摘要
通过分析BP神经网络用于检测系统存在的问题,在传统BP算法基础上,采用自动变速率学习法,引入遗忘因子、随机优化算子,并将其用于网络入侵检测系统。仿真实验结果表明,改进的BP神经网络算法用于入侵检测,速度快、易收敛,目标精度0.02很快达到。改进的BP神经网络算法的检测率、漏测率、误报率分别为96.17%,3.83%,4.15%,检测率比未改进的BP算法要高出11.65%,漏测率比未改进的BP算法要低10.66%,误报率比未改进的BP算法要低4.07%,改进算法优越性明显。
By analyzing the problems of BP neural network applied to the detection system,the automatic variable-rate learning method,forgetting factor and random optimization operator are introduced into the BP algorithm on the basis of traditional BP algorithm. The BP algorithm is applied to the network intrusion detection system. The simulation results show that the improved BP neural network algorithm applied to intrusion detection has the characteristics of fast speed and easy convergence,and can quickly obtain the target accuracy of 0.02. The detection rate,missed detection rate and false alarm rate of the improved BP neural network algorithm can reach up to 96.17%,3.83% and 4.15% respectively,whose detection rate is 11.65% higher than that of the traditional BP algorithm,the missed detection rate is 10.66% lower than that of the traditional BP algorithm,and the false alarm rate is 4.07% lower than that of the traditional BP algorithm. The superiority of the algorithm is obvious.
出处
《现代电子技术》
北大核心
2017年第11期91-94,共4页
Modern Electronics Technique
关键词
BP算法
入侵检测
神经网络
随机优化算子
BP algorithm
intrusion detection
neural network
random optimization operator