively realize cloud computing malicious behavior trajectory mining. A Bayesian game based trajectory mining algorithm for malicious behavior of cloud computing is proposed. The user transmits location data to the server every other time and sets a time period to complete the segmentation operation. The data points in high dimensional space are mapped to low dimensional space by random projection transformation. The state space of malicious attack signal is updated and iterated, and the importance of moving object in cloud computing environment is processed by smoothing the malicious attack signal and filtering noise. Time interval TI is regarded as a game period. According to the detection results of the previous period, the probability of malicious behavior moving object is adjusted. Simulation results show that the proposed method has a high detection accuracy.
【作者单位】：郑州工业应用技术学院信息工程学院;武汉大学计算机学院;
【基金】：河南省高等学校重点科研项目(15B520036)
【分类号】：TP393.08

【相似文献】