Secure and Resilient Artificial Intelligence of Things: A HoneyNet Approach for Threat Detection and Situational Awareness
Liang Tan, Keping Yu, Fangpeng Ming, Xiaofan Cheng, Gautam Srivastava
Abstract
Artificial Intelligence of Things (AIoT) is emerging as the future of Industry 4.0 and will be widely applied in consumer, commercial, and industrial fields. In AIoT, intelligent objects (smart devices), smart gateways, and edge/cloud nodes are subject to a large number of security threats and attacks. However, the traditional network security approaches are not fully suitable for AIoT. To address this issue, this article proposes a HoneyNet approach that includes both threat detection and situational awareness to enhance the security and resilience of AIoT. We first design a HoneyNet based on Docker technology that collects data to detect adversaries and monitor their attack behaviors. The collected data are then converted into images and used as samples to train a deep learning model. Finally, the trained model is deployed in AIoT to perform threat detection and provide situational awareness. To validate our scheme, we conduct HoneyNet deployment and model training on the SiteWhere AIoT platform and construct a simulation environment on this platform for threat detection and situational awareness. The experimental results demonstrate the feasibility and effectiveness of our solution.