An attack-aware shipping enterprise cybersecurity framework based on deep learning
Zhongdai Wu, Junxiang Wang, Qiangqiang Shi, Jinxu Zhang, Jin Liu, Xiliang Zhang · 2023
Digitalization of shipping aims to enhance shipping efficiency, safety, and sustainability through advanced technology. However, fully digitalized shipping relies heavily on the integration of various communication devices, which exposes shipping enterprises to increasing cyber risks, such as DDoS attacks, malware attacks, and data leakage threats. Unlike ground networks, shipping networks have high latency and low bandwidth, making traditional methods less accurate and challenging to meet the network protection requirements of shipping enterprises. In contrast, deep learning (DL)-based methods can effectively identify and respond to new types of shipping network attacks by simulating human expert decision-making. Therefore, In this paper, we propose a DL-based security framework that integrates attack type awareness to enhance the network security protection capability of shipping enterprises. The proposed framework performs real-time monitoring and analysis of network data to identify various types of network attacks and take corresponding defense measures timely. This framework was implemented and tested in COSCO Shipping Group, effectively improving the overall network security protection capability.