面向MANETs按需路由协议的黑洞攻击解析改进模型

传统移动Ad Hoc网络(mobile Ad Hoc network,MANETs)黑洞攻击解析模型存在网络拓扑结构固定、网络传输性能预测精确度低的问题.针对使用按需路由协议的MANETs网络,提出一种基于随机拓扑近似技术的黑洞攻击解析改进模型(improved black hole attack analytical model,IBAAM).IBAAM协议使用随机模型代替传统解析模型使用的n元2立方体模型,并将网络结构扩展至随机拓扑结构,使用最短跳距离概率描述表示网络拓扑结构的随机拓扑信息,再使用K均值聚类法实现跳距离文件配置以求解任意拓扑结构下的攻击概率问题,从而在不利用任何实际拓扑先验信息条件下有效预测MANETs网络平均丢包数目.IBAAM实验结果表明,在多种不同固定Ad Hoc拓扑结构下,IBAAM的网络丢包预测精确度在仿真结果的95%置信区间内,能够有效预测网络传输性能.

Improved Black hole attack analytical model for MANETs on demand routing protocols

The problem of the traditional analytical model of black hole attack for mobile Ad Hoc network (MANETs) is lower prediction accuracy for network transmission performance. Hence, for an ad hoc network with on demand routing protocol, a novel black hole analytical model was proposed based on stochastic topology approximation technology. IBAAM employs a stochastic analytical model instead of n ary 2 cube model used in the traditional analytical method, and expands the topology into the any arbitrary network topologies. It employs the shortest hop distance probability to represent the random information of network topology, and uses K-means clustering method to realize the hop distance file configuration to compute the attack probability, which can predict the average packet loss of the networks without using any given information about practical network topology. The IBAAM experiments showed, under variety of the fixed Ad hoc topology, the average prediction accuracy rate of the network packet loss of the analytical model proposed is about 95% of packet loss in the simulation results.