摘要
传统人工免疫方法中对自体的定义不是动态的,这并不符合真实的系统环境。因为随着时间的积累,有些正常与异常的区别会变小,甚至相互转化。所以,必须根据环境的变化动态调整自体集。另外,采用有限差异变异和基因重排机制产生初始检测器集,将遗传算法引入到成熟检测器的免疫进化过程中,由于遗传算法具有良好的全局寻优能力,可有效地提高优质成熟检测器的产生并缩短成熟时间。基于KDDcup数据集上的仿真实验结果证明了本方法在检测异常和入侵方面能表现出良好的性能。
In traditional way,self-set isn't dynamic,but this situation don't correspond to the real network circumstance.With time going by,the difference bettwen the abnormal and normal is becoming small,even convert to one another,so,we must adjust the self-set to adapt to the real circumstance.In the other hand,we adpot the limt mutation machanism and gene recombination to generate the unmature detector,and then,genetic algorithm is to be taken in production of mature detector.Since genetic algorithm's ability of searching in overall situation,we can shorten the time of detector maturing process.In the simulative experiment in Kddcup data-set,the result show that the method we proposed is effective in instution detection aspects.
出处
《软件导刊》
2011年第6期52-54,共3页
Software Guide
关键词
人工免疫
入侵检测
检测器
Artificial Immune
Intrusion Detection
Detector