摘要： 针对Ad Hoc网络中虫洞攻击检测可预测的特点,提出一种基于模糊预测的虫洞检测算法。分析当节点受到虫洞攻击时,其邻节点数会因虫洞影响而异样增多的现象,利用节点发送数据包确定其周围邻节点数,并将收集到的一系列位置邻节点数存储在滑动窗口中,根据前期邻节点数的统计,运用模糊预测理论,在节点运动到下一位置前预测出其邻节点数的上限阈值。当节点实际运动到下一位置后,检测出邻节点数,并与上限阈值进行比较,当检测的邻节点数大于阈值时则认为此节点受到虫洞攻击。仿真结果表明,该算法时效性较强,检测率较SWAN算法有所提高。

关键词: 虫洞攻击, 模糊预测, 滑动窗口, Ad Hoc网络, 邻节点, 虫洞检测

Abstract: For Ad Hoc network wormhole attack detection predictability,this paper presents a wormhole detection algorithm based on fuzzy prediction.When a node is attacked by a wormhole,its neighbor node number abnormally increases.This algorithm uses the node to send packets,to determine its surrounding neighbor node number,then stores the total number of nodes in a sliding window.According to preliminary statistics of neighbor nodes,using fuzzy predictive theory,it predicts the upper threshold before the node moves to the next position.When the node actual moves to the next position,it detects neighbor nodes and compares with the upper threshold.When the neighbor node is greater than the threshold,the point is considered be attacked by wormhole.Simulation results show that the detection rate is improved over the contrast algorithm compared with Statistical Wormhole Apprehension using Neighbors(SWAN) algorithm.

Key words: wormhole attack, fuzzy prediction;sliding window;Ad Hoc network;neighbor node;wormhole detection

中图分类号: 

• TP393