无线网络传输数据节点入侵优化检测仿真

摘要

In this paper,we propose an optimization detection method for intrusion of transmission data node in wireless network based on the reputation mechanism.Firstly,the behavior attribute of wireless network node is analyzed and the packet loss rate and forwarding rate of node is quantized to build model and analyze different attributes of different nodes.Then,the nodes with different attribute are analyzed and judged through building the reputation mechanism.Moreover,the node behavior is quantized and the pre-established credit threshold is compared with the credit value of network node.Finally,we confirmed the magnitude relationship between the credit value of tested node and pre-established threshold value and detected out whether the node is abnormal.The simulation results show that the modified method can detect out the node intrusion state effectively compared with traditional threshold method of reputation mechanism.It provides the support for the safe service of network data.%对无线网络传输数据节点入侵进行准确检测,可保证无线网络的安全运行.进行入侵检测时,应将节点行为进行量化后运用预先设定的信誉阈值与网络节点的信誉值比较完成检测,但是传统方法通过对无线网络某一区域环境中变量自发性的临界变化状态进行分析,找出最脆弱的网络节点完成人侵检测,但是无法将待检测的节点信誉阈值与正常网络节点的信誉值进行比对,降低了入侵检测的精确度.提出一种基于信誉机制的无线网络传输数据节点入侵优化检测方法.首先,对无线网络节点的行为属性进行分析,通过对节点的丢包率、转发率等数据实现量化后构建模型,分析不同节点所具有的不同属性.然后,对不同属性节点通过建立的信誉机制进行分析判断,将节点行为进行量化后运用预先设定的信誉阈值与网络节点的信誉值比较,确定被检测节点信誉值与预先设定阈值的大小关系,最终检测出节点是否异常.仿真结果表明,与传统的信誉机制的阈值法进行比较,改进方法可以有效地检测出节点节点入侵状况,为网络数据的安全使用提供了支持.