动态网络环境下电力无线通信网络信息异常值快速检测

Fast Detection of Outlier of Power Wireless Communication Network Information in Dynamic Network Environment

常规的电力网络信息异常值检测方法以熵阈值检测为主,熵阈值的取值范围受到充分数据的影响,降低了异常值的检测精准度。因此,设计了动态网络环境下电力无线通信网络信息异常值快速检测方法。提取电力无线通信网络动态异常信息文本特征,将网络信息中的异常关键词进行量化,剔除重复无意义的异常文本特征,更好地通过文本特征区分网络信息异常值。构建动态通信网络信息异常值快速检测模型,动态评估网络的安全环境,通过设定网络异常值检测阈值,保障网络信息安全。均衡无线通信网络信息异常值检测负载,将电力无线通信网络信息采样节点随机化,检测节点被通信网络完全覆盖,均匀随机地检测网络中的任意节点,从而避免漏检、错检等问题。采用对比实验,验证了该检测方法的精准度更高,能够应用于实际生活。

Conventional Outlier detection methods of power network information mainly focus on entropy threshold detection.The value range of entropy threshold is affected by sufficient data,which reduces the accuracy of Outlier detection.Therefore,a fast detection method for Outlier of power wireless communication network information in dynamic network environment is designed.Extract the text features of dynamic abnormal information of power wireless communication network,quantify the abnormal keywords in the network information,eliminate the repeated and meaningless abnormal text features,and better distinguish the Outlier of network information through the text features.Build a fast detection model of dynamic communication network information Outlier,dynamically evaluate the network security environment,and improve the network information security by setting the network Outlier detection threshold.Balance the detection load of Outlier of wireless communication network information,randomize the sampling nodes of power wireless communication network information,and the detection nodes are completely covered by the communication network,so as to detect any node in the network evenly and randomly,so as to avoid problems such as missed detection and wrong detection.Through comparative experiments,it was verified that the detection method has higher accuracy and can be applied in practical life.