CAJ | 학술논문

ipsweep攻击信号广泛存在于复杂进化网络系统中,对其实现自校正分离,进行攻击信号准确检测.传统算法采用PSO辨别树算法进行复杂进化网络的ipsweep攻击自校正分离算法设计,存在自适应差和检测性能不好的问题.提出一种基于Dopplerlet变换匹配投影ipsweep攻击自校正分离算法.提取攻击信号传播路径的随机概率密度特征,计算自校正分离属性的信息熵的预测值,提取信号在Dopplerle空间中相干点积功率累积尺度坐标,得到ipsweep攻击信号的自校正分离模型改进设计.仿真结果表明,该算法能有效提高对ipsweep攻击信号检测性能,传统算法的检测概率为63%,而该算法的检测概率为92%,检测性能提升明显,且能有效实现对ipsweep攻击信号Dopplerle空间谱检测.
ipsweep공격신호엄범존재우복잡진화망락계통중,대기실현자교정분리,진행공격신호준학검측.전통산법채용PSO변별수산법진행복잡진화망락적ipsweep공격자교정분리산법설계,존재자괄응차화검측성능불호적문제.제출일충기우Dopplerlet변환필배투영ipsweep공격자교정분리산법.제취공격신호전파로경적수궤개솔밀도특정,계산자교정분리속성적신식적적예측치,제취신호재Dopplerle공간중상간점적공솔루적척도좌표,득도ipsweep공격신호적자교정분리모형개진설계.방진결과표명,해산법능유효제고대ipsweep공격신호검측성능,전통산법적검측개솔위63%,이해산법적검측개솔위92%,검측성능제승명현,차능유효실현대ipsweep공격신호Dopplerle공간보검측.
The ipsweep attacks signal exists widely and complex evolution network system, the realization of self correcting separation, attack accurate signal detection. The traditional algorithm using PSO tree algorithm to identify complex evolu-tionary neural network ipsweep attack self-tuning separation algorithm design, the presence of adaptive difference and the detection performance is not good problem. Put forward a kind of Dopplerlet transform, projection self-tuning separation al-gorithm based on ipsweep attack.Calculated from the predictor corrector separation belongs to the information entropy val-ues of the extracted signal in Dopplerle space coherent dot product the cumulative power scaling coordinates, get the ip-sweep attack signal separation model of improved design of self correction. The simulation results show that the algorithm can effectively improve the signal detection performance of the ipsweep attack, the detection probability of the traditional al-gorithm is 63%, and the detection probability of this algorithm is 92%, the detection performance significantly, and can ef-fectively realize spectrum detection of attacks on ipsweep signal in Dopplerle space.