基于生物地理优化算法的网络入侵检测
|
摘要
特征提取和选择是网络入侵检测的一个关键步骤,针对当前特征提取算法存在的缺陷,以提高入侵检测正确率为目标,提出了一种基于生物地理优化算法的网络入侵检测模型。对当前网络入侵检测研究现状进行分析,找到当前模型检测正确率低的原因,建立网络入侵检测特征提取和选择的数学模型,采用生物地理优化算法对该数学模型进行求解,并根据求解结果建立最优的入侵检测模型,在Matlab 2014平台上进行了仿真测试。结果表明,该模型可以找到最优的特征,提高了入侵检测的正确率,具有较高的实际应用价值。
Feature extraction and selection are key steps of network intrusion detection.In view of the current defects of feature extraction algorithm,in order to improve the intrusion detection accuracy,this paper put forward a kind of network intrusion detection model based on the biological geographical optimization algorithm.First the current status of network intrusion detection research was analyzed,the cause of the low accuracy of the current model was found,and then a mathematical model of network intrusion detection feature extraction and selection was established,and then the biogeography optimization algorithm was used to solve the mathematical model.According to the result,an optimal model of intrusion detection was given,and finally in the Matlab simulation test is carried out.Results show that the model can find the optimal characteristics,improve the accuracy of intrusion detection,and has higher application value.
Feature extraction and selection are key steps of network intrusion detection.In view of the current defects of feature extraction algorithm,in order to improve the intrusion detection accuracy,this paper put forward a kind of network intrusion detection model based on the biological geographical optimization algorithm.First the current status of network intrusion detection research was analyzed,the cause of the low accuracy of the current model was found,and then a mathematical model of network intrusion detection feature extraction and selection was established,and then the biogeography optimization algorithm was used to solve the mathematical model.According to the result,an optimal model of intrusion detection was given,and finally in the Matlab simulation test is carried out.Results show that the model can find the optimal characteristics,improve the accuracy of intrusion detection,and has higher application value.
引文
[1]卿斯汉,蒋建春,马恒太,等.入侵检测技术研究综述[J].通信学报,2004,25(7):19-29.
[2]王永全.入侵检测系统(IDS)的研究现状和展望[J].通信技术,2008,41(11):139-143.
[3]周荃,王崇骏.基于人工智能技术的网络入侵检测的若干方法[J].计算机应用研究,2007,24(5):144-148.
[4]郭亚周,高德远.模糊聚类分析在入侵检测系统中的应用研究[J]沈阳理工大学学报,2005,24(4):26-28.
[5]饶鲜,董春曦,杨绍全.基于支持向量机的入侵检测系统[J].软件学报,2003,14(4):798-803.
[6]汪兴东,佘堃,周明天,等.基于BP神经网络的智能入侵检测系统[J].成都信息工程学院学报,2005,20(1):1-4.
[7]李文华.基于聚类分析的网络入侵检测模型[J].计算机工程,2011,37(17):96-98.
[8]孙晓艳,郑淑丽,沈洪伟.优化的RBF神经网络在入侵检测中的应用[J].合肥工业大学学报(自然科学版),2008,31(11):1794-1797.
[9]杨富华,彭刚.PCA-SVM在网络入侵检测中的仿真研究[J].计算机仿真,2011,28(7):146-149.
[10]朱永宣,单莘,郭军.基于免疫算法的入侵检测系统特征选择[J].微电子与计算机,2007,24(3):20-22.
[11]饶鲜,杨绍全,魏青,等.基于熵的入侵检测特征参数选择[J].系统工程与电子技术,2006,28(4):599-601.
[12]黄辰林,赵辉,胡华平.基于分布自治代理的层次入侵检测系统设计[J].计算机工程与应用,2001,6(37):47-49.
[13]宋宇翔,刘琰.特征和分类器联合优化的网络入侵检测算法[J].计算机工程与应用,2012,48(19):77-81.
[2]王永全.入侵检测系统(IDS)的研究现状和展望[J].通信技术,2008,41(11):139-143.
[3]周荃,王崇骏.基于人工智能技术的网络入侵检测的若干方法[J].计算机应用研究,2007,24(5):144-148.
[4]郭亚周,高德远.模糊聚类分析在入侵检测系统中的应用研究[J]沈阳理工大学学报,2005,24(4):26-28.
[5]饶鲜,董春曦,杨绍全.基于支持向量机的入侵检测系统[J].软件学报,2003,14(4):798-803.
[6]汪兴东,佘堃,周明天,等.基于BP神经网络的智能入侵检测系统[J].成都信息工程学院学报,2005,20(1):1-4.
[7]李文华.基于聚类分析的网络入侵检测模型[J].计算机工程,2011,37(17):96-98.
[8]孙晓艳,郑淑丽,沈洪伟.优化的RBF神经网络在入侵检测中的应用[J].合肥工业大学学报(自然科学版),2008,31(11):1794-1797.
[9]杨富华,彭刚.PCA-SVM在网络入侵检测中的仿真研究[J].计算机仿真,2011,28(7):146-149.
[10]朱永宣,单莘,郭军.基于免疫算法的入侵检测系统特征选择[J].微电子与计算机,2007,24(3):20-22.
[11]饶鲜,杨绍全,魏青,等.基于熵的入侵检测特征参数选择[J].系统工程与电子技术,2006,28(4):599-601.
[12]黄辰林,赵辉,胡华平.基于分布自治代理的层次入侵检测系统设计[J].计算机工程与应用,2001,6(37):47-49.
[13]宋宇翔,刘琰.特征和分类器联合优化的网络入侵检测算法[J].计算机工程与应用,2012,48(19):77-81.