被入侵网络中的活跃节点检测方法研究

It Was an Invasion in Active Nodes in Network Detection Method Research

对被入侵网络中的活跃节点进行检测,可以保证在网络在瘫痪的情况下,恢复通信能力。被入侵网络与正常网络不同,活跃节点分布具有较大随机性,节点之间的可检测活跃特征的关联较弱,传统的检测方法需要通过节点的关联性才能完成是否活跃的判断,只能以随机检测的方式完成,准确度较低。提出改进约束粒子群算法的被入侵网络中的活跃节点检测方法。上述方法先对采集的被入侵网络中各个活跃节点的原始信号进行特征提取,再将每个活跃节点的特征进行数据标准化处理,将标准化处理后的活跃节点特征向量依据一定的顺序进行特征组合,得到一个被入侵网络中的活跃节点二维特征向量矩阵,并进行活跃节点的数据特征融合,并将被入侵网络中的活跃节点定位问题转换成约束优化问题,融合粒子群优化算法采用设定约束适应度函数和距离适应度函数的方式对上述问题进行求解。仿真结果表明,改进约束粒子群算法的被入侵网络中的活跃节点检测方法定位精确度高。

The detection of active nodes in network can ensure communication restoration ability of the network in case of paralysis. The paper proposed an active node detection method for intruded network based on improved constrained particle swarm optimization algorithm. In this method,the collected original signals of all active nodes in the intruded network were extracted for to obtain the feature,and then the feature of each active node was standardized.After standard treatment,the active node＇s feature vectors were combined according to a certain sequence to get a two-dimensional feature vector matrix of the active node in intruded network,and the data feature of the active node was fused. The active node localization problem in intruded network was transmitted into constraint optimization problem in combination with particle swarm optimization algorithm. The way to set constraint fitness function and distance fitness function was used to solve the above problems. The simulation results show that the active node detection method for intruded network has high location accuracy.