An Intrusion Detection Method for Cyber Monintoring Using Attention based Hierarchical LSTM
Haixia Hou, Zijun Di, Mingqiang Zhang, Dongfeng Yuan · 2022
The large-scale adoption of the Internet of Things (IoT) is accompanied by significant challenges posed by cyber security issues. Therefore, it is crucial to develop intrusion detection system (IDS) that can guarantee the security of IoT networks. In this paper, in view of the problem that the network traffic data has multidimensional features, and different feature units represent different meanings, which have different levels of contribution to the type of traffic data classification, an intrusion detection method based on Hierarchical LSTM and attention mechanism (HLSTM + Attention) is proposed. First, the HLSTM is used to extract sequence features across multiple hierarchical structures on the network record sequence. After that, the attention layer is used to capture the correlation between the features, and redistribute the weights of the features, which adaptively map the different importance of each feature to different network attack categories into the network learning process. The verification experiment results on the intrusion detection benchmark data set NSL-KDD show that the proposed algorithm has a better detection performance on network intrusion detection.