一种用于实时探测网络虫洞节点检测模型研究

A Detection Model for Real-Time Detection Network Worm Hole Nodes

针对传统的检测建模算法在建立网络虫洞节点检测模型时,存在检测时间过长,耗费能源较高的问题。提出采用基于模糊预测的实时探测网络虫洞节点检测建模方法。先利用节点的邻居数查询出可能受虫洞影响的路由节点,得到可能受虫洞影响节点的集合,利用节点发送数据包确定周围的邻居节点数,将收集到的不同位置邻居节点数存储在滑动窗口中,在节点移动到下一位置前预测出其邻居节点数的上限阈值,当节点实时移动到该位置时,检测出邻节点数,与上限值进行比较,当检测的邻居点数大于阈值时,则认为此节点受到虫洞攻击。仿真证明,采用改进的检测建模方法时效性较强,有效地提升网络虫洞节点检测效率。

In view of the problems of too long detection time and high energy consumption when traditional detection modeling algorithm in establishing the detection model for network wormhole nodes, a detection modeling method for wormhole nodes in real - time detection network based on fuzzy prediction is proposed. The number of neighbors of a node is used to query the routing nodes which may be affected by the wormhole, and the set of these nodes is obtained. Then, the nodes are used to send data package, the number of neighbor nodes around is determined, and the number of the collected neighbor nodes in different positions are stored in the sliding window. Before the node moves to a position, the upper threshold of the number of neighbor nodes is predicted. When the nodes move to the position in real time, the number of neighbor nodes is detected, and which is compared to the upper limit value; when the detected number of neighbor points is greater than the threshold, it is considered that the node is attacked by wormhole. Simulation shows that the improved detection modeling method has strong timeliness, and can effectively enhance the detection efficiency of network wormhole nodes.