论文部分内容阅读
针对目前多数入侵检测系统的低检测率问题,提出一种自适应进化神经网络算法AENNA。基于遗传算法和BP神经网络算法,利用模拟退火算法的概率突跳和局部搜索强的特性对遗传算法进行改进,采用双种群策略的遗传进化规则实现BP神经网络权值和结构的双重优化;通过对遗传算法的交叉算子与变异算子的改进,设计一种自适应的神经网络训练方法。实验结果表明,基于AENNA的入侵检测方法能够有效提高系统的检测率并降低误报率。
Aiming at the low detection rate of most intrusion detection systems, this paper proposes an adaptive evolutionary neural network algorithm AENNA. Based on genetic algorithm and BP neural network algorithm, genetic algorithm is improved by utilizing the characteristics of simulated annealing algorithm such as probability kurtosis and strong local search. The genetic algorithm of double population strategy is used to optimize the weight and structure of BP neural network. On the improvement of crossover operator and mutation operator in genetic algorithm, an adaptive neural network training method is designed. The experimental results show that the ANANA-based intrusion detection method can effectively improve the detection rate and reduce the false alarm rate.