云计算中网络通信服务目标数据高效检测仿真

Efficient Detection and Simulation of Network Communication Service Target Data in Cloud Computing

对通信服务目标数据的检测,能够有效提高网络通信服务质量。对目标数据的高效检测,需要获得网络数据非零元素的幅度信息,对重构后的数据进行特征选取,完成目标数据的检测。传统方法对网络测量数据时空特性进行分析,建立高维随机矩阵模型,但忽略了对数据特征进行选取,导致检测精度偏低。提出联合压缩感知重构算法和支持向量机算法的目标数据高效检测方法。利用压缩感知技术对下层检测节点通信服务数据进行处理,构造通信服务数据稀疏矩阵和测量矩阵,完成对稀疏度未知信号的精确重构。利用最小二乘法获得网络数据非零元素的幅度信息,对重构后的数据进行特征选取,计算通信服务数据特征间的近似度,最后采用支持向量机来判断通信服务数据是否异常。实验结果表明,所提方法通信服务数据分类线性边界定位准确,检测精度、查全率较高,误报率较低。

An efficient detection method of target data based on compressive sensing reconstruction algorithm and support vector machine algorithm is proposed. Firstly, the compressive sensing technology is used to process the com- munication service data of detection nodes in lower level and construct the sparse matrix and the measurement matrix of communication service data, thus the accurate reconstruction of sparse unknown signal is completed. Secondly, the least squares method is used to obtain the amplitude information of nonzero elements in network data. Moreover, the feature of reconstructed data is selected to calculate degree of approximation among features of communication service data. Finally, support vector machine is used to determine whether the communication service data is abnormal or not. Simulation results show that the proposed method can locate the linear boundary of communication service data accurately, which has high detection accuracy and recall level. The false positive rate is low.