优化网络入侵特征库的量子进化算法

Quantum evolutionary algorithm for optimizing network intrusion signature database

针对网络入侵检测系统中入侵特征库的性能普遍较差的缺点,提出了一种优化网络入侵特征库的改进量子进化算法(IQEA)。采用特征向量表示染色体结构,借鉴小生境协同进化思想初始化种群,以个体的匹配程度设计适应度函数,使用动态更新和"优体交叉"策略进化种群。仿真实验表明,IQEA的寻优能力和收敛速度均优于量子进化算法和进化算法,经IQEA优化后的入侵特征库,检测能力强,并具有较好的自适应性。

Concerning the poor performance of the intrusion signature database in network intrusion detection system the Improved Quantum Evolutionary Algorithm IQEA of optimizing network intrusion signature database was proposed in this paper.The IQEA adopted eigenvector to express chromosome structure initialized population based on the idea of niche cooperative evolutionary designed the fitness function based on the matching degree of individual and used the strategy about the dynamic update of quantum rotation corner and the cross of excellent individuals to evolve population.The simulation results show that the IQEA is superior to QEA and EA in search ability and convergence rate and the intrusion signature database optimized by IQEA has better detection ability and self adaptability.