通信网络中故障数据优化检测仿真研究

发布时间：2019-05-26 21:00

【摘要】：在对通信网络中故障数据检测时,由于通信网络中数据量多,差异性大,造成相关检测数据关联性低。传统的网络故障数据检测时,在低关联性环境下对相关参数的确定需要先验知识,导致故障数据检测精度差。提出基于最小最大概率机的故障数据高效检测方法,分析了最小最大概率机的基本理论,对最小最大概率机算法进行扩展,转换成单类最小最大概率机,以适应卫星通信网络中故障数据的挖掘问题。获取优化目标函数中参数的最优值,依据得到的最优值塑造判断新测试数据是否是故障数据的函数,实现卫星通信网络中故障数据的高效检测。仿真结果表明,所提模型在卫星通信网络中,和传统模型相比,不仅运算效率高,而且可提高检测故障数据的精度。
[Abstract]:When detecting fault data in communication network, because of the large amount of data and great difference in communication network, the correlation of related detection data is low. In the traditional network fault data detection, the determination of related parameters in low correlation environment needs prior knowledge, which leads to the poor accuracy of fault data detection. An efficient fault data detection method based on minimum and maximum probability machine is proposed, the basic theory of minimum maximum probability machine is analyzed, and the algorithm of minimum and maximum probability machine is extended to a single class of minimum and maximum probability machine. In order to adapt to the problem of fault data mining in satellite communication network. The optimal value of the parameters in the optimization objective function is obtained, and the new test data is judged to be a function of fault data according to the obtained optimal value, so as to realize the efficient detection of fault data in satellite communication network. The simulation results show that the proposed model is not only more efficient than the traditional model in satellite communication network, but also can improve the accuracy of fault detection data.
【作者单位】：北华大学师范分院;东北电力大学自动化工程学院;
【分类号】：TN927.2

【参考文献】