基于LSTM的卷积神经网络异常流量检测方法
信息技术与网络安全 7期
陈解元
(国家计算机网络与信息安全管理中心,北京100032)
摘要: 针对传统机器学习方法依赖人工特征提取,存在检测算法准确率低、无法应对0day漏洞利用等未知类型攻击等问题,提出一种基于卷积神经网络(Convolutional Neural Networks,CNN)和长短期记忆网络(Long-Short Term Memory,LSTM)混合算法的异常流量检测方法,充分发掘攻击流量的结构化特点,提取流量数据的时空特征,提高了异常流量检测系统性能。实验结果表明,在CIC-IDS2017数据集上,多种异常流量检测的准确率均超过96.9%,总体准确率达到98.8%,与其他机器学习算法相比准确率更高,同时保持了极低的误警率。
中图分类号: TP393.08
文献标识码: A
DOI: 10.19358/j.issn.2096-5133.2021.07.007
引用格式: 陈解元. 基于LSTM的卷积神经网络异常流量检测方法[J].信息技术与网络安全,2021,40(7):42-46.
文献标识码: A
DOI: 10.19358/j.issn.2096-5133.2021.07.007
引用格式: 陈解元. 基于LSTM的卷积神经网络异常流量检测方法[J].信息技术与网络安全,2021,40(7):42-46.
Network intrusion detection based on convolutional neural networks with LSTM
Chen Xieyuan
(National Computer Network Emergency Response Technical Team/Coordination Center of China(CNCERT/CC), Beijing 100032,China)
Abstract: As traditional machine learning methods rely on artificial feature extraction,there are problems such as low accuary and inability to deal with unknown types of attacks such as 0day vulnerability exploitation,this paper proposed a hybrid algorithm based on Convolutional Neural Networks(CNN) and Long-Short Term Memory(LSTM) to fully explore the structural characteristics of attack traffic, extract the spatiotemporal characteristics of traffic data, and improve the performance of abnormal traffic detection system.The experimental results show that on the CIC-IDS2017 data set, the accuracy of various abnormal traffic detection is more than 96.9%, and the overall accuracy reaches 98.8%, which is higher than other machine learning algorithms, while maintaining a very low false alarm rate.
Key words : network intrusion detection;Convolutional Neural Networks(CNN);Long-Short Term Memory(LSTM);deep learning
0 引言
信息技术的广泛应用和网络空间的兴起发展,极大促进了经济社会繁荣进步,同时也带来新的安全风险和挑战。网络安全威胁逐步从信息窃听、篡改、传播病毒等方式上升为更新颖的高强度DDoS攻击、0day漏洞利用、APT攻击等形式,造成的大规模数据泄露和网络黑产行业大规模增长严重危害信息系统运营者权益和用户个人隐私[1]。网络空间中信息传输与交互均以流量为载体,通过异常流量检测,及时发现网络异常情况和攻击行为,对于强化网络安全应急响应能力,维护网络空间安全具有重要意义[2]。
本文详细内容请下载:http://www.chinaaet.com/resource/share/2000003676
作者信息:
陈解元
(国家计算机网络与信息安全管理中心,北京100032)
此内容为AET网站原创,未经授权禁止转载。