ACO结合P2P信任模型的无线传感器网络Sinkhole攻击检测

Sinkhole attack detection based on fusion of ACO and P2P trust model in WSN

针对无线传感器网络中的Sinkhole攻击问题,提出了一种基于蚁群优化(ACO)结合P2P信任模型的Sinkhole攻击检测(P-ACO)算法。首先,使用蚁群优化算法检测路由中是否存在Sinkhole攻击,并生成传感器节点的警报信息;然后,利用布尔表达式进化标记生成算法为群组警报节点分发密钥,并使用密钥标记可疑节点;最后,计算可疑节点列表中各节点的信任值,将信任值低于预设阈值的节点视为攻击节点。分析表明,相比二分查找算法与基于规则匹配的神经网络(RMNN)算法,该算法在匹配过程中需要更少的匹配搜索次数,提高了算法执行效率。实验结果显示,相比RMNN算法,该算法可以更加准确地检测Sinkhole攻击。

For the Sinkhole attack problem of wireless sensor network（WSN）, a detection algorithm based on fusion of ant colony optimization（ACO） and P2 P trust model was proposed. Firstly, ant colony optimization algorithm was used to detect the existence of a Sinkhole attack in route and generate the alarm information of sensor nodes. Then,Boolean expression evolve sign generation algorithm was used to distributed keys for group alarm nodes, and it was used to mark suspicious nodes. Finally, P2 P trust model was used to compute trust value of each suspicious node,and the node with trust value was lower than the preset threshold as invasion of node. The analysis shows that the proposed algorithm need less matching search times in matching process than binary search algorithm and the rules matching based neural network（RMNN） algorithm, which indicates that it has improved the efficiency of the algorithm.Experimental results show that the proposed algorithm has higher detection accuracy on Sinkhole attack than RMNN algorithm.