中文摘要：

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。

英文摘要：

The cloud computing malicious behavior trajectories of excavation process,it＇s easy to have a large amount of calculation,the problem of large randomness,lead to the traditional a semi-supervised clustering based cloud computing trajectory mining method,malicious behavior with fixed parameters,unable to effectively implement cloud computing trajectory of digging,malicious behavior presents a cloud computing based on bayesian game malicious behavior path mining method,the client every once in a while to the server location data transmission a unified set a time period,to complete the section operation. By random projection transformation to map data points in high dimensional space to low dimension space. State space of malicious ACTS against signal update iteration,by doppler frequency shift fuzzy search to complete to malicious ACTS against signal smoothing operation. By IIR filter algorithm for noise filtering of signal processing. Based on the moving objects in cloud computing energy,such factors as the number of connections,the number of passing information to determine the importance of the moving objects in cloud computing environment. The cycle time interval TI as a game,on the basis of test results on a cycle,through the bayesian method to adjust the probability of moving objects exist for malicious behavior. The simulation results show that the proposed method has high accuracy.