移动网络入侵特征信息提取检测研究

发布时间：2018-01-22 23:14

本文关键词： 移动网络入侵检测基于特征　出处：《计算机仿真》2017年03期 　论文类型：期刊论文

【摘要】：对移动网络入侵特征信息进行提取与检测,可以提高移动网络的安全防御能力。进行入侵特征信息提取检测时,应对入侵特征进行准确提取,通过构建训练样本进行检测,但是传统的基于机器学习方法是构建正常网络状态下的关系模式,将网络入侵事件与正常关系模式不断比对,完成入侵检测,但不能准确提取入侵特征,存在移动网络入侵特征信息检测误差大的问题。提出一种新的移动网络入侵特征信息提取检测方法。上述方法先基于PCA对输入移动网络动态变量进行潜在特征提取,将入侵特征编码变为粒子的位置向量,组建基于粒子信息共享的移动网络入侵最优特征子集,利用模糊均值聚类算法选取最优入侵特征训练样本,构建基于支持向量机的网络入侵检测器,利用上述检测器有效地完成移动网络入侵检测。仿真结果表明,所提算法移动网络入侵检测的精确度和检测效率均有效提高。
[Abstract]:Extracting and detecting intrusion feature information of mobile network can improve the security and defense ability of mobile network. When extracting and detecting intrusion feature information, intrusion feature should be extracted accurately. By constructing training samples to detect, but the traditional machine learning method is to build a normal network under the state of the relationship model, network intrusion events and the normal relationship between the mode of continuous comparison, the completion of intrusion detection. But it can not accurately extract intrusion features. This paper presents a new intrusion feature information extraction and detection method for mobile network. Firstly, based on PCA, the dynamic variables of the input mobile network are potentially special. Sign extraction. The intrusion feature coding is changed into the particle position vector, and the optimal feature subset of mobile network intrusion based on particle information sharing is constructed. The fuzzy mean clustering algorithm is used to select the optimal intrusion feature training sample. A network intrusion detector based on support vector machine (SVM) is constructed, and the mobile network intrusion detection is accomplished effectively by using the above detectors. The simulation results show that. The proposed algorithm improves the accuracy and detection efficiency of mobile network intrusion detection.
【作者单位】：西安理工大学高等技术学院;
【分类号】：TN915.08
【正文快照】： 1引言在下一代网络技术水平飞速发展的同时,人侵检测技术作为一种动态的移动网络安全检测技术成为当前网络安全研究的重点[1-3]。而在对移动网络的人侵检测过程中,提取网络入侵特征质量的优劣可以决定网络人侵检测的精度[4—61。移动网络入侵检测优化方法通过选择最优训练样

【相似文献】