基于多元节点属性分类的光纤网络入侵中未感染节点检测

摘要

The optical fiber network adopts a strong open distributed structure , which is vulnerable to malicious data and code intrusion , and proposes a study on the detection algorithm of uninfected nodes based on the attribute classification of multiple nodes .According to the principle of node distance measurement , the location information of all nodes in the optical fiber network is extracted , the characteristic attributes of the optical fiber nodes related to the type of uninfected nodes are selected , and the node attributes and intrusion types are modeled , and the sampled spatial sampling features of the extracted optical fiber nodes are studied and classified according to the multiple classification algorithm .The intrusion of uninfected nodes in the optical fiber network .The simulation experiment shows that the proposed node detection algorithm overcomes the disadvantages and shortcomings of the traditional al -gorithm.It can effectively reduce the communication cost and energy consumption , improve the intrusion detection rate and prolong the life cycle of the optical fiber network .%光纤网络采用开放性较强的分布式结构,易受到恶意数据和代码的入侵.提出基于多元节点属性分类的光纤网络入侵未感染节点检测算法研究.依据节点测距原理,提取光纤网络中全部节点的位置信息;选定与未感染节点类型相关的光纤节点特征属性;并针对节点属性和入侵类型建模.依据多元分类算法对提取的光纤节点样本空间采样特征数据进行学习和分类,检测光纤网络中的入侵未感染节点.仿真实验表明,提出的节点检测算法克服了传统算法的弊端和不足,能够有效降低通信成本和节点能耗、提高入侵检测率、延长光纤网络生命周期.